Clenow’s Stocks on the Move, Tested Apart Ten Years After the Book
Stocks on the Move is a 2015 book by Andreas Clenow. It gives simple rules for picking stocks from the S&P 500. In short: buy the stocks that rose fastest and most smoothly over the last three months, and stop buying new stocks while the whole market is falling. The book's promise is to beat the market.
We ran these rules on the S&P 500 from 2006 to 2025. On each day we used the companies that were in the index that day, wherever we have their prices. Then we tested the rules one at a time: we turned one rule off, ran everything again, and looked at the difference.
From 2006 to 2014, the rules beat the market. But almost all of that came from one year. In 2008 the market lost more than a third of its value, and that year the rule that stops buying in a falling market kept the money in cash. Before the book and after it, the rules did not find better stocks than average. Simply owning all S&P 500 stocks in equal amounts did slightly better. After 2015, the rules fell behind the market. Over the full twenty years, every version we tested did worse than simply holding SPY, the most common S&P 500 fund.
Why we tested this book
Stocks on the Move came out in 2015 and became popular with people who invest by fixed rules. The rules are simple enough to follow by hand, and many people have copied them into their own trading programs. In 2015, the trader Nick Radge tested the rules on the years 1999 to 2014. He found that the rules made much more money than the S&P 500, and lost much less in the 2008 crash.
A book is written using the years before it came out. The years after it are a fairer test, because nobody could have adjusted the rules to fit them. We also wanted to use the S&P 500 as it was on each day. Many copies of this strategy use today's list of S&P 500 companies, which leaves out the companies that dropped out of the index over the years. That makes past results look better than they were. We use the list as it stood on each day. For about a third of the 2006 members, mostly companies that later left the index, we have no prices, and this gap closes by 2020.
The rules: the book's version and our three versions
The book's rules, as Clenow describes them:
1. Once a week, look at every stock in the S&P 500.
2. Skip a stock if its price is below its average price of the last 100 days, because its price is going down.
3. Skip a stock if it moved more than 15% up or down in a single day in the last 90 days. One big jump can make a stock look stronger than it is.
4. Rank the remaining stocks by two things: how fast their price went up over the last 90 days, and how smoothly it went up. A stock that went up a little almost every day ranks higher than a stock that went up the same amount in a few big jumps.
5. Buy from the top of the list until the money is used up. Put more money into calm stocks and less into stocks whose price jumps around a lot.
6. Sell a stock when it is no longer in the top fifth of the list.
7. The market rule: buy new stocks only while the S&P 500 is above its average of the last 200 days. When it is below that average, keep the stocks you have, but do not buy new ones.
Version A is our copy of these rules. Our platform runs it with four differences:
• It checks the stocks once a month, not once a week.
• It always holds the best 20 stocks, instead of buying until the money is used up. (It sells a stock when the stock leaves the top 100, which is about the top fifth, as in the book.)
• Like the book, it puts more money into calmer stocks, but it measures calmness in a slightly different way.
• The market rule changes only after the S&P 500 has stayed above or below its 200-day average for five days.
Version B is version A without the market rule. It keeps buying new stocks even when the market is falling.
Version C is version A without the smoothness in rule 4. It ranks stocks only by how fast their price went up.
Comparing A with B shows what the market rule did, and comparing A with C shows what the smoothness did. Both comparisons were written down and locked before the first run: A against B in this study, and A against C in a companion study run beside it. The record at the end of this page lists both. Table 1 shows the four versions side by side.
1 Methodology, in detail (click to open)
1 Methodology
Universe. The S&P 500 as its members stood on each day the stocks are chosen. Every member with a price history in our data is included.
Version A, each rule on a catalog card's setting:
• Skip a stock below its 100-day moving average (Distance to Moving Average).
• Skip a stock whose largest one-day move in the last 90 days is more than 15% (Volatility, largest one-day move).
• Rank the rest whose 90-day trend points up. The score is the yearly growth of a straight line fitted to the logs of the last 90 closing prices, multiplied by the line's R2. R2 measures how closely the prices follow the line (Trend Slope, with its setting that counts how straight the line is).
• Hold the best 20, and keep a held stock while it ranks in the top 100 (Top-N).
• Give each stock money in inverse proportion to its volatility, with at most 15% in one stock (Portfolio Sigma).
• The market rule: buy new stocks only while SPY is above its 200-day moving average. A cross counts once it has held for five days (SPY DMA Regime, the line only). Below the line, A keeps the stocks it holds while they still qualify, and money from sales stays in cash.
Versions B and C. B is A with the market rule held open, so it buys new stocks whatever SPY does. C is A ranked by the line's growth without R2. Each is the same circuit with one setting changed. A against B is this study; A against C is its companion study (477be926d0a9), locked and run the same way.
Trading. The stocks are chosen at the first close of each year and then every month. Trades happen at the next close. Every trade pays 0.1% of its value. Returns include dividends, SPY's too. Cash earns nothing.
The test. Twenty one-year periods from 2006 to 2025, with every setting fixed before the first period ran. Each year starts in cash and buys at its first close. So the first trading day of each later year is carried with each portfolio as it stood at the previous year's last close, and SPY with its own closes. That way SPY's years are its calendar years. The record's own tables leave that first day out, so they read a little lower: A grows 9.2% a year there and 9.5% here.
For comparison only. RSP, the fund that holds every S&P 500 stock in equal amounts, appears in Tables 2 to 5 and Figure 2. Its returns come from its own closes, with dividends, on the same dates. It was not part of the test.
2 Results
Findings
How the money grew. Figure 1 shows what one dollar became in each version and in SPY. Until 2014, version A was ahead of SPY, and most of that lead came from 2008. That year the market rule kept A out of the market. A lost almost nothing, while SPY lost more than a third of its value. If we leave 2008 out, A was behind SPY from 2006 to 2014 as well. A made 13.6% a year, and SPY made 15.1% a year. After 2015, SPY grew faster than A. It passed A in 2023 and stayed ahead. By the end of 2025, one dollar in A had become $6.11. One dollar in SPY had become $7.80. None of our three versions beat simply holding SPY over the twenty years.
Where the difference came from. Figure 2 splits the difference between A and SPY into three parts, and the three parts add up to the whole difference (Table 5). To explain them, we use one more fund: RSP. RSP owns every S&P 500 stock in equal amounts, so it shows how the average stock did.
• Stock picking. Here we compare version B with RSP. Version B picks stocks by the book's rules, without the market rule. Version B did slightly worse than RSP, both before 2015 and after. So the book's way of picking stocks did not beat simply owning all of them, before the book came out or after.
• The market rule. Before 2015, it added almost three percentage points a year, nearly all of it in 2008. After 2015, it cost about two and a half points a year.
• Big companies. SPY puts most of its money into the largest companies, while RSP puts the same amount into every company. Before 2015, RSP did a little better than SPY. After 2015, the largest companies grew much faster, and SPY did about three points a year better than RSP. Version A holds 20 stocks with similar amounts in each, so it could not profit from this.
Each rule, year by year. Figure 3 shows what each rule added in each year. The market rule helped a lot in 2008, and a little in two other years. It cost money in ten years. It cost the most in 2009, 2012, 2019 and 2020. In each of those years, the market fell and then recovered quickly, and the market rule stopped A from buying while prices were already rising again. From 2006 to 2014, the smoothness made almost no difference overall. In single years it moved the result by up to six points, up or down. From 2015 to 2021, it helped in six of the seven years. Since 2022, it has cost a little every year.
The biggest falls. A's biggest fall from a high point was 30%, in March 2020. Without the market rule, the biggest fall was 48%, during the 2008 crash. SPY's biggest fall was 55% (Table 6). Since 2010, A's biggest fall was about as large as SPY's, and A's value moved up and down a little more from day to day.
2.1 Headline
Table 1. The book's rules and the three versions we ran. A is our copy of the book, with four differences. B and C each change one rule of A. Every version pays 0.1% on every trade and includes dividends.
| Rule | Clenow's book | A: the book's rules | B: no market rule | C: no smoothness |
|---|---|---|---|---|
| Stocks | the S&P 500 | the S&P 500 as it was each day | same as A | same as A |
| Ranked by | speed and smoothness, 90 days | speed and smoothness, 90 days | same as A | speed only |
| Skipped | below 100-day average; 15% jumps | below 100-day average; 15% jumps | same as A | same as A |
| How many | until the money is used up | the best 20 | same as A | same as A |
| Sold when | out of the top fifth | out of the top 100 (a fifth) | same as A | same as A |
| Money per stock | more in calm stocks | more in calm stocks | same as A | same as A |
| Market rule | buy only above 200-day average | the same, after 5 days | off | same as A |
| Checked | every week | every month | same as A | same as A |
Table 2. Each year's return before the book came out, dividends included. A, B and C pay 0.1% on every trade. RSP owns every S&P 500 stock in equal amounts and is shown for comparison. In 2008, A and C held cash all year; their small losses come from that year's first trading day.
| Year | A: book's rules | B: no market rule | C: no smoothness | RSP (equal amounts) | Holding SPY |
|---|---|---|---|---|---|
| 2006 | +5.2% | +4.9% | +0.5% | +13.5% | +13.8% |
| 2007 | +11.6% | +14.3% | +12.5% | +0.9% | +5.1% |
| 2008 | −0.5% | −40.5% | −0.4% | −40.0% | −36.8% |
| 2009 | +32.5% | +47.3% | +37.2% | +44.7% | +26.4% |
| 2010 | +9.3% | +12.0% | +9.4% | +21.5% | +15.1% |
| 2011 | −9.4% | −4.3% | −9.6% | +0.2% | +1.9% |
| 2012 | +7.0% | +16.1% | +5.4% | +18.5% | +16.0% |
| 2013 | +42.1% | +42.1% | +47.9% | +36.7% | +32.3% |
| 2014 | +18.5% | +18.5% | +19.2% | +15.1% | +13.5% |
Table 3. The same for the years after the book came out.
| Year | A: book's rules | B: no market rule | C: no smoothness | RSP (equal amounts) | Holding SPY |
|---|---|---|---|---|---|
| 2015 | +4.5% | +8.1% | −0.6% | −1.7% | +1.3% |
| 2016 | +11.9% | +10.5% | +9.8% | +14.5% | +12.0% |
| 2017 | +18.3% | +18.3% | +13.3% | +18.5% | +21.7% |
| 2018 | −1.5% | −4.9% | −7.3% | −7.8% | −4.6% |
| 2019 | +4.6% | +14.8% | +0.5% | +28.9% | +31.2% |
| 2020 | +9.5% | +18.8% | +11.0% | +12.7% | +18.4% |
| 2021 | +32.2% | +32.4% | +26.1% | +29.4% | +28.7% |
| 2022 | −14.3% | −10.2% | −13.4% | −11.6% | −18.2% |
| 2023 | +1.3% | +4.7% | +2.5% | +13.7% | +26.2% |
| 2024 | +9.6% | +9.4% | +12.9% | +12.8% | +24.9% |
| 2025 | +12.7% | +13.0% | +15.4% | +11.2% | +17.7% |
Table 4. How much each one made per year, dividends included. Without 2008, from 2006 to 2014, A made 13.6% a year and SPY made 15.1% a year.
| Growth a year | 2006 to 2014 | 2015 to 2025 | 2006 to 2025 |
|---|---|---|---|
| A: the book's rules | 12.0% | 7.5% | 9.5% |
| B: without the market rule | 9.2% | 9.9% | 9.6% |
| C: without the smoothness | 12.3% | 5.8% | 8.7% |
| RSP (equal amounts) | 9.5% | 10.2% | 9.9% |
| Holding SPY | 7.7% | 13.4% | 10.8% |
Table 5. The difference between A and SPY, in three parts that add up exactly. Stock picking: version B against RSP, the fund that owns every S&P 500 stock in equal amounts. Market rule: version A against version B. Big companies: RSP against SPY. Each part comes from the growth rates in Table 4.
| Percentage points a year | 2006 to 2014 | 2015 to 2025 |
|---|---|---|
| Stock picking | −0.3 | −0.3 |
| Market rule | +2.8 | −2.4 |
| Big companies | +1.8 | −3.2 |
| Total: A against SPY | +4.3 | −5.9 |
Table 6. The biggest fall from a high point, and how much the value typically moves up and down in a year (the standard deviation of daily returns, scaled to a year). Shown for all twenty years and since 2010. A's biggest fall came in March 2020. B's and SPY's came in the 2008 crash.
| Biggest fall 2006-25 | Biggest fall 2010-25 | Ups and downs 2006-25 | Ups and downs 2010-25 | |
|---|---|---|---|---|
| A: the book's rules | −30% | −30% | 18% | 19% |
| B: without the market rule | −48% | −30% | 21% | 20% |
| C: without the smoothness | −37% | −37% | 19% | 20% |
| Holding SPY | −55% | −34% | 19% | 17% |
Sections 2.2 to 4, the full record: every year, every test, and how each one was run (click to open)
The registration names the two groups compared The book and Without the market switch; this paper calls them A: the book's rules and B: without the market rule.
2.2 Per-step results
| # | Out-of-sample window | A: the book's rules SR | B: without the market rule SR |
|---|---|---|---|
| 1 | 2006-01-03 → 2006-12-29 | 0.41 | 0.38 |
| 2 | 2007-01-03 → 2007-12-31 | 0.64 | 0.73 |
| 3 | 2008-01-02 → 2008-12-31 | n/a | -1.41 |
| 4 | 2009-01-02 → 2009-12-31 | 1.64 | 1.48 |
| 5 | 2010-01-04 → 2010-12-31 | 0.47 | 0.53 |
| 6 | 2011-01-03 → 2011-12-30 | -0.36 | -0.08 |
| 7 | 2012-01-03 → 2012-12-31 | 0.51 | 1.03 |
| 8 | 2013-01-02 → 2013-12-31 | 2.17 | 2.17 |
| 9 | 2014-01-02 → 2014-12-31 | 1.17 | 1.17 |
| 10 | 2015-01-02 → 2015-12-31 | 0.39 | 0.57 |
| 11 | 2016-01-04 → 2016-12-30 | 0.91 | 0.77 |
| 12 | 2017-01-03 → 2017-12-29 | 1.20 | 1.20 |
| 13 | 2018-01-02 → 2018-12-31 | -0.05 | -0.21 |
| 14 | 2019-01-02 → 2019-12-31 | 0.41 | 1.20 |
| 15 | 2020-01-02 → 2020-12-31 | 0.43 | 0.67 |
| 16 | 2021-01-04 → 2021-12-31 | 1.37 | 1.37 |
| 17 | 2022-01-03 → 2022-12-30 | -0.72 | -0.39 |
| 18 | 2023-01-03 → 2023-12-29 | 0.16 | 0.45 |
| 19 | 2024-01-02 → 2024-12-31 | 0.67 | 0.67 |
| 20 | 2025-01-02 → 2025-12-31 | 0.70 | 0.66 |
2.2b Every test, in numbers
Every test this paper registered, two rows each, the paper’s own test first.
| Walk | Windows | Span | Growth | CAGR | Worst drawdown | Pooled Sharpe |
|---|---|---|---|---|---|---|
| A: the book's rules (in the market rule test) (this paper, the one the lab opens) | 20 | 2006-01-03 → 2025-12-31 | +482.9% | +9.2% | -30.2% | 0.57 |
| B: without the market rule (this paper, the one the lab opens) | 20 | 2006-01-03 → 2025-12-31 | +503.2% | +9.4% | -48.2% | 0.53 |
| A: the book's rules (in the smoothness test) | 20 | 2006-01-03 → 2025-12-31 | +482.9% | +9.2% | -30.2% | 0.57 |
| C: without the smoothness | 20 | 2006-01-03 → 2025-12-31 | +398.1% | +8.4% | -37.5% | 0.52 |
| platform reference (SPY) (benchmark) | 2006-01-03 → 2025-12-31 | +628.8% | +10.4% | -56.1% |
2.3 Search accounting
This paper's search is a declared family: the paper's 2 walks, counted at N = 2 evaluated books. Every member is either a registered walk with its own hypothesis and frozen record, or a derived average computed from those frozen records; every member is reported, in the family table, and none was selected away. The count is declared by the author rather than derived from one project's ledger, because the members are sibling registered studies; the declaration names them and is frozen in this artifact. What the source strategy's author searched before publishing is not knowable from here and is not counted. The registered per-step record below still guarantees each window's hypothesis was hashed and registered before that window was scored.
2.4 The comparison
Both arms trade the same registered windows, so their returns can be PAIRED: inside each window the two return series are inner-joined date by date and the difference rA: the book's rules − rB: without the market rule is the object under test. Because this is ONE pre-declared contrast, frozen at registration before any window was scored, the paired statistic needs no multiple-testing deflation; the arm-level records carry the declared family count of §2.3 as their search accounting, and this contrast, registered per window before scoring, is not multiplied by it.
In the table: Arm A = A: the book's rules · Arm B = B: without the market rule.
| # | Window | Paired bars | Arm A | Arm B | Δ | Leader |
|---|---|---|---|---|---|---|
| 1 | 2006-01-04 → 2006-12-29 | 250 | +5.2% | +4.9% | +0.3 pp | Arm A |
| 2 | 2007-01-04 → 2007-12-31 | 250 | +11.3% | +13.7% | -2.4 pp | Arm B |
| 3 | 2008-01-03 → 2008-12-31 | 252 | +0.0% | -39.8% | +39.8 pp | Arm A |
| 4 | 2009-01-05 → 2009-12-31 | 251 | +32.5% | +44.0% | -11.4 pp | Arm B |
| 5 | 2010-01-05 → 2010-12-31 | 251 | +7.2% | +9.9% | -2.7 pp | Arm B |
| 6 | 2011-01-04 → 2011-12-30 | 251 | -10.4% | -5.2% | -5.2 pp | Arm B |
| 7 | 2012-01-04 → 2012-12-31 | 249 | +7.2% | +16.4% | -9.2 pp | Arm B |
| 8 | 2013-01-03 → 2013-12-31 | 251 | +37.6% | +37.6% | +0.0 pp | tie |
| 9 | 2014-01-03 → 2014-12-31 | 251 | +19.2% | +19.2% | +0.0 pp | tie |
| 10 | 2015-01-05 → 2015-12-31 | 251 | +4.9% | +8.5% | -3.6 pp | Arm B |
| 11 | 2016-01-05 → 2016-12-30 | 251 | +13.3% | +12.5% | +0.8 pp | Arm A |
| 12 | 2017-01-04 → 2017-12-29 | 250 | +17.0% | +17.0% | +0.0 pp | tie |
| 13 | 2018-01-03 → 2018-12-31 | 250 | -2.3% | -5.6% | +3.4 pp | Arm A |
| 14 | 2019-01-03 → 2019-12-31 | 251 | +4.7% | +16.5% | -11.8 pp | Arm B |
| 15 | 2020-01-03 → 2020-12-31 | 252 | +9.0% | +18.3% | -9.3 pp | Arm B |
| 16 | 2021-01-05 → 2021-12-31 | 251 | +34.1% | +34.1% | +0.0 pp | tie |
| 17 | 2022-01-04 → 2022-12-30 | 250 | -15.3% | -11.2% | -4.1 pp | Arm B |
| 18 | 2023-01-04 → 2023-12-29 | 249 | +1.3% | +6.2% | -4.8 pp | Arm B |
| 19 | 2024-01-03 → 2024-12-31 | 251 | +10.9% | +10.9% | +0.0 pp | tie |
| 20 | 2025-01-03 → 2025-12-31 | 249 | +12.2% | +12.6% | -0.3 pp | Arm B |
Paired Sharpe of the difference track: -0.07 · block bootstrap (2000 paths, block 10, seed 1234): P(A: the book's rules beats B: without the market rule) = 34.5%.
Window win-rate. A: the book's rules led 4 of 20 windows (20.0%), B: without the market rule led 11, and 5 windows were ties, and the mean window gap of -1.03 pp points the same way. Widest single window: 2008 at +39.8 pp.
| Period | Windows | A: the book's rules | B: without the market rule | Mean gap | A: the book's rules led |
|---|---|---|---|---|---|
| All windows | 20 | +9.99% | +11.02% | -1.03 pp | 4/20 |
| Before 2015 | 9 | +12.21% | +11.19% | +1.02 pp | 2/9 |
| 2015 onward | 11 | +8.18% | +10.89% | -2.71 pp | 2/11 |
The two eras disagree by 3.73 pp. The pooled figure is therefore not a standing property of either method, it is dominated by the later period. Read the two rows, not the average.
3 The circuit
The strategy is a circuit of platform primitives, frozen when the study is registered. Below is the circuit as wired on the canvas, the objective it encodes and how the search runs through it, followed by the mathematics each primitive actually computes, the same formulas the execution engine runs. The complete parameterisation is preserved in the study ledger (Appendix A).
The sentence below is the registration record, generated when the circuit was registered and printed verbatim; the authored description of the design is Section 1.
A COMPARATIVE study: The book vs Without the market switch, walked on the same registered out-of-sample windows. The book: S&P 500, selected by statistical / factor criteria, rebalanced monthly across the selected basket, conditioned on the wired regime classifier, and run out-of-sample from the anchor: anything the design estimates from history, where it estimates at all, is re-estimated at each anchor from pre-anchor data only, and the walk advances through registered out-of-sample windows; its disposition is the realized forward path versus the benchmark. Without the market switch: S&P 500, selected by statistical / factor criteria, rebalanced monthly across the selected basket, conditioned on the wired regime classifier, and run out-of-sample from the anchor: anything the design estimates from history, where it estimates at all, is re-estimated at each anchor from pre-anchor data only, and the walk advances through registered out-of-sample windows; its disposition is the realized forward path versus the benchmark. The arms differ in: SPY DMA Regime, Force regime (override): auto → calm. The contrast under test: whether The book generates better risk-adjusted returns than Without the market switch over the identical out-of-sample windows.
Every block in this study is a card from the platform's catalog. Each of the book's rules is one card's setting: the 100-day average, the 15% jump limit, the 90-day speed with or without the smoothness, the best 20 kept while in the top 100, more money in calmer stocks, and the market rule on SPY's 200-day average. You can rebuild every test and change any setting.
Envelopes show counts, ratios, dates, and the parameters the author chose. Full price and per-name data series are not republished: the underlying market data is licensed to QuanterLab. Point figures quoted in the prose, a named holding's return over a stated span, are summary facts derived from public market prices, not redistributed series.
The objective and the search
The book
| Universe | S&P 500 index constituents. |
|---|---|
| Selection | serial-gated across Distance from the 100-day average, Largest one-day move (90 days), Trend speed times how straight it is (Clenow, 90 days) → highest 20 kept by Trend speed times how straight it is (Clenow, 90 days), a name already held kept while it ranks in the top 100. |
| Validation & out-of-sample | portfolio forward test (holds the weights it is given, re-set at each rebalance) (1y horizon from the anchor, monthly rebalance, each decision traded at the next day's close, dividends counted, the regime deciding only whether new names are bought at a rebalance (a held name stays while it is still selected)); overlays: Transaction Cost. |
| Regime layer | macro regime detected via SPY above or below its 200-day average (the line only). |
| Other components | Combine: Portfolio Σ. |
Without the market switch
The specification is identical to The book's table above, row for row; the one registered difference between the arms is itemized below.
What differs between the arms, one difference; the comparison is clean:
- paramSPY DMA Regime, Force regime (override): auto → calm
Everything else is held identical, so an out-of-sample gap between the arms is attributable to this one change.
Cost elements are wired into the circuit, the realised drag is reported per step in Appendix B.
Show the mathematics, 10 primitives, formulas and parity notes
3.1 Universe
The starting set of tickers, resolved point-in-time from the index change-log, so names delisted or removed later still compete on the dates they traded.
Before any math, you need a list of stocks. An index preset (S&P 500, Nasdaq-100, Dow 30) is reconstructed as it stood ON your anchor date by replaying the historical add/drop change-log backwards, so a 2018 backtest sees the 2018 membership, not today's winners.
Start from today's constituents and un-apply every membership change after the anchor t:
\mathcal{U}(t) = \mathcal{U}_{\text{now}} \;\ominus\; \{\text{adds after } t\} \;\oplus\; \{\text{drops after } t\}3.2 Price Loader
Bulk OHLCV fetch for the whole universe, point-in-time, no future bars.
Momentum, volatility, trend, every price-based metric needs history. This loads open/high/low/close/volume for all names in parallel, clipped so nothing after the anchor can leak in. The lookback window is derived automatically from the deepest metric you wired.
It loads exactly enough history for the hungriest downstream metric plus a warm-up buffer:
W = \max_k(\text{lookback}_k) + \text{buffer}, \qquad \text{bars} \le \text{anchor } t3.3 Filter Dist 200Dma
Distance from its moving average, the long-term trend gauge.
How far above (or below) its moving average a stock trades, 200 days unless you set the length. Positive means it's in an uptrend; the 200-day line is the classic line that separates bull from bear for a single name, and Clenow's Stocks on the Move uses 100.
d = \frac{P_t}{\text{SMA}_{n}(P)_t} - 1, \qquad \text{SMA}_{n} = \frac{1}{n}\sum_{i=0}^{n-1} P_{t-i}3.4 Filter Volatility
Annualized volatility, the size of the wiggle.
The standard deviation of daily log-returns, scaled to a yearly number by √252 (trading days). A risk and regime descriptor used everywhere downstream. Set the measure to "Largest one-day move" for the biggest single-day jump instead: Clenow's Stocks on the Move skips a stock that moved more than 15% in a day in the last 90.
\sigma_{\text{ann}} = \operatorname{std}(r_t)\,\sqrt{252}, \qquad r_t = \ln\frac{P_t}{P_{t-1}}m = \max_{t-n < \tau \le t} \left|\frac{P_\tau}{P_{\tau-1}} - 1\right|3.5 Filter Trend Slope
Trend slope, the annualized direction and steepness.
The slope of a straight line fitted to log-price, scaled to a yearly rate. Signed: strongly positive is a powerful uptrend, negative a downtrend. The line is fitted over about six months unless you set the window. With "Steadiness counts" on, it is the ranking of Andreas Clenow's Stocks on the Move: the yearly speed times how straight the line is, so a steady climb ranks above a jumpy one of the same speed.
\text{slope}_{\text{ann}} = b \times 252, \qquad \log P_\tau = a + b\,\tau\text{score} = \left(e^{\,252\,b} - 1\right) \times R^23.6 Top N
Keep the best N, rank, then cut.
Sort the survivors by the Composite Σ (or, if none is wired, the last metric in the chain) and keep the top (or bottom) N. The final narrowing from a scored list to a committed basket.
\text{Top-}N = \{\, i : \operatorname{rank}(\text{score}_i) \le N \,\}N = \operatorname{round}(f \cdot n_{\text{ranked}})\text{kept} = \{\, i \in H : \operatorname{rank}_i \le K \,\} \;\cup\; \text{best ranked } i \notin H \text{ up to } N3.7 Portfolio Builder
From "which names" to "how much of each", the weighting rules.
Turns a basket into capital weights by a simple rule: equal weight, score-weighted, inverse-volatility, or cap-weighted. An optional max-weight cap waterfills the excess onto the under-cap names so nothing is over-concentrated.
w_i^{\text{eq}}=\tfrac1N,\quad w_i^{\text{score}}\propto \text{score}_i,\quad w_i^{\text{ivol}}\propto \tfrac{1}{\sigma_i},\quad w_i^{\text{cap}}\propto \text{mktcap}_iClip any weight above the cap, redistribute the excess proportionally to names still under the cap, and repeat until none exceed it (≤ 24 rounds).
w_i \leftarrow \min(w_i, w_{\max}); \;\; \text{redistribute excess} \propto w_j\,[w_j < w_{\max}]3.8 Portfolio Backtest
Replay the portfolio forward, rebalanced, point-in-time, with costs.
Holds the basket and rebalances on schedule, re-selecting and re-optimizing point-in-time at each rebalance (so it only ever uses information available then), and reports the equity curve, Sharpe, drawdown and trade stats, optionally net of cost and risk overlays.
E_t = E_{t-1}\big(1 + \mathbf w_{t}^{\top}\mathbf r_t - \text{costs}_t\big)\text{DD}_t = \frac{E_t}{\max_{\tau\le t}E_\tau} - 1, \qquad \text{MaxDD} = \min_t \text{DD}_t\text{charge}_t \;=\; \text{loan}_t \cdot \frac{\text{spread}}{252}, \qquad \text{loan}_t = \begin{cases}\max(0,\,-\text{cash}_t) & \text{institutional (netted)}\\ \max(0,\,\text{long MV}_t - E_t) & \text{retail (no netting)}\end{cases}\mathbf w_k = \{\, w_{k,i} : i \in \text{held} \,\}, \qquad \text{cash}_k = 1 - \textstyle\sum_{i \in \text{held}} w_{k,i}3.9 Transaction Cost
Charge for trading, slippage + commission on every turn.
Real trading isn't free. This deducts a cost proportional to how much you trade (turnover), in basis points, so the backtest reflects net, not gross, performance.
\text{cost}_t = \frac{\text{bps}}{10{,}000}\;\times\;\text{turnover}_t, \qquad \text{turnover}_t = \tfrac12\sum_i \lvert w_{i,t}-w_{i,t^-}\rvert3.10 Spy Dma Regime
Price vs the 200-day average, the classic bull/bear line.
Looks at the benchmark relative to its 200-day moving average and the slope of that average. Above and rising is calm (bull); below and falling is stressed (bear); mixed is choppy.
P > \text{SMA}_{200}\ \wedge\ \tfrac{d}{dt}\text{SMA}_{200} > 0 \;\Rightarrow\; \text{calm}P > \text{SMA}_{200} \;\Rightarrow\; \text{calm}, \qquad P \le \text{SMA}_{200} \;\Rightarrow\; \text{stressed}4 Projection calibration, pooled across the walk
Every rebalance carried a Monte Carlo cone and a 95% VaR estimated before the segment it is scored against. Two questions, pooled over the whole study: did realized outcomes land inside the band as often as the band claims, and were VaR breaches as frequent as 5%?
This section is produced by the forward tester itself: every portfolio backtest fits the cone and the VaR estimate at each rebalance and scores them against the segment that followed. It does not require, and this circuit does not contain, a Monte Carlo primitive; that primitive is a separate, standalone analysis.
| Arm | Steps | Rebalances | In band | Coverage | Expected | VaR days | Breach rate | Expected |
|---|---|---|---|---|---|---|---|---|
| A: the book's rules | 20 | 255 | 195 / 212 | 92.0% ±1.87 | 90.0% | 4237 | 6.02% ±0.365 | 5.0% |
| B: without the market rule | 20 | 255 | 216 / 240 | 90.0% ±1.94 | 90.0% | 4791 | 6.35% ±0.352 | 5.0% |
± values are binomial standard errors on the estimate. A coverage figure below the expected band means the projection was over-confident; a breach rate above 5% means the same of the risk model. Both forecasts used only data prior to the segment scored.
5 Discussion
5.1 Interpretation
Why did the rules fall behind after 2015? A published rule can do worse after it is published for two common reasons. The first is that many people start to use it, and the advantage gets used up. Research on published investing rules finds this often (McLean and Pontiff, 2016). The second is that the rules fit the years they were made from better than they fit the years after. This is called overfitting, and we explain it below.
Our test does not support the first reason. The book's stock picking did slightly worse than RSP both before and after the book came out, so there was no advantage in the picking that people could use up. The lead before 2015 came from one year, 2008, and from the market rule. After 2015 there was no slow crash like 2008: the falls were short, and the market recovered quickly. And after 2015 the largest companies grew much faster than the rest of the market, while SPY holds mostly those companies.
Is the market rule worth it? It works like insurance: it costs a little in most years and pays off in a crash. In twenty years it paid off once, in 2008, and then it paid off well. It cost the most when the market fell and then recovered quickly. The reason: version A buys again only after the S&P 500 has been back above its average for five days, and then it still waits for its next monthly check. The book checks every week, so it would have started buying again up to three weeks earlier, and its market rule probably cost less in those recoveries. Whether this insurance is worth having depends on how much another crash like 2008 would hurt you.
The smoothness. We expected the smoothness to make the least difference. In our earlier study, the frog in the pan, we looked at last year's winning stocks and checked how they had gone up during that year. In the S&P 500, the stocks that went up in a few big jumps did better than the stocks that went up smoothly. Here the smoothness is measured over three months, and it helped in most years from 2015 to 2021. Since 2022, ranking by speed only has done better. We read this as a small tendency that may not last.
5.2 Limitations
Our price data does not include companies that later left the S&P 500, for example after a bankruptcy. So a stock that collapsed and was removed from the index is missing from the months when the rules might have picked it. This gap is biggest in the early years. In 2006, about a third of the index's members have no prices in our data, and by 2020 almost all of them do. The missing companies probably make A, B and C look a little better than they were. The records of SPY and RSP are complete, so the comparison with SPY stands, and the stock picking before 2015 was, if anything, a little worse than the comparison with RSP shows.
Version A differs from the book in four ways, listed in the rules section. The weekly check matters most: it would have bought back sooner after the quick recoveries. Version C's comparison was locked in a companion study (477be926d0a9), run beside this one with the same settings; its numbers are in the record's table of every test. Cash earns no interest in our test. With the interest rates of 2008 and 2023, it would have added a little. In 2008, versions A and C held cash all year. Their small loss that year comes from the first trading day of 2008, when they still held the stocks they owned at the end of 2007. Every trade costs 0.1% of its value. Twenty years give us only twenty yearly results, and the market rule's record depends mostly on one crash.
Why a strategy can look better in a backtest than later
Online you can find many charts of strategies that beat the market by a wide margin. Some of them hold up later. Many do not, and often nobody meant to mislead anyone. The usual cause is overfitting.
Overfitting happens when rules are chosen while looking at the same years they are tested on. If you try enough rules on the past, some of them will fit the past very well, partly by luck. The test then looks excellent, but the luck does not come back in the years after.
This is not a judgment of this book or its author. Any rule tested on the years it was made from can look better than it will later, and this paper shows what that can look like. Up to 2014, the rules were well ahead of the market. Most of that lead came from one year and one rule: the market rule kept the money out of the 2008 crash. The stock picking itself did slightly worse than owning every stock in equal amounts.
Three checks help with any strategy chart you see:
• Look at the years after the rules were made. Those years are the fair test.
• Look at the results year by year. If one or two years make most of the lead, the lead is fragile.
• Compare with simple choices: holding SPY, and owning every stock in equal amounts. A strategy should beat both to be worth the trouble.
Our lab does all three for any rule you build. It tests on years the rules never saw, shows every year on its own, and puts SPY beside every result.
Try it yourself
Every test in this paper can be run again on quanterlab.com, and you can change any setting. Open it in the lab and try, for example:
• turn the market rule off;
• rank by speed only;
• use a 50-day average instead of 100 days;
• change the 15% jump limit to 10%;
• run the rules on the Nasdaq 100.
Then see what each rule does in the years you choose.
Computed, not featured
The run produced the results below. The author chose not to feature them in this paper; they were computed all the same, are part of the frozen record, and are reproducible from the study's frozen circuit.
- The measurement ladder, 20 windows. The forward tester, per window: the same walk chained price-only, with dividends and net of costs, against the benchmark measured both ways.
References
- Clenow (2015), Stocks on the Move: Beating the Market with Hedge Fund Momentum Strategies. The rules tested here.
- Radge (2015), Momentum Investing, The Chartist, 3 October 2015. A test of the rules on 1999 to 2014: 12.3% a year against 5.2% for the S&P 500, biggest fall 24% against 55%. https://www.thechartist.com.au/momentum-investing/
- McLean and Pontiff (2016), Does Academic Research Destroy Stock Return Predictability?, Journal of Finance 71(1): 5-32. Published investing rules often do worse after publication. https://doi.org/10.1111/jofi.12365
- QuanterLab (2026), The Frog in the Pan Says Keep the Steady Climbers. In the S&P 500, It Works the Other Way Around. How last year's winning stocks went up, measured over a whole year. https://quanterlab.com/research/the-frog-in-the-pan-says-keep-the-steady-climbers-in-the-s-p-500-it-works-the
- Jegadeesh and Titman (1993), Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, Journal of Finance 48(1): 65-91. Stocks that went up over the past months tended to keep going up. https://doi.org/10.1111/j.1540-6261.1993.tb04702.x
- Faber (2007), A Quantitative Approach to Tactical Asset Allocation, Journal of Wealth Management 9(4): 69-79. Holding the market only while it is above its long-term average. https://doi.org/10.3905/jwm.2007.674809
Appendix A Reproducibility in QuanterLab
Each step is backed by a frozen run report. The study is re-derivable from the ledger below.
| # | Commit | Report | Anchor | OOS window |
|---|---|---|---|---|
| 1 | 22769d86996a | 14824 | 2006-01-01 | 2006-01-03 → 2006-12-29 |
| 2 | 737f3ba32851 | 14827 | 2007-01-01 | 2007-01-03 → 2007-12-31 |
| 3 | f064e5360098 | 14829 | 2008-01-01 | 2008-01-02 → 2008-12-31 |
| 4 | 3e95beccbf90 | 14831 | 2009-01-01 | 2009-01-02 → 2009-12-31 |
| 5 | 87b26501047d | 14833 | 2010-01-01 | 2010-01-04 → 2010-12-31 |
| 6 | 5360ee105525 | 14835 | 2011-01-01 | 2011-01-03 → 2011-12-30 |
| 7 | 09b822e03585 | 14837 | 2012-01-01 | 2012-01-03 → 2012-12-31 |
| 8 | 6e02ef26270f | 14838 | 2013-01-01 | 2013-01-02 → 2013-12-31 |
| 9 | bf6af9209a41 | 14840 | 2014-01-01 | 2014-01-02 → 2014-12-31 |
| 10 | 4e1db4101a73 | 14843 | 2015-01-01 | 2015-01-02 → 2015-12-31 |
| 11 | 15daf7abc085 | 14844 | 2016-01-01 | 2016-01-04 → 2016-12-30 |
| 12 | 96a1224e9368 | 14846 | 2017-01-01 | 2017-01-03 → 2017-12-29 |
| 13 | 05588bc85e0c | 14849 | 2018-01-01 | 2018-01-02 → 2018-12-31 |
| 14 | fedcdae88aa9 | 14851 | 2019-01-01 | 2019-01-02 → 2019-12-31 |
| 15 | b510ceefc4f5 | 14853 | 2020-01-01 | 2020-01-02 → 2020-12-31 |
| 16 | 830699089a9a | 14854 | 2021-01-01 | 2021-01-04 → 2021-12-31 |
| 17 | b0bddc39a3b5 | 14857 | 2022-01-01 | 2022-01-03 → 2022-12-30 |
| 18 | 1868e3283751 | 14858 | 2023-01-01 | 2023-01-03 → 2023-12-29 |
| 19 | 524a92cacb08 | 14860 | 2024-01-01 | 2024-01-02 → 2024-12-31 |
| 20 | 75628ecaad53 | 14863 | 2025-01-01 | 2025-01-02 → 2025-12-31 |
The methods the platform implements, for every study made on it, rest on these works.
- Gelman, A., & Loken, E. (2013). The garden of forking paths: Why multiple comparisons can be a problem, even when there is no “fishing expedition.” Working paper, Columbia University.
- Harvey, C. R., Liu, Y., & Zhu, H. (2016). … and the Cross-Section of Expected Returns. Review of Financial Studies, 29(1), 5–68. doi:10.1093/rfs/hhv059
- Lo, A. W. (2002). The Statistics of Sharpe Ratios. Financial Analysts Journal, 58(4), 36–52. doi:10.2469/faj.v58.n4.2453
Appendix A2 Registration record
What this record does and does not establish. Every window in this study is historical: the data existed before the study began, so this is sequential registration on past windows, not pre-registration in the clinical-trial sense, and no procedure could make it so. What the platform does enforce is order, each step's specification was frozen and hashed before that step was scored, and the walk cannot advance past a step that was never run or close one with a result registered for a different window. The two timestamp columns below are the evidence: read them together and each registration precedes its own run, and each run precedes the next registration. A study whose registrations all post-date its runs would show it here. Wall-clock spacing between registrations varies with the author's schedule and queue latency; the ordering, not the tempo, is the claim.
“A COMPARATIVE study: The book vs Without the market switch, walked on the same registered out-of-sample windows. The book: S&P 500, selected by statistical / factor criteria, rebalanced monthly across the selected basket, conditioned on the wired regime classifier, and run out-of-sample from the anchor: anything the design estimates from history, where it estimates at all, is re-estimated at each anchor from pre-anchor data only, and the walk advances through registered out-of-sample windows; its disposition is the realized forward path versus the benchmark. Without the market switch: S&P 500, selected by statistical / factor criteria, rebalanced monthly across the selected basket, conditioned on the wired regime classifier, and run out-of-sample from the anchor: anything the design estimates from history, where it estimates at all, is re-estimated at each anchor from pre-anchor data only, and the walk advances through registered out-of-sample windows; its disposition is the realized forward path versus the benchmark. The arms differ in: SPY DMA Regime, Force regime (override): auto → calm. The contrast under test: whether The book generates better risk-adjusted returns than Without the market switch over the identical out-of-sample windows.”
The same hypothesis was registered independently at every step, hashed before each step's out-of-sample window was scored:
| # | Anchor | Registered at (UTC) | Run completed (UTC) |
|---|---|---|---|
| 1 | 2006-01-01 | 2026-10-14 15:10:10 | 2026-10-14 15:26:38 |
| 2 | 2007-01-01 | 2026-10-14 15:26:39 | 2026-10-14 15:39:35 |
| 3 | 2008-01-01 | 2026-10-14 15:39:35 | 2026-10-14 15:50:46 |
| 4 | 2009-01-01 | 2026-10-14 15:50:46 | 2026-10-14 15:57:56 |
| 5 | 2010-01-01 | 2026-10-14 15:57:56 | 2026-10-14 16:07:11 |
| 6 | 2011-01-01 | 2026-10-14 16:07:12 | 2026-10-14 16:16:36 |
| 7 | 2012-01-01 | 2026-10-14 16:16:36 | 2026-10-14 16:25:00 |
| 8 | 2013-01-01 | 2026-10-14 16:25:00 | 2026-10-14 16:36:02 |
| 9 | 2014-01-01 | 2026-10-14 16:36:02 | 2026-10-14 16:46:12 |
| 10 | 2015-01-01 | 2026-10-14 16:46:12 | 2026-10-14 16:56:01 |
| 11 | 2016-01-01 | 2026-10-14 16:56:01 | 2026-10-14 17:05:44 |
| 12 | 2017-01-01 | 2026-10-14 17:05:44 | 2026-10-14 17:16:58 |
| 13 | 2018-01-01 | 2026-10-14 17:16:58 | 2026-10-14 17:26:55 |
| 14 | 2019-01-01 | 2026-10-14 17:26:56 | 2026-10-14 17:37:46 |
| 15 | 2020-01-01 | 2026-10-14 17:37:46 | 2026-10-14 17:46:11 |
| 16 | 2021-01-01 | 2026-10-14 17:46:11 | 2026-10-14 17:57:45 |
| 17 | 2022-01-01 | 2026-10-14 17:57:45 | 2026-10-14 18:08:34 |
| 18 | 2023-01-01 | 2026-10-14 18:08:34 | 2026-10-14 18:16:34 |
| 19 | 2024-01-01 | 2026-10-14 18:16:34 | 2026-10-14 18:27:06 |
| 20 | 2025-01-01 | 2026-10-14 18:27:06 | 2026-10-14 18:37:03 |
Appendix B Per-step diagnostics
Realized in the projection tables below is the risk engine scoring its own forecast: the buy-and-hold return of the segment that followed each rebalance, on the same gross basis the cone was projected on. It is deliberately not the charged, calendar-window total return the study’s tables print, so the two will not reconcile line by line; the cone and its outcome share one basis, which is what a calibration test requires. Each row names its segment’s span so a boundary session is visible.
Names held is the union across the window: the count of distinct instruments the book touched between the window’s first and last session, not the number it held at one time. A book that rotates monthly touches more names than it holds.
What each step's run actually did beyond its return: the portfolio book's rebalancing and cost drag. Harvested from the frozen run reports, present where the circuit produced them. Cost drag is the gap between the step's return before and after its trading costs, in percentage points of the step's starting capital, so on a book that trades every session and compounds it can exceed the step's own net return.
Open the full per-step grid (20 steps: every rebalance, per window)
Step 1 · 2006-01-03 → 2006-12-29
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 82 names held · selection: reselect · 0.4% in cash · turnover 8.8× · cost drag 0.88% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 83.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 8.79% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2006-01-01 → 2006-02-01 | -4.4502% | 1.8648% | 8.8624% | 6.1547% | yes | 1.4693% | 1 / 19 |
| 2006-02-01 → 2006-03-01 | -4.2213% | 2.4027% | 9.8949% | -0.0104% | yes | 1.5733% | 2 / 18 |
| 2006-03-01 → 2006-04-01 | -5.5672% | 2.8816% | 11.7704% | 4.2211% | yes | 1.8475% | 1 / 22 |
| 2006-04-01 → 2006-05-01 | -5.0634% | 2.3324% | 10.7678% | -2.0475% | yes | 1.8172% | 1 / 18 |
| 2006-05-01 → 2006-06-01 | -4.8882% | 2.7824% | 11.5981% | -6.5876% | no | 1.7768% | 5 / 21 |
| 2006-06-01 → 2006-07-01 | -3.8733% | 2.3879% | 9.4773% | 0.3322% | yes | 1.4667% | 4 / 21 |
| 2006-07-01 → 2006-08-01 | -4.3189% | 2.3315% | 9.7247% | -7.9097% | no | 1.5104% | 5 / 19 |
| 2006-08-01 → 2006-09-01 | -3.0034% | 1.7301% | 6.5273% | 2.7307% | yes | 0.9654% | 0 / 22 |
| 2006-09-01 → 2006-10-01 | -2.7602% | 1.5143% | 6.1506% | -0.7338% | yes | 0.9094% | 0 / 19 |
| 2006-10-01 → 2006-11-01 | -2.9862% | 1.8786% | 7.3068% | 1.5007% | yes | 0.9672% | 0 / 21 |
| 2006-11-01 → 2006-12-01 | -3.7571% | 1.6997% | 7.5866% | 4.0818% | yes | 1.2202% | 1 / 20 |
| 2006-12-01 → 2007-01-01 | -4.2495% | 2.0118% | 8.9452% | 0.988% | yes | 1.3911% | 1 / 19 |
| 2007-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 83 names held · selection: reselect · 0.4% in cash · turnover 8.8× · cost drag 0.88%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 83.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.95% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2006-01-01 → 2006-02-01 | -4.4502% | 1.8648% | 8.8624% | 6.1547% | yes | 1.4693% | 1 / 19 |
| 2006-02-01 → 2006-03-01 | -4.2213% | 2.4027% | 9.8949% | -0.0104% | yes | 1.5733% | 2 / 18 |
| 2006-03-01 → 2006-04-01 | -5.5672% | 2.8816% | 11.7704% | 4.2211% | yes | 1.8475% | 1 / 22 |
| 2006-04-01 → 2006-05-01 | -5.0634% | 2.3324% | 10.7678% | -2.0475% | yes | 1.8172% | 1 / 18 |
| 2006-05-01 → 2006-06-01 | -4.8882% | 2.7824% | 11.5981% | -6.5876% | no | 1.7768% | 5 / 21 |
| 2006-06-01 → 2006-07-01 | -3.8733% | 2.3879% | 9.4773% | 0.3322% | yes | 1.4667% | 4 / 21 |
| 2006-07-01 → 2006-08-01 | -4.2038% | 2.1264% | 9.1407% | -6.3289% | no | 1.4936% | 4 / 19 |
| 2006-08-01 → 2006-09-01 | -3.1157% | 1.7859% | 6.7617% | 1.8078% | yes | 1.0424% | 0 / 22 |
| 2006-09-01 → 2006-10-01 | -2.8958% | 1.4477% | 6.1622% | -0.5705% | yes | 0.934% | 0 / 19 |
| 2006-10-01 → 2006-11-01 | -3.1072% | 1.8062% | 7.2917% | 1.6662% | yes | 0.9659% | 0 / 21 |
| 2006-11-01 → 2006-12-01 | -3.5989% | 1.7157% | 7.4406% | 3.8602% | yes | 1.1836% | 1 / 20 |
| 2006-12-01 → 2007-01-01 | -4.1069% | 1.9334% | 8.6067% | 0.9793% | yes | 1.323% | 0 / 19 |
| 2007-01-01 | no segment follows this rebalance, not scored | ||||||
Step 2 · 2007-01-03 → 2007-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 68 names held · selection: reselect · 0.4% in cash · turnover 7.3× · cost drag 0.73% · regime gate active · 1 name dropped at load (69 selected, 68 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 10.46% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2007-01-01 → 2007-02-01 | -3.8608% | 1.8365% | 8.1083% | 0.8563% | yes | 1.3307% | 1 / 19 |
| 2007-02-01 → 2007-03-01 | -3.614% | 1.7808% | 7.8053% | -2.9187% | yes | 1.3482% | 2 / 18 |
| 2007-03-01 → 2007-04-01 | -4.6329% | 1.9074% | 9.3374% | 1.5006% | yes | 1.5679% | 1 / 21 |
| 2007-04-01 → 2007-05-01 | -4.0317% | 1.7451% | 8.1102% | 1.9338% | yes | 1.3451% | 0 / 19 |
| 2007-05-01 → 2007-06-01 | -3.85% | 2.3248% | 9.3101% | 3.3062% | yes | 1.3612% | 1 / 21 |
| 2007-06-01 → 2007-07-01 | -3.5423% | 2.7081% | 9.5035% | -4.1773% | no | 1.4349% | 3 / 20 |
| 2007-07-01 → 2007-08-01 | -4.2334% | 3.2734% | 11.5398% | -1.884% | yes | 1.62% | 2 / 20 |
| 2007-08-01 → 2007-09-01 | -4.6376% | 2.9084% | 10.7733% | 3.1196% | yes | 1.6742% | 5 / 22 |
| 2007-09-01 → 2007-10-01 | -5.0859% | 2.3849% | 10.9125% | 4.4479% | yes | 1.8214% | 1 / 18 |
| 2007-10-01 → 2007-11-01 | -5.0339% | 3.0537% | 11.5295% | 1.8323% | yes | 1.6957% | 2 / 22 |
| 2007-11-01 → 2007-12-01 | -5.009% | 2.8591% | 11.5596% | -0.3841% | yes | 1.7438% | 5 / 20 |
| 2007-12-01 → 2008-01-01 | -5.4189% | 2.6007% | 11.6459% | 3.626% | yes | 1.9121% | 2 / 19 |
| 2008-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 73 names held · selection: reselect · 0.4% in cash · turnover 7.8× · cost drag 0.78% · 1 name dropped at load (74 selected, 73 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 10.46% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2007-01-01 → 2007-02-01 | -3.8608% | 1.8365% | 8.1083% | 0.8563% | yes | 1.3307% | 1 / 19 |
| 2007-02-01 → 2007-03-01 | -3.614% | 1.7808% | 7.8053% | -2.9187% | yes | 1.3482% | 2 / 18 |
| 2007-03-01 → 2007-04-01 | -4.6329% | 1.9074% | 9.3374% | 1.5006% | yes | 1.5679% | 1 / 21 |
| 2007-04-01 → 2007-05-01 | -4.0317% | 1.7451% | 8.1102% | 1.9338% | yes | 1.3451% | 0 / 19 |
| 2007-05-01 → 2007-06-01 | -3.85% | 2.3248% | 9.3101% | 3.3062% | yes | 1.3612% | 1 / 21 |
| 2007-06-01 → 2007-07-01 | -3.5423% | 2.7081% | 9.5035% | -4.1773% | no | 1.4349% | 3 / 20 |
| 2007-07-01 → 2007-08-01 | -4.2334% | 3.2734% | 11.5398% | -1.884% | yes | 1.62% | 2 / 20 |
| 2007-08-01 → 2007-09-01 | -4.6376% | 2.9084% | 10.7733% | 3.1196% | yes | 1.6742% | 5 / 22 |
| 2007-09-01 → 2007-10-01 | -4.4803% | 2.2422% | 9.8545% | 3.7089% | yes | 1.6217% | 1 / 18 |
| 2007-10-01 → 2007-11-01 | -5.0339% | 3.0537% | 11.5295% | 1.8323% | yes | 1.6957% | 2 / 22 |
| 2007-11-01 → 2007-12-01 | -5.009% | 2.8591% | 11.5596% | -0.3841% | yes | 1.7438% | 5 / 20 |
| 2007-12-01 → 2008-01-01 | -5.1251% | 2.4519% | 10.9583% | 4.4045% | yes | 1.7035% | 2 / 19 |
| 2008-01-01 | no segment follows this rebalance, not scored | ||||||
Step 3 · 2008-01-02 → 2008-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 12 constructions · selection: reselect · 100.0% in cash · regime gate active
B: without the market rule
Portfolio book, rebalanced monthly · 12 constructions · 84 names held · selection: reselect · 0.4% in cash · turnover 9.4× · cost drag 0.94%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 58.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 15.77% of 241 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2008-01-01 → 2008-02-01 | -5.1219% | 2.6229% | 11.1774% | -9.1259% | no | 1.825% | 5 / 20 |
| 2008-02-01 → 2008-03-01 | -5.1495% | 1.6543% | 9.234% | -0.3402% | yes | 1.7154% | 2 / 19 |
| 2008-03-01 → 2008-04-01 | -6.1389% | 1.4277% | 9.9286% | -0.3294% | yes | 1.9104% | 3 / 19 |
| 2008-04-01 → 2008-05-01 | -6.9364% | 1.5723% | 11.4517% | 4.261% | yes | 2.1789% | 0 / 21 |
| 2008-05-01 → 2008-06-01 | -7.0152% | 1.757% | 11.5612% | 2.0601% | yes | 2.1823% | 0 / 20 |
| 2008-06-01 → 2008-07-01 | -7.0612% | 2.1387% | 12.4651% | -0.6555% | yes | 2.1818% | 1 / 20 |
| 2008-07-01 → 2008-08-01 | -7.1502% | 2.6125% | 14.0949% | -12.3734% | no | 2.1931% | 5 / 21 |
| 2008-08-01 → 2008-09-01 | -5.778% | 1.4822% | 9.4669% | -0.6183% | yes | 1.832% | 1 / 20 |
| 2008-09-01 → 2008-10-01 | -5.7945% | 0.898% | 8.2169% | -6.9792% | no | 1.6976% | 5 / 20 |
| 2008-10-01 → 2008-11-01 | -5.3718% | 0.8648% | 7.2866% | -11.6793% | no | 1.3885% | 9 / 22 |
| 2008-11-01 → 2008-12-01 | -6.9068% | 0.0175% | 7.8885% | -7.596% | no | 1.6016% | 7 / 18 |
| 2008-12-01 → window end | -20.6074% | 1.66% | 32.1494% | 7.8244% | yes | 4.2032% | 0 / 21 |
Step 4 · 2009-01-02 → 2009-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 49 names held · selection: reselect · 49.6% in cash · turnover 5.1× · cost drag 0.51% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 6 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 0.0% of 122 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2009-01-01 | no segment follows this rebalance, not scored | ||||||
| 2009-02-01 | no segment follows this rebalance, not scored | ||||||
| 2009-03-01 | no segment follows this rebalance, not scored | ||||||
| 2009-04-01 | no segment follows this rebalance, not scored | ||||||
| 2009-05-01 | no segment follows this rebalance, not scored | ||||||
| 2009-06-01 | no segment follows this rebalance, not scored | ||||||
| 2009-07-01 → 2009-08-01 | -19.401% | -1.5158% | 21.8174% | 8.3208% | yes | 4.4981% | 0 / 21 |
| 2009-08-01 → 2009-09-01 | -19.4367% | -0.9178% | 22.3722% | 1.0526% | yes | 5.0229% | 0 / 20 |
| 2009-09-01 → 2009-10-01 | -18.5248% | -0.6122% | 21.73% | 8.6165% | yes | 4.9791% | 0 / 20 |
| 2009-10-01 → 2009-11-01 | -19.172% | -0.0129% | 25.3018% | -1.8489% | yes | 5.0379% | 0 / 21 |
| 2009-11-01 → 2009-12-01 | -19.7129% | -0.371% | 24.6516% | 4.4325% | yes | 5.0853% | 0 / 19 |
| 2009-12-01 → 2010-01-01 | -18.4998% | 0.0061% | 24.2545% | 3.3798% | yes | 4.5853% | 0 / 21 |
| 2010-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 74 names held · selection: reselect · 0.4% in cash · turnover 11.0× · cost drag 1.1%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 2.5% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2009-01-01 → 2009-02-01 | -12.9214% | 1.5957% | 19.2315% | -2.9988% | yes | 3.0584% | 1 / 19 |
| 2009-02-01 → 2009-03-01 | -9.8802% | 0.0947% | 11.8279% | -10.3321% | no | 2.4023% | 4 / 18 |
| 2009-03-01 → 2009-04-01 | -12.9271% | -0.8125% | 13.8884% | 12.901% | yes | 2.6499% | 1 / 21 |
| 2009-04-01 → 2009-05-01 | -13.4341% | -0.5915% | 14.4781% | 4.9835% | yes | 2.882% | 0 / 20 |
| 2009-05-01 → 2009-06-01 | -13.9825% | -0.1519% | 16.5628% | 1.2875% | yes | 3.1447% | 0 / 19 |
| 2009-06-01 → 2009-07-01 | -15.2682% | -0.2404% | 18.6275% | -3.4258% | yes | 3.5149% | 0 / 21 |
| 2009-07-01 → 2009-08-01 | -17.8812% | -0.9054% | 20.9573% | 11.082% | yes | 4.3861% | 0 / 21 |
| 2009-08-01 → 2009-09-01 | -18.4289% | -0.5679% | 21.6933% | 1.0168% | yes | 4.5974% | 0 / 20 |
| 2009-09-01 → 2009-10-01 | -18.3708% | -0.4281% | 21.9509% | 9.6439% | yes | 4.7453% | 0 / 20 |
| 2009-10-01 → 2009-11-01 | -19.172% | -0.0129% | 25.3018% | -1.8489% | yes | 5.0379% | 0 / 21 |
| 2009-11-01 → 2009-12-01 | -19.7129% | -0.371% | 24.6516% | 4.4325% | yes | 5.0853% | 0 / 19 |
| 2009-12-01 → 2010-01-01 | -18.4998% | 0.0061% | 24.2545% | 3.3798% | yes | 4.5853% | 0 / 21 |
| 2010-01-01 | no segment follows this rebalance, not scored | ||||||
Step 5 · 2010-01-04 → 2010-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 65 names held · selection: reselect · 25.8% in cash · turnover 6.4× · cost drag 0.64% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 9 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 1.68% of 179 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2010-01-01 → 2010-02-01 | -14.6073% | 0.5841% | 19.5672% | -6.4127% | yes | 3.8765% | 0 / 18 |
| 2010-02-01 → 2010-03-01 | -14.5095% | 0.6738% | 19.6418% | 5.0721% | yes | 3.7375% | 0 / 18 |
| 2010-03-01 → 2010-04-01 | -15.9585% | 1.2151% | 21.1536% | 4.5032% | yes | 3.9316% | 0 / 22 |
| 2010-04-01 → 2010-05-01 | -15.9178% | 1.6204% | 23.2912% | 3.1228% | yes | 4.3242% | 0 / 20 |
| 2010-05-01 → 2010-06-01 | -16.3412% | 1.4544% | 23.9429% | -10.2341% | yes | 4.5201% | 2 / 19 |
| 2010-06-01 → 2010-07-01 | -20.4805% | 1.2314% | 30.7819% | -11.3816% | yes | 5.1893% | 1 / 21 |
| 2010-07-01 | no segment follows this rebalance, not scored | ||||||
| 2010-08-01 | no segment follows this rebalance, not scored | ||||||
| 2010-09-01 | no segment follows this rebalance, not scored | ||||||
| 2010-10-01 → 2010-11-01 | -14.8453% | 1.1399% | 20.5472% | 3.2281% | yes | 3.9508% | 0 / 20 |
| 2010-11-01 → 2010-12-01 | -14.3333% | 1.4848% | 20.6389% | 0.6635% | yes | 3.9202% | 0 / 20 |
| 2010-12-01 → 2011-01-01 | -15.4409% | 2.8903% | 26.7019% | 4.0532% | yes | 4.4341% | 0 / 21 |
| 2011-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 87 names held · selection: reselect · 0.4% in cash · turnover 9.1× · cost drag 0.91%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 1.25% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2010-01-01 → 2010-02-01 | -14.6073% | 0.5841% | 19.5672% | -6.4127% | yes | 3.8765% | 0 / 18 |
| 2010-02-01 → 2010-03-01 | -14.5095% | 0.6738% | 19.6418% | 5.0721% | yes | 3.7375% | 0 / 18 |
| 2010-03-01 → 2010-04-01 | -15.9585% | 1.2151% | 21.1536% | 4.5032% | yes | 3.9316% | 0 / 22 |
| 2010-04-01 → 2010-05-01 | -15.9178% | 1.6204% | 23.2912% | 3.1228% | yes | 4.3242% | 0 / 20 |
| 2010-05-01 → 2010-06-01 | -16.3412% | 1.4544% | 23.9429% | -10.2341% | yes | 4.5201% | 2 / 19 |
| 2010-06-01 → 2010-07-01 | -20.549% | 0.8578% | 29.908% | -8.3929% | yes | 5.7515% | 1 / 21 |
| 2010-07-01 → 2010-08-01 | -11.8573% | 1.0616% | 16.197% | 3.6491% | yes | 2.9247% | 0 / 20 |
| 2010-08-01 → 2010-09-01 | -12.2019% | 1.1601% | 17.5666% | -2.474% | yes | 3.1066% | 0 / 21 |
| 2010-09-01 → 2010-10-01 | -12.2616% | 1.1222% | 16.8847% | 4.6816% | yes | 3.2612% | 0 / 20 |
| 2010-10-01 → 2010-11-01 | -13.2103% | 1.2841% | 18.5711% | 2.4532% | yes | 3.5127% | 0 / 20 |
| 2010-11-01 → 2010-12-01 | -14.7298% | 1.5476% | 21.3629% | 2.1524% | yes | 3.9541% | 0 / 20 |
| 2010-12-01 → 2011-01-01 | -14.8673% | 3.0917% | 26.3042% | 3.7776% | yes | 4.1894% | 0 / 21 |
| 2011-01-01 | no segment follows this rebalance, not scored | ||||||
Step 6 · 2011-01-03 → 2011-12-30
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 66 names held · selection: reselect · 0.4% in cash · turnover 7.2× · cost drag 0.72% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.92% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2011-01-01 → 2011-02-01 | -13.643% | 3.7879% | 25.6141% | 2.1331% | yes | 4.2575% | 0 / 19 |
| 2011-02-01 → 2011-03-01 | -11.7096% | 3.9307% | 23.4619% | 3.5329% | yes | 4.3561% | 0 / 18 |
| 2011-03-01 → 2011-04-01 | -14.0495% | 4.2273% | 25.5903% | 3.6169% | yes | 4.255% | 0 / 22 |
| 2011-04-01 → 2011-05-01 | -11.7811% | 4.3096% | 24.1249% | -0.7722% | yes | 3.7229% | 0 / 19 |
| 2011-05-01 → 2011-06-01 | -9.5985% | 5.1145% | 22.5976% | -1.6246% | yes | 3.2839% | 0 / 20 |
| 2011-06-01 → 2011-07-01 | -5.3244% | 3.4388% | 13.6246% | -1.0273% | yes | 1.856% | 2 / 21 |
| 2011-07-01 → 2011-08-01 | -5.814% | 2.581% | 12.0884% | -4.0946% | yes | 1.9075% | 2 / 19 |
| 2011-08-01 → 2011-09-01 | -5.274% | 3.1366% | 11.9798% | -4.1196% | yes | 1.7703% | 5 / 22 |
| 2011-09-01 → 2011-10-01 | -6.0187% | 2.8355% | 12.7303% | -5.9026% | yes | 1.8219% | 5 / 20 |
| 2011-10-01 → 2011-11-01 | -5.7565% | 2.0255% | 10.6286% | 4.5688% | yes | 1.7121% | 3 / 20 |
| 2011-11-01 → 2011-12-01 | -5.9405% | 1.4133% | 9.5097% | 3.7551% | yes | 1.6115% | 2 / 20 |
| 2011-12-01 → 2012-01-01 | -5.9712% | 2.2432% | 11.3648% | 2.8825% | yes | 1.777% | 0 / 20 |
| 2012-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 87 names held · selection: reselect · 0.4% in cash · turnover 8.6× · cost drag 0.86%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.5% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2011-01-01 → 2011-02-01 | -13.643% | 3.7879% | 25.6141% | 2.1331% | yes | 4.2575% | 0 / 19 |
| 2011-02-01 → 2011-03-01 | -11.7096% | 3.9307% | 23.4619% | 3.5329% | yes | 4.3561% | 0 / 18 |
| 2011-03-01 → 2011-04-01 | -14.0495% | 4.2273% | 25.5903% | 3.6169% | yes | 4.255% | 0 / 22 |
| 2011-04-01 → 2011-05-01 | -11.7811% | 4.3096% | 24.1249% | -0.7722% | yes | 3.7229% | 0 / 19 |
| 2011-05-01 → 2011-06-01 | -9.5985% | 5.1145% | 22.5976% | -1.6246% | yes | 3.2839% | 0 / 20 |
| 2011-06-01 → 2011-07-01 | -5.3244% | 3.4388% | 13.6246% | -1.0273% | yes | 1.856% | 2 / 21 |
| 2011-07-01 → 2011-08-01 | -5.814% | 2.581% | 12.0884% | -4.0946% | yes | 1.9075% | 2 / 19 |
| 2011-08-01 → 2011-09-01 | -5.274% | 3.1366% | 11.9798% | -4.1196% | yes | 1.7703% | 5 / 22 |
| 2011-09-01 → 2011-10-01 | -5.4374% | 2.7868% | 11.916% | -5.0448% | yes | 1.6548% | 5 / 20 |
| 2011-10-01 → 2011-11-01 | -4.2986% | 1.8887% | 8.6146% | 6.0939% | yes | 1.3322% | 1 / 20 |
| 2011-11-01 → 2011-12-01 | -4.2831% | 1.8294% | 8.4688% | 2.8611% | yes | 1.3265% | 3 / 20 |
| 2011-12-01 → 2012-01-01 | -4.4134% | 2.0337% | 9.0603% | 1.82% | yes | 1.4232% | 0 / 20 |
| 2012-01-01 | no segment follows this rebalance, not scored | ||||||
Step 7 · 2012-01-03 → 2012-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 12 constructions · 72 names held · selection: reselect · 8.4% in cash · turnover 7.4× · cost drag 0.74% · regime gate active · 1 name dropped at load (73 selected, 72 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 90.9% of 11 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 1.83% of 219 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2012-01-01 | no segment follows this rebalance, not scored | ||||||
| 2012-02-01 → 2012-03-01 | -10.5551% | 1.503% | 15.7431% | 3.6624% | yes | 3.0869% | 0 / 19 |
| 2012-03-01 → 2012-04-01 | -10.7085% | 1.656% | 16.6513% | 4.0382% | yes | 3.1116% | 0 / 21 |
| 2012-04-01 → 2012-05-01 | -10.163% | 1.6507% | 15.5607% | -2.0549% | yes | 3.0171% | 1 / 19 |
| 2012-05-01 → 2012-06-01 | -9.3939% | 2.2097% | 16.1493% | -11.5393% | no | 2.7424% | 2 / 21 |
| 2012-06-01 → 2012-07-01 | -4.8586% | 1.8287% | 9.1369% | 6.8483% | yes | 1.4946% | 1 / 20 |
| 2012-07-01 → 2012-08-01 | -4.8219% | 2.0102% | 9.4871% | 1.0203% | yes | 1.535% | 0 / 20 |
| 2012-08-01 → 2012-09-01 | -4.3397% | 2.3308% | 9.2237% | -0.0655% | yes | 1.3524% | 0 / 22 |
| 2012-09-01 → 2012-10-01 | -3.8689% | 2.0749% | 8.7503% | 1.5511% | yes | 1.4395% | 0 / 18 |
| 2012-10-01 → 2012-11-01 | -7.893% | 2.4652% | 14.2358% | -0.577% | yes | 2.3837% | 0 / 20 |
| 2012-11-01 → 2012-12-01 | -8.2573% | 1.997% | 13.6423% | 3.2513% | yes | 2.3883% | 0 / 20 |
| 2012-12-01 → window end | -8.6021% | 1.6977% | 13.6195% | 3.4773% | yes | 2.3651% | 0 / 19 |
B: without the market rule
Portfolio book, rebalanced monthly · 12 constructions · 74 names held · selection: reselect · 0.4% in cash · turnover 8.1× · cost drag 0.81% · 1 name dropped at load (75 selected, 74 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 1.68% of 238 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2012-01-01 → 2012-02-01 | -8.3098% | 1.9835% | 13.893% | 4.4022% | yes | 2.3415% | 0 / 19 |
| 2012-02-01 → 2012-03-01 | -8.679% | 2.021% | 14.4568% | 4.0395% | yes | 2.4471% | 0 / 19 |
| 2012-03-01 → 2012-04-01 | -9.552% | 2.0139% | 15.9059% | 5.7472% | yes | 2.6879% | 0 / 21 |
| 2012-04-01 → 2012-05-01 | -9.452% | 1.8468% | 15.0699% | -1.7847% | yes | 2.6911% | 1 / 19 |
| 2012-05-01 → 2012-06-01 | -8.9444% | 2.3091% | 15.7729% | -10.4701% | no | 2.6085% | 2 / 21 |
| 2012-06-01 → 2012-07-01 | -4.8586% | 1.8287% | 9.1369% | 6.8483% | yes | 1.4946% | 1 / 20 |
| 2012-07-01 → 2012-08-01 | -4.8219% | 2.0102% | 9.4871% | 1.0203% | yes | 1.535% | 0 / 20 |
| 2012-08-01 → 2012-09-01 | -4.3397% | 2.3308% | 9.2237% | -0.0655% | yes | 1.3524% | 0 / 22 |
| 2012-09-01 → 2012-10-01 | -3.8689% | 2.0749% | 8.7503% | 1.5511% | yes | 1.4395% | 0 / 18 |
| 2012-10-01 → 2012-11-01 | -7.893% | 2.4652% | 14.2358% | -0.577% | yes | 2.3837% | 0 / 20 |
| 2012-11-01 → 2012-12-01 | -8.2573% | 1.997% | 13.6423% | 3.2513% | yes | 2.3883% | 0 / 20 |
| 2012-12-01 → window end | -8.6021% | 1.6977% | 13.6195% | 3.4773% | yes | 2.3651% | 0 / 19 |
Step 8 · 2013-01-02 → 2013-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 78 names held · selection: reselect · 0.4% in cash · turnover 9.6× · cost drag 0.96% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 2.92% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2013-01-01 → 2013-02-01 | -8.5415% | 1.4584% | 12.7895% | 6.3253% | yes | 2.3346% | 0 / 20 |
| 2013-02-01 → 2013-03-01 | -8.4985% | 1.3326% | 12.8619% | -1.4611% | yes | 2.3855% | 2 / 18 |
| 2013-03-01 → 2013-04-01 | -9.9132% | 1.0375% | 13.8157% | 7.0382% | yes | 2.6722% | 0 / 19 |
| 2013-04-01 → 2013-05-01 | -9.7298% | 1.96% | 16.0217% | -2.9223% | yes | 2.735% | 1 / 21 |
| 2013-05-01 → 2013-06-01 | -6.6344% | 1.5913% | 11.1121% | 3.276% | yes | 1.96% | 0 / 21 |
| 2013-06-01 → 2013-07-01 | -6.0917% | 1.7293% | 10.5381% | -0.0931% | yes | 1.7999% | 2 / 19 |
| 2013-07-01 → 2013-08-01 | -7.158% | 1.9426% | 12.5751% | 6.3638% | yes | 2.1026% | 0 / 21 |
| 2013-08-01 → 2013-09-01 | -8.4269% | 2.0024% | 14.369% | -4.634% | yes | 2.3086% | 1 / 21 |
| 2013-09-01 → 2013-10-01 | -9.3256% | 1.2578% | 13.5531% | 3.9435% | yes | 2.4691% | 0 / 19 |
| 2013-10-01 → 2013-11-01 | -8.1456% | 2.3402% | 13.6177% | 2.7502% | yes | 2.3021% | 0 / 22 |
| 2013-11-01 → 2013-12-01 | -6.7446% | 2.2859% | 12.5866% | 1.5095% | yes | 2.1872% | 0 / 19 |
| 2013-12-01 → 2014-01-01 | -5.3767% | 2.5634% | 11.3522% | 3.5028% | yes | 1.8866% | 1 / 20 |
| 2014-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 78 names held · selection: reselect · 0.4% in cash · turnover 9.6× · cost drag 0.96%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 2.92% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2013-01-01 → 2013-02-01 | -8.5415% | 1.4584% | 12.7895% | 6.3253% | yes | 2.3346% | 0 / 20 |
| 2013-02-01 → 2013-03-01 | -8.4985% | 1.3326% | 12.8619% | -1.4611% | yes | 2.3855% | 2 / 18 |
| 2013-03-01 → 2013-04-01 | -9.9132% | 1.0375% | 13.8157% | 7.0382% | yes | 2.6722% | 0 / 19 |
| 2013-04-01 → 2013-05-01 | -9.7298% | 1.96% | 16.0217% | -2.9223% | yes | 2.735% | 1 / 21 |
| 2013-05-01 → 2013-06-01 | -6.6344% | 1.5913% | 11.1121% | 3.276% | yes | 1.96% | 0 / 21 |
| 2013-06-01 → 2013-07-01 | -6.0917% | 1.7293% | 10.5381% | -0.0931% | yes | 1.7999% | 2 / 19 |
| 2013-07-01 → 2013-08-01 | -7.158% | 1.9426% | 12.5751% | 6.3638% | yes | 2.1026% | 0 / 21 |
| 2013-08-01 → 2013-09-01 | -8.4269% | 2.0024% | 14.369% | -4.634% | yes | 2.3086% | 1 / 21 |
| 2013-09-01 → 2013-10-01 | -9.3256% | 1.2578% | 13.5531% | 3.9435% | yes | 2.4691% | 0 / 19 |
| 2013-10-01 → 2013-11-01 | -8.1456% | 2.3402% | 13.6177% | 2.7502% | yes | 2.3021% | 0 / 22 |
| 2013-11-01 → 2013-12-01 | -6.7446% | 2.2859% | 12.5866% | 1.5095% | yes | 2.1872% | 0 / 19 |
| 2013-12-01 → 2014-01-01 | -5.3767% | 2.5634% | 11.3522% | 3.5028% | yes | 1.8866% | 1 / 20 |
| 2014-01-01 | no segment follows this rebalance, not scored | ||||||
Step 9 · 2014-01-02 → 2014-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 81 names held · selection: reselect · 0.4% in cash · turnover 8.6× · cost drag 0.86% · regime gate active · 1 name dropped at load (82 selected, 81 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.5% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2014-01-01 → 2014-02-01 | -5.0048% | 2.8954% | 11.6339% | -0.2069% | yes | 1.7276% | 2 / 20 |
| 2014-02-01 → 2014-03-01 | -4.323% | 2.6437% | 10.5511% | 7.5102% | yes | 1.5823% | 0 / 18 |
| 2014-03-01 → 2014-04-01 | -3.4812% | 3.6319% | 11.4303% | 0.836% | yes | 1.3485% | 2 / 20 |
| 2014-04-01 → 2014-05-01 | -4.0076% | 3.4174% | 11.5858% | -1.0973% | yes | 1.5982% | 3 / 20 |
| 2014-05-01 → 2014-06-01 | -4.9778% | 1.7897% | 9.1919% | 2.8463% | yes | 1.5542% | 0 / 20 |
| 2014-06-01 → 2014-07-01 | -4.7342% | 1.8545% | 9.0472% | 4.1792% | yes | 1.5087% | 1 / 20 |
| 2014-07-01 → 2014-08-01 | -4.2324% | 2.2295% | 9.5627% | -5.247% | no | 1.4888% | 2 / 21 |
| 2014-08-01 → 2014-09-01 | -3.8458% | 2.685% | 9.8062% | 5.4475% | yes | 1.437% | 1 / 20 |
| 2014-09-01 → 2014-10-01 | -3.463% | 2.7567% | 9.5162% | -1.5045% | yes | 1.3459% | 1 / 20 |
| 2014-10-01 → 2014-11-01 | -3.6203% | 2.7772% | 9.3676% | 5.7344% | yes | 1.2819% | 4 / 22 |
| 2014-11-01 → 2014-12-01 | -3.3544% | 2.2894% | 8.6068% | 4.3301% | yes | 1.4062% | 0 / 18 |
| 2014-12-01 → 2015-01-01 | -3.1947% | 3.0246% | 10.0604% | 1.1982% | yes | 1.4709% | 2 / 21 |
| 2015-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 81 names held · selection: reselect · 0.4% in cash · turnover 8.6× · cost drag 0.86% · 1 name dropped at load (82 selected, 81 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.5% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2014-01-01 → 2014-02-01 | -5.0048% | 2.8954% | 11.6339% | -0.2069% | yes | 1.7276% | 2 / 20 |
| 2014-02-01 → 2014-03-01 | -4.323% | 2.6437% | 10.5511% | 7.5102% | yes | 1.5823% | 0 / 18 |
| 2014-03-01 → 2014-04-01 | -3.4812% | 3.6319% | 11.4303% | 0.836% | yes | 1.3485% | 2 / 20 |
| 2014-04-01 → 2014-05-01 | -4.0076% | 3.4174% | 11.5858% | -1.0973% | yes | 1.5982% | 3 / 20 |
| 2014-05-01 → 2014-06-01 | -4.9778% | 1.7897% | 9.1919% | 2.8463% | yes | 1.5542% | 0 / 20 |
| 2014-06-01 → 2014-07-01 | -4.7342% | 1.8545% | 9.0472% | 4.1792% | yes | 1.5087% | 1 / 20 |
| 2014-07-01 → 2014-08-01 | -4.2324% | 2.2295% | 9.5627% | -5.247% | no | 1.4888% | 2 / 21 |
| 2014-08-01 → 2014-09-01 | -3.8458% | 2.685% | 9.8062% | 5.4475% | yes | 1.437% | 1 / 20 |
| 2014-09-01 → 2014-10-01 | -3.463% | 2.7567% | 9.5162% | -1.5045% | yes | 1.3459% | 1 / 20 |
| 2014-10-01 → 2014-11-01 | -3.6203% | 2.7772% | 9.3676% | 5.7344% | yes | 1.2819% | 4 / 22 |
| 2014-11-01 → 2014-12-01 | -3.3544% | 2.2894% | 8.6068% | 4.3301% | yes | 1.4062% | 0 / 18 |
| 2014-12-01 → 2015-01-01 | -3.1947% | 3.0246% | 10.0604% | 1.1982% | yes | 1.4709% | 2 / 21 |
| 2015-01-01 | no segment follows this rebalance, not scored | ||||||
Step 10 · 2015-01-02 → 2015-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 72 names held · selection: reselect · 0.4% in cash · turnover 7.2× · cost drag 0.72% · regime gate active · 1 name dropped at load (73 selected, 72 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 83.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 6.67% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2015-01-01 → 2015-02-01 | -3.1005% | 2.5105% | 8.6793% | 0.0721% | yes | 1.2104% | 2 / 19 |
| 2015-02-01 → 2015-03-01 | -3.1442% | 2.4895% | 8.794% | 5.7808% | yes | 1.3011% | 0 / 18 |
| 2015-03-01 → 2015-04-01 | -2.6259% | 3.4074% | 10.2176% | 0.499% | yes | 1.2654% | 2 / 21 |
| 2015-04-01 → 2015-05-01 | -2.8451% | 3.479% | 10.3562% | -4.4872% | no | 1.399% | 1 / 20 |
| 2015-05-01 → 2015-06-01 | -2.8913% | 3.076% | 9.6588% | -0.4715% | yes | 1.3951% | 0 / 19 |
| 2015-06-01 → 2015-07-01 | -3.5877% | 2.9934% | 10.4674% | -0.0237% | yes | 1.486% | 1 / 21 |
| 2015-07-01 → 2015-08-01 | -3.431% | 3.1069% | 10.5278% | -0.5825% | yes | 1.4729% | 1 / 21 |
| 2015-08-01 → 2015-09-01 | -3.538% | 2.3515% | 8.7315% | -4.7782% | no | 1.3563% | 3 / 20 |
| 2015-09-01 → 2015-10-01 | -5.0449% | 2.1685% | 10.0937% | 7.1031% | yes | 1.6984% | 1 / 20 |
| 2015-10-01 → 2015-11-01 | -4.9605% | 2.3791% | 10.7867% | 10.2026% | yes | 1.7332% | 1 / 21 |
| 2015-11-01 → 2015-12-01 | -3.4782% | 2.4854% | 9.0662% | 0.4108% | yes | 1.348% | 2 / 19 |
| 2015-12-01 → 2016-01-01 | -3.4913% | 2.5207% | 9.3091% | -1.4657% | yes | 1.3452% | 2 / 21 |
| 2016-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 78 names held · selection: reselect · 0.4% in cash · turnover 7.9× · cost drag 0.79% · 1 name dropped at load (79 selected, 78 held across the window), weights renormalised onto the rest
Projection accuracy, realized outcome fell inside the P5–P95 cone in 83.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.08% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2015-01-01 → 2015-02-01 | -3.1005% | 2.5105% | 8.6793% | 0.0721% | yes | 1.2104% | 2 / 19 |
| 2015-02-01 → 2015-03-01 | -3.1442% | 2.4895% | 8.794% | 5.7808% | yes | 1.3011% | 0 / 18 |
| 2015-03-01 → 2015-04-01 | -2.6259% | 3.4074% | 10.2176% | 0.499% | yes | 1.2654% | 2 / 21 |
| 2015-04-01 → 2015-05-01 | -2.8451% | 3.479% | 10.3562% | -4.4872% | no | 1.399% | 1 / 20 |
| 2015-05-01 → 2015-06-01 | -2.8913% | 3.076% | 9.6588% | -0.4715% | yes | 1.3951% | 0 / 19 |
| 2015-06-01 → 2015-07-01 | -3.5877% | 2.9934% | 10.4674% | -0.0237% | yes | 1.486% | 1 / 21 |
| 2015-07-01 → 2015-08-01 | -3.431% | 3.1069% | 10.5278% | -0.5825% | yes | 1.4729% | 1 / 21 |
| 2015-08-01 → 2015-09-01 | -3.538% | 2.3515% | 8.7315% | -4.7782% | no | 1.3563% | 3 / 20 |
| 2015-09-01 → 2015-10-01 | -4.0169% | 2.2361% | 9.0364% | 3.455% | yes | 1.4184% | 1 / 20 |
| 2015-10-01 → 2015-11-01 | -4.1199% | 2.0964% | 9.1328% | 6.9389% | yes | 1.4547% | 1 / 21 |
| 2015-11-01 → 2015-12-01 | -3.9168% | 2.3687% | 9.329% | -0.1932% | yes | 1.4759% | 2 / 19 |
| 2015-12-01 → 2016-01-01 | -3.8602% | 2.4284% | 9.5507% | -2.5236% | yes | 1.4659% | 3 / 21 |
| 2016-01-01 | no segment follows this rebalance, not scored | ||||||
Step 11 · 2016-01-04 → 2016-12-30
A: the book's rules
Portfolio book, rebalanced monthly · 12 constructions · 71 names held · selection: reselect · 24.6% in cash · turnover 7.0× · cost drag 0.7% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 88.9% of 9 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 6.04% of 182 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2016-01-01 | no segment follows this rebalance, not scored | ||||||
| 2016-02-01 | no segment follows this rebalance, not scored | ||||||
| 2016-03-01 | no segment follows this rebalance, not scored | ||||||
| 2016-04-01 → 2016-05-01 | -4.6017% | 1.2797% | 7.6547% | -2.0883% | yes | 1.2796% | 3 / 20 |
| 2016-05-01 → 2016-06-01 | -5.0112% | 0.9638% | 7.448% | -1.0207% | yes | 1.3131% | 2 / 20 |
| 2016-06-01 → 2016-07-01 | -10.2074% | -0.0289% | 12.0344% | 3.2354% | yes | 2.3055% | 3 / 21 |
| 2016-07-01 → 2016-08-01 | -10.2925% | 0.0731% | 12.1024% | 0.4075% | yes | 2.5209% | 0 / 19 |
| 2016-08-01 → 2016-09-01 | -6.583% | 1.6238% | 10.2454% | 0.5221% | yes | 1.799% | 0 / 22 |
| 2016-09-01 → 2016-10-01 | -6.8798% | 1.2643% | 10.3087% | 1.5693% | yes | 1.9882% | 2 / 20 |
| 2016-10-01 → 2016-11-01 | -6.7712% | 1.3429% | 10.3505% | -4.1189% | yes | 1.9044% | 1 / 20 |
| 2016-11-01 → 2016-12-01 | -8.0628% | 0.9209% | 10.9927% | 12.9009% | no | 2.0472% | 0 / 20 |
| 2016-12-01 → window end | -8.4939% | 0.8511% | 11.3706% | 1.7328% | yes | 2.1404% | 0 / 20 |
B: without the market rule
Portfolio book, rebalanced monthly · 12 constructions · 89 names held · selection: reselect · 0.4% in cash · turnover 9.1× · cost drag 0.91%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.08% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2016-01-01 → 2016-02-01 | -4.3853% | 1.5715% | 8.2644% | -3.2087% | yes | 1.456% | 4 / 18 |
| 2016-02-01 → 2016-03-01 | -4.5544% | 1.3959% | 7.9657% | -0.6652% | yes | 1.4007% | 3 / 19 |
| 2016-03-01 → 2016-04-01 | -4.0065% | 1.3498% | 7.3598% | 4.517% | yes | 1.1961% | 0 / 21 |
| 2016-04-01 → 2016-05-01 | -3.8253% | 1.4074% | 7.0402% | -2.9753% | yes | 1.0053% | 3 / 20 |
| 2016-05-01 → 2016-06-01 | -4.7224% | 0.9261% | 7.0349% | -1.2848% | yes | 1.1788% | 1 / 20 |
| 2016-06-01 → 2016-07-01 | -9.4588% | -0.1773% | 10.7123% | 2.3293% | yes | 2.2725% | 3 / 21 |
| 2016-07-01 → 2016-08-01 | -9.2696% | 0.4761% | 11.7022% | 1.14% | yes | 2.7105% | 0 / 19 |
| 2016-08-01 → 2016-09-01 | -5.8694% | 2.0248% | 10.288% | 0.0357% | yes | 1.6214% | 0 / 22 |
| 2016-09-01 → 2016-10-01 | -6.3549% | 1.3968% | 9.968% | 1.9323% | yes | 1.8126% | 2 / 20 |
| 2016-10-01 → 2016-11-01 | -6.3195% | 1.4422% | 10.025% | -4.3021% | yes | 1.8204% | 1 / 20 |
| 2016-11-01 → 2016-12-01 | -8.1475% | 0.9621% | 11.1889% | 12.6488% | no | 2.0128% | 0 / 20 |
| 2016-12-01 → window end | -8.4939% | 0.8511% | 11.3706% | 1.7328% | yes | 2.1404% | 0 / 20 |
Step 12 · 2017-01-03 → 2017-12-29
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 79 names held · selection: reselect · 0.4% in cash · turnover 8.1× · cost drag 0.81% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 3.35% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2017-01-01 → 2017-02-01 | -8.7503% | 1.0504% | 12.3395% | -0.6511% | yes | 2.396% | 1 / 19 |
| 2017-02-01 → 2017-03-01 | -8.5927% | 0.9424% | 12.092% | 4.0702% | yes | 2.4192% | 0 / 18 |
| 2017-03-01 → 2017-04-01 | -9.4289% | 1.2304% | 12.7301% | -6.2503% | yes | 2.4383% | 1 / 22 |
| 2017-04-01 → 2017-05-01 | -6.266% | 1.3498% | 10.064% | 0.4562% | yes | 1.9762% | 0 / 18 |
| 2017-05-01 → 2017-06-01 | -6.668% | 1.2542% | 10.3957% | 2.798% | yes | 1.9433% | 1 / 21 |
| 2017-06-01 → 2017-07-01 | -5.4357% | 2.046% | 10.6318% | -4.1538% | yes | 1.6325% | 2 / 21 |
| 2017-07-01 → 2017-08-01 | -5.5772% | 2.1372% | 10.8131% | 4.5349% | yes | 1.5648% | 0 / 19 |
| 2017-08-01 → 2017-09-01 | -5.4463% | 2.4471% | 10.7065% | 3.2512% | yes | 1.5863% | 2 / 22 |
| 2017-09-01 → 2017-10-01 | -4.656% | 1.7157% | 8.7809% | 2.1421% | yes | 1.3834% | 0 / 19 |
| 2017-10-01 → 2017-11-01 | -4.8788% | 1.6321% | 9.0279% | 3.0263% | yes | 1.4501% | 0 / 21 |
| 2017-11-01 → 2017-12-01 | -4.446% | 1.9106% | 8.8326% | 4.1435% | yes | 1.4215% | 0 / 20 |
| 2017-12-01 → 2018-01-01 | -3.8867% | 2.3234% | 9.195% | 0.5092% | yes | 1.4046% | 1 / 19 |
| 2018-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 79 names held · selection: reselect · 0.4% in cash · turnover 8.1× · cost drag 0.81%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 3.35% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2017-01-01 → 2017-02-01 | -8.7503% | 1.0504% | 12.3395% | -0.6511% | yes | 2.396% | 1 / 19 |
| 2017-02-01 → 2017-03-01 | -8.5927% | 0.9424% | 12.092% | 4.0702% | yes | 2.4192% | 0 / 18 |
| 2017-03-01 → 2017-04-01 | -9.4289% | 1.2304% | 12.7301% | -6.2503% | yes | 2.4383% | 1 / 22 |
| 2017-04-01 → 2017-05-01 | -6.266% | 1.3498% | 10.064% | 0.4562% | yes | 1.9762% | 0 / 18 |
| 2017-05-01 → 2017-06-01 | -6.668% | 1.2542% | 10.3957% | 2.798% | yes | 1.9433% | 1 / 21 |
| 2017-06-01 → 2017-07-01 | -5.4357% | 2.046% | 10.6318% | -4.1538% | yes | 1.6325% | 2 / 21 |
| 2017-07-01 → 2017-08-01 | -5.5772% | 2.1372% | 10.8131% | 4.5349% | yes | 1.5648% | 0 / 19 |
| 2017-08-01 → 2017-09-01 | -5.4463% | 2.4471% | 10.7065% | 3.2512% | yes | 1.5863% | 2 / 22 |
| 2017-09-01 → 2017-10-01 | -4.656% | 1.7157% | 8.7809% | 2.1421% | yes | 1.3834% | 0 / 19 |
| 2017-10-01 → 2017-11-01 | -4.8788% | 1.6321% | 9.0279% | 3.0263% | yes | 1.4501% | 0 / 21 |
| 2017-11-01 → 2017-12-01 | -4.446% | 1.9106% | 8.8326% | 4.1435% | yes | 1.4215% | 0 / 20 |
| 2017-12-01 → 2018-01-01 | -3.8867% | 2.3234% | 9.195% | 0.5092% | yes | 1.4046% | 1 / 19 |
| 2018-01-01 | no segment follows this rebalance, not scored | ||||||
Step 13 · 2018-01-02 → 2018-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 71 names held · selection: reselect · 0.4% in cash · turnover 7.6× · cost drag 0.76% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 75.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 11.72% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2018-01-01 → 2018-02-01 | -5.0519% | 1.5332% | 8.7234% | 4.1742% | yes | 1.4923% | 0 / 20 |
| 2018-02-01 → 2018-03-01 | -5.021% | 1.4478% | 8.7569% | -6.3611% | no | 1.5326% | 4 / 18 |
| 2018-03-01 → 2018-04-01 | -5.0709% | 1.8494% | 9.4308% | -1.6705% | yes | 1.5478% | 4 / 20 |
| 2018-04-01 → 2018-05-01 | -4.3271% | 2.4008% | 9.7536% | 2.3262% | yes | 1.3681% | 2 / 20 |
| 2018-05-01 → 2018-06-01 | -3.7063% | 2.4096% | 9.3234% | 2.9413% | yes | 1.2361% | 0 / 21 |
| 2018-06-01 → 2018-07-01 | -3.7576% | 2.7367% | 9.8151% | -0.9886% | yes | 1.2799% | 2 / 20 |
| 2018-07-01 → 2018-08-01 | -4.4322% | 2.727% | 10.5851% | 2.4664% | yes | 1.5025% | 0 / 20 |
| 2018-08-01 → 2018-09-01 | -4.4307% | 2.7072% | 10.1166% | 6.6433% | yes | 1.4035% | 1 / 22 |
| 2018-09-01 → 2018-10-01 | -2.9332% | 2.1735% | 7.8574% | -0.0849% | yes | 1.0823% | 0 / 18 |
| 2018-10-01 → 2018-11-01 | -3.3553% | 2.3087% | 8.1017% | -8.4509% | no | 1.0832% | 7 / 22 |
| 2018-11-01 → 2018-12-01 | -4.7062% | 1.3562% | 7.9398% | 5.9762% | yes | 1.271% | 2 / 20 |
| 2018-12-01 → 2019-01-01 | -4.4737% | 1.553% | 8.3297% | -7.7487% | no | 1.357% | 6 / 18 |
| 2019-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 84 names held · selection: reselect · 0.4% in cash · turnover 8.5× · cost drag 0.85%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 75.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 12.13% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2018-01-01 → 2018-02-01 | -5.0519% | 1.5332% | 8.7234% | 4.1742% | yes | 1.4923% | 0 / 20 |
| 2018-02-01 → 2018-03-01 | -5.021% | 1.4478% | 8.7569% | -6.3611% | no | 1.5326% | 4 / 18 |
| 2018-03-01 → 2018-04-01 | -5.0709% | 1.8494% | 9.4308% | -1.6705% | yes | 1.5478% | 4 / 20 |
| 2018-04-01 → 2018-05-01 | -4.3271% | 2.4008% | 9.7536% | 2.3262% | yes | 1.3681% | 2 / 20 |
| 2018-05-01 → 2018-06-01 | -3.7063% | 2.4096% | 9.3234% | 2.9413% | yes | 1.2361% | 0 / 21 |
| 2018-06-01 → 2018-07-01 | -3.7576% | 2.7367% | 9.8151% | -0.9886% | yes | 1.2799% | 2 / 20 |
| 2018-07-01 → 2018-08-01 | -4.4322% | 2.727% | 10.5851% | 2.4664% | yes | 1.5025% | 0 / 20 |
| 2018-08-01 → 2018-09-01 | -4.4307% | 2.7072% | 10.1166% | 6.6433% | yes | 1.4035% | 1 / 22 |
| 2018-09-01 → 2018-10-01 | -2.9332% | 2.1735% | 7.8574% | -0.0849% | yes | 1.0823% | 0 / 18 |
| 2018-10-01 → 2018-11-01 | -3.3553% | 2.3087% | 8.1017% | -8.4509% | no | 1.0832% | 7 / 22 |
| 2018-11-01 → 2018-12-01 | -3.5138% | 1.5653% | 7.0236% | 4.0336% | yes | 1.0568% | 3 / 20 |
| 2018-12-01 → 2019-01-01 | -3.0056% | 1.5032% | 6.4917% | -7.9696% | no | 1.0245% | 6 / 18 |
| 2019-01-01 | no segment follows this rebalance, not scored | ||||||
Step 14 · 2019-01-02 → 2019-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 83 names held · selection: reselect · 16.3% in cash · turnover 7.7× · cost drag 0.77% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 90.0% of 10 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 6.44% of 202 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2019-01-01 | no segment follows this rebalance, not scored | ||||||
| 2019-02-01 | no segment follows this rebalance, not scored | ||||||
| 2019-03-01 → 2019-04-01 | -4.9698% | 1.6782% | 8.9411% | 1.295% | yes | 1.5818% | 1 / 20 |
| 2019-04-01 → 2019-05-01 | -5.0113% | 1.7432% | 9.1304% | 1.5445% | yes | 1.5824% | 0 / 20 |
| 2019-05-01 → 2019-06-01 | -5.1025% | 2.0991% | 10.3382% | -9.0227% | no | 1.742% | 5 / 21 |
| 2019-06-01 → 2019-07-01 | -4.4241% | 1.5358% | 8.1162% | 6.6502% | yes | 1.4243% | 0 / 19 |
| 2019-07-01 → 2019-08-01 | -3.8227% | 2.0246% | 8.6174% | 0.4047% | yes | 1.3414% | 0 / 21 |
| 2019-08-01 → 2019-09-01 | -4.2254% | 1.8558% | 8.7304% | 2.9049% | yes | 1.4401% | 3 / 21 |
| 2019-09-01 → 2019-10-01 | -3.6439% | 1.8761% | 7.9412% | -3.2894% | yes | 1.19% | 3 / 19 |
| 2019-10-01 → 2019-11-01 | -5.849% | 1.3659% | 8.8685% | 0.5795% | yes | 1.6731% | 1 / 22 |
| 2019-11-01 → 2019-12-01 | -5.2681% | 1.2375% | 8.4641% | 0.8825% | yes | 1.4884% | 0 / 19 |
| 2019-12-01 → 2020-01-01 | -6.078% | 1.1588% | 9.1179% | 2.2738% | yes | 1.7711% | 0 / 20 |
| 2020-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 91 names held · selection: reselect · 0.4% in cash · turnover 9.7× · cost drag 0.97%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 5.0% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2019-01-01 → 2019-02-01 | -3.1092% | 1.3032% | 6.0125% | 4.349% | yes | 0.9456% | 2 / 20 |
| 2019-02-01 → 2019-03-01 | -2.9579% | 1.1795% | 5.7398% | 2.5494% | yes | 0.9573% | 0 / 18 |
| 2019-03-01 → 2019-04-01 | -3.5958% | 1.4738% | 6.9214% | 1.8942% | yes | 1.1196% | 2 / 20 |
| 2019-04-01 → 2019-05-01 | -4.3639% | 1.7722% | 8.4393% | 1.94% | yes | 1.5167% | 0 / 20 |
| 2019-05-01 → 2019-06-01 | -4.396% | 2.0732% | 9.416% | -5.8821% | no | 1.6259% | 2 / 21 |
| 2019-06-01 → 2019-07-01 | -3.9014% | 1.7033% | 7.8675% | 6.0169% | yes | 1.334% | 0 / 19 |
| 2019-07-01 → 2019-08-01 | -3.4319% | 2.2172% | 8.5724% | 0.5962% | yes | 1.3117% | 0 / 21 |
| 2019-08-01 → 2019-09-01 | -4.0638% | 1.9445% | 8.7309% | 3.1628% | yes | 1.3964% | 3 / 21 |
| 2019-09-01 → 2019-10-01 | -3.5745% | 1.9452% | 8.0099% | -3.5698% | yes | 1.1661% | 3 / 19 |
| 2019-10-01 → 2019-11-01 | -5.7105% | 1.4458% | 8.8824% | 1.2968% | yes | 1.6928% | 0 / 22 |
| 2019-11-01 → 2019-12-01 | -5.1253% | 1.3118% | 8.4567% | 1.0246% | yes | 1.5129% | 0 / 19 |
| 2019-12-01 → 2020-01-01 | -6.078% | 1.1588% | 9.1179% | 2.2738% | yes | 1.7711% | 0 / 20 |
| 2020-01-01 | no segment follows this rebalance, not scored | ||||||
Step 15 · 2020-01-02 → 2020-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 12 constructions · 75 names held · selection: reselect · 0.4% in cash · turnover 7.0× · cost drag 0.7% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 83.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 9.96% of 241 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2020-01-01 → 2020-02-01 | -7.2773% | 1.04% | 10.296% | -5.1454% | yes | 2.0456% | 2 / 20 |
| 2020-02-01 → 2020-03-01 | -6.5688% | 1.4212% | 10.6005% | -6.0089% | yes | 2.1605% | 3 / 18 |
| 2020-03-01 → 2020-04-01 | -6.3438% | 2.0395% | 11.7558% | -10.8773% | no | 2.0849% | 9 / 21 |
| 2020-04-01 → 2020-05-01 | -8.5441% | 1.5564% | 13.0131% | 15.1845% | no | 1.9918% | 2 / 20 |
| 2020-05-01 → 2020-06-01 | -8.1411% | 1.907% | 13.5024% | 5.1914% | yes | 1.9812% | 1 / 19 |
| 2020-06-01 → 2020-07-01 | -7.7492% | 2.5531% | 14.7439% | 4.8181% | yes | 1.8783% | 3 / 21 |
| 2020-07-01 → 2020-08-01 | -8.0898% | 2.4719% | 15.0073% | 7.4974% | yes | 2.421% | 0 / 21 |
| 2020-08-01 → 2020-09-01 | -9.542% | 2.1517% | 15.6429% | 5.0731% | yes | 2.5407% | 0 / 20 |
| 2020-09-01 → 2020-10-01 | -9.0813% | 2.6563% | 16.1961% | -3.779% | yes | 2.6257% | 2 / 20 |
| 2020-10-01 → 2020-11-01 | -10.3448% | 2.4171% | 17.9486% | 0.1624% | yes | 2.9204% | 1 / 21 |
| 2020-11-01 → 2020-12-01 | -10.9673% | 1.7246% | 16.8174% | 8.3448% | yes | 2.9396% | 1 / 19 |
| 2020-12-01 → window end | -12.2272% | 1.9338% | 19.4638% | 3.9061% | yes | 3.1901% | 0 / 21 |
B: without the market rule
Portfolio book, rebalanced monthly · 12 constructions · 86 names held · selection: reselect · 0.4% in cash · turnover 8.8× · cost drag 0.88%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 83.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 9.54% of 241 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2020-01-01 → 2020-02-01 | -7.2773% | 1.04% | 10.296% | -5.1454% | yes | 2.0456% | 2 / 20 |
| 2020-02-01 → 2020-03-01 | -6.5688% | 1.4212% | 10.6005% | -6.0089% | yes | 2.1605% | 3 / 18 |
| 2020-03-01 → 2020-04-01 | -6.3438% | 2.0395% | 11.7558% | -10.8773% | no | 2.0849% | 9 / 21 |
| 2020-04-01 → 2020-05-01 | -7.1476% | 1.1583% | 10.3996% | 11.5323% | no | 1.5637% | 2 / 20 |
| 2020-05-01 → 2020-06-01 | -6.653% | 1.4273% | 10.5563% | 5.981% | yes | 1.5805% | 2 / 19 |
| 2020-06-01 → 2020-07-01 | -6.2121% | 2.0039% | 11.5089% | 1.933% | yes | 1.6361% | 1 / 21 |
| 2020-07-01 → 2020-08-01 | -6.4486% | 2.5342% | 13.0094% | 6.6034% | yes | 1.6338% | 1 / 21 |
| 2020-08-01 → 2020-09-01 | -10.1442% | 2.0826% | 16.2746% | 4.6946% | yes | 2.5181% | 0 / 20 |
| 2020-09-01 → 2020-10-01 | -9.8238% | 2.4201% | 16.6281% | -2.091% | yes | 2.5684% | 2 / 20 |
| 2020-10-01 → 2020-11-01 | -10.8434% | 2.1389% | 17.9852% | -0.5059% | yes | 2.842% | 1 / 21 |
| 2020-11-01 → 2020-12-01 | -11.4695% | 1.473% | 16.9139% | 6.731% | yes | 2.888% | 0 / 19 |
| 2020-12-01 → window end | -13.8089% | 1.4542% | 20.613% | 3.7749% | yes | 3.3154% | 0 / 21 |
Step 16 · 2021-01-04 → 2021-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 78 names held · selection: reselect · 0.4% in cash · turnover 10.0× · cost drag 1.0% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 2.92% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2021-01-01 → 2021-02-01 | -13.0156% | 1.9413% | 20.5343% | 3.1339% | yes | 3.4341% | 1 / 18 |
| 2021-02-01 → 2021-03-01 | -12.5959% | 2.5161% | 21.3173% | 11.9321% | yes | 3.4937% | 1 / 18 |
| 2021-03-01 → 2021-04-01 | -12.7623% | 4.3239% | 24.0236% | -2.472% | yes | 3.5318% | 0 / 22 |
| 2021-04-01 → 2021-05-01 | -13.9601% | 2.7776% | 23.2162% | 2.1733% | yes | 3.6422% | 0 / 20 |
| 2021-05-01 → 2021-06-01 | -14.2467% | 2.2376% | 22.71% | 3.4215% | yes | 3.7192% | 0 / 19 |
| 2021-06-01 → 2021-07-01 | -13.4473% | 2.5712% | 22.8189% | -4.5025% | yes | 3.5504% | 0 / 21 |
| 2021-07-01 → 2021-08-01 | -10.4193% | 2.6097% | 17.859% | 2.578% | yes | 2.6071% | 0 / 20 |
| 2021-08-01 → 2021-09-01 | -7.9026% | 4.1399% | 18.6424% | 6.454% | yes | 2.4985% | 0 / 21 |
| 2021-09-01 → 2021-10-01 | -7.66% | 4.3452% | 18.2048% | -6.2763% | yes | 2.4102% | 1 / 20 |
| 2021-10-01 → 2021-11-01 | -8.6134% | 4.4096% | 19.6122% | 8.9266% | yes | 2.6696% | 0 / 20 |
| 2021-11-01 → 2021-12-01 | -8.8518% | 4.8783% | 21.0218% | -1.7416% | yes | 2.7668% | 1 / 20 |
| 2021-12-01 → 2022-01-01 | -9.6356% | 4.8272% | 22.7104% | 0.7559% | yes | 2.6625% | 3 / 21 |
| 2022-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 78 names held · selection: reselect · 0.4% in cash · turnover 10.0× · cost drag 1.0%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 2.92% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2021-01-01 → 2021-02-01 | -13.0156% | 1.9413% | 20.5343% | 3.1339% | yes | 3.4341% | 1 / 18 |
| 2021-02-01 → 2021-03-01 | -12.5959% | 2.5161% | 21.3173% | 11.9321% | yes | 3.4937% | 1 / 18 |
| 2021-03-01 → 2021-04-01 | -12.7623% | 4.3239% | 24.0236% | -2.472% | yes | 3.5318% | 0 / 22 |
| 2021-04-01 → 2021-05-01 | -13.9601% | 2.7776% | 23.2162% | 2.1733% | yes | 3.6422% | 0 / 20 |
| 2021-05-01 → 2021-06-01 | -14.2467% | 2.2376% | 22.71% | 3.4215% | yes | 3.7192% | 0 / 19 |
| 2021-06-01 → 2021-07-01 | -13.4473% | 2.5712% | 22.8189% | -4.5025% | yes | 3.5504% | 0 / 21 |
| 2021-07-01 → 2021-08-01 | -10.4193% | 2.6097% | 17.859% | 2.578% | yes | 2.6071% | 0 / 20 |
| 2021-08-01 → 2021-09-01 | -7.9026% | 4.1399% | 18.6424% | 6.454% | yes | 2.4985% | 0 / 21 |
| 2021-09-01 → 2021-10-01 | -7.66% | 4.3452% | 18.2048% | -6.2763% | yes | 2.4102% | 1 / 20 |
| 2021-10-01 → 2021-11-01 | -8.6134% | 4.4096% | 19.6122% | 8.9266% | yes | 2.6696% | 0 / 20 |
| 2021-11-01 → 2021-12-01 | -8.8518% | 4.8783% | 21.0218% | -1.7416% | yes | 2.7668% | 1 / 20 |
| 2021-12-01 → 2022-01-01 | -9.6356% | 4.8272% | 22.7104% | 0.7559% | yes | 2.6625% | 3 / 21 |
| 2022-01-01 | no segment follows this rebalance, not scored | ||||||
Step 17 · 2022-01-03 → 2022-12-30
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 35 names held · selection: reselect · 0.4% in cash · turnover 3.4× · cost drag 0.34% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 6.28% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2022-01-01 → 2022-02-01 | -9.9361% | 3.2808% | 19.0569% | -7.9897% | yes | 2.5326% | 1 / 19 |
| 2022-02-01 → 2022-03-01 | -10.4791% | 2.6392% | 18.5827% | -2.2341% | yes | 2.4504% | 0 / 18 |
| 2022-03-01 → 2022-04-01 | -19.2672% | 3.2289% | 30.929% | 7.6496% | yes | 4.1054% | 0 / 22 |
| 2022-04-01 → 2022-05-01 | -15.1193% | 2.6829% | 25.1232% | -4.0433% | yes | 3.549% | 1 / 19 |
| 2022-05-01 → 2022-06-01 | -15.1353% | 3.2794% | 26.193% | 11.4221% | yes | 3.5949% | 2 / 20 |
| 2022-06-01 → 2022-07-01 | -10.8847% | 5.4353% | 25.1717% | -17.4435% | no | 3.1856% | 5 / 20 |
| 2022-07-01 → 2022-08-01 | -9.4295% | 2.8189% | 17.2892% | 5.6535% | yes | 2.7641% | 1 / 19 |
| 2022-08-01 → 2022-09-01 | -8.0127% | 3.2798% | 15.5172% | 4.918% | yes | 2.4772% | 1 / 22 |
| 2022-09-01 → 2022-10-01 | -7.3831% | 3.3897% | 15.6743% | -6.5855% | yes | 2.278% | 1 / 20 |
| 2022-10-01 → 2022-11-01 | -7.5709% | 2.9247% | 14.8634% | 12.2945% | yes | 2.2352% | 0 / 20 |
| 2022-11-01 → 2022-12-01 | -7.0426% | 3.4554% | 15.3899% | -0.0942% | yes | 2.2352% | 2 / 20 |
| 2022-12-01 → 2023-01-01 | -6.9672% | 3.6665% | 15.7703% | -1.2114% | yes | 2.2352% | 1 / 20 |
| 2023-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 83 names held · selection: reselect · 0.4% in cash · turnover 7.6× · cost drag 0.76%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 83.3% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 7.11% of 239 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2022-01-01 → 2022-02-01 | -9.9361% | 3.2808% | 19.0569% | -7.9897% | yes | 2.5326% | 1 / 19 |
| 2022-02-01 → 2022-03-01 | -8.4148% | 2.2146% | 14.78% | -1.4917% | yes | 2.2062% | 0 / 18 |
| 2022-03-01 → 2022-04-01 | -11.8996% | 2.2051% | 17.9894% | 7.5746% | yes | 2.653% | 0 / 22 |
| 2022-04-01 → 2022-05-01 | -11.7718% | 2.3917% | 19.505% | -3.3012% | yes | 2.8655% | 1 / 19 |
| 2022-05-01 → 2022-06-01 | -12.3513% | 3.1532% | 21.8034% | 8.3172% | yes | 2.896% | 2 / 20 |
| 2022-06-01 → 2022-07-01 | -9.3642% | 5.0546% | 22.1335% | -15.242% | no | 2.8328% | 5 / 20 |
| 2022-07-01 → 2022-08-01 | -5.397% | 1.5987% | 9.4088% | 1.9559% | yes | 1.5961% | 0 / 19 |
| 2022-08-01 → 2022-09-01 | -5.4279% | 1.8243% | 9.366% | -2.4806% | yes | 1.5063% | 2 / 22 |
| 2022-09-01 → 2022-10-01 | -5.0961% | 1.7901% | 9.3318% | -5.267% | no | 1.5794% | 2 / 20 |
| 2022-10-01 → 2022-11-01 | -6.5327% | 1.6649% | 10.7708% | 8.6503% | yes | 1.9496% | 2 / 20 |
| 2022-11-01 → 2022-12-01 | -5.9876% | 2.6007% | 12.1728% | 2.2983% | yes | 1.9595% | 1 / 20 |
| 2022-12-01 → 2023-01-01 | -6.211% | 3.2847% | 13.9653% | -3.1908% | yes | 1.969% | 1 / 20 |
| 2023-01-01 | no segment follows this rebalance, not scored | ||||||
Step 18 · 2023-01-03 → 2023-12-29
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 79 names held · selection: reselect · 8.4% in cash · turnover 7.9× · cost drag 0.79% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 11 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 1.83% of 219 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2023-01-01 | no segment follows this rebalance, not scored | ||||||
| 2023-02-01 → 2023-03-01 | -8.9505% | 1.2765% | 13.3246% | -2.8513% | yes | 2.3478% | 0 / 18 |
| 2023-03-01 → 2023-04-01 | -10.2901% | 1.4159% | 14.1871% | 1.6379% | yes | 2.3731% | 1 / 22 |
| 2023-04-01 → 2023-05-01 | -9.6027% | 1.0847% | 13.7437% | 0.1476% | yes | 2.5013% | 0 / 18 |
| 2023-05-01 → 2023-06-01 | -9.8264% | 1.5097% | 15.0989% | 2.1104% | yes | 2.5109% | 1 / 21 |
| 2023-06-01 → 2023-07-01 | -8.6777% | 1.3455% | 12.7075% | 6.1129% | yes | 2.2599% | 0 / 20 |
| 2023-07-01 → 2023-08-01 | -9.2136% | 1.2967% | 13.4964% | 1.4943% | yes | 2.5281% | 0 / 19 |
| 2023-08-01 → 2023-09-01 | -9.8752% | 1.7107% | 14.3295% | -1.2754% | yes | 2.6204% | 0 / 22 |
| 2023-09-01 → 2023-10-01 | -9.2166% | 1.0066% | 12.8388% | -6.4134% | yes | 2.2995% | 1 / 19 |
| 2023-10-01 → 2023-11-01 | -8.1931% | 1.6131% | 13.1646% | -3.5833% | yes | 2.2252% | 0 / 21 |
| 2023-11-01 → 2023-12-01 | -7.4514% | 2.0091% | 12.6595% | 5.263% | yes | 2.1021% | 0 / 20 |
| 2023-12-01 → 2024-01-01 | -5.8702% | 1.9124% | 10.673% | -1.7413% | yes | 1.8135% | 1 / 19 |
| 2024-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 84 names held · selection: reselect · 0.4% in cash · turnover 8.8× · cost drag 0.88%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 100.0% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 1.68% of 238 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2023-01-01 → 2023-02-01 | -6.5228% | 1.8038% | 11.2331% | 4.6892% | yes | 1.8179% | 0 / 19 |
| 2023-02-01 → 2023-03-01 | -6.9103% | 1.4484% | 11.0903% | -3.5373% | yes | 1.9418% | 1 / 18 |
| 2023-03-01 → 2023-04-01 | -9.9708% | 1.6056% | 14.2145% | 0.5303% | yes | 2.4194% | 1 / 22 |
| 2023-04-01 → 2023-05-01 | -9.1529% | 1.2857% | 13.6121% | 0.1517% | yes | 2.5203% | 0 / 18 |
| 2023-05-01 → 2023-06-01 | -9.3652% | 1.6222% | 14.7391% | 1.5998% | yes | 2.438% | 0 / 21 |
| 2023-06-01 → 2023-07-01 | -8.6777% | 1.3455% | 12.7075% | 6.1129% | yes | 2.2599% | 0 / 20 |
| 2023-07-01 → 2023-08-01 | -9.2136% | 1.2967% | 13.4964% | 1.4943% | yes | 2.5281% | 0 / 19 |
| 2023-08-01 → 2023-09-01 | -9.8752% | 1.7107% | 14.3295% | -1.2754% | yes | 2.6204% | 0 / 22 |
| 2023-09-01 → 2023-10-01 | -9.2166% | 1.0066% | 12.8388% | -6.4134% | yes | 2.2995% | 1 / 19 |
| 2023-10-01 → 2023-11-01 | -8.1931% | 1.6131% | 13.1646% | -3.5833% | yes | 2.2252% | 0 / 21 |
| 2023-11-01 → 2023-12-01 | -5.8282% | 1.4735% | 9.5077% | 4.7088% | yes | 1.6506% | 0 / 20 |
| 2023-12-01 → 2024-01-01 | -5.5169% | 1.4038% | 9.125% | -1.4422% | yes | 1.6009% | 1 / 19 |
| 2024-01-01 | no segment follows this rebalance, not scored | ||||||
Step 19 · 2024-01-02 → 2024-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 12 constructions · 97 names held · selection: reselect · 0.4% in cash · turnover 10.5× · cost drag 1.05% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 2.5% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2024-01-01 → 2024-02-01 | -8.1607% | 1.4942% | 12.3926% | 4.063% | yes | 2.0748% | 0 / 20 |
| 2024-02-01 → 2024-03-01 | -9.7281% | 1.5926% | 14.8489% | 3.1487% | yes | 2.5912% | 0 / 19 |
| 2024-03-01 → 2024-04-01 | -10.7784% | 1.5229% | 16.0902% | 1.8547% | yes | 2.9029% | 0 / 19 |
| 2024-04-01 → 2024-05-01 | -8.6643% | 2.6307% | 16.1452% | -4.8786% | yes | 2.5161% | 0 / 21 |
| 2024-05-01 → 2024-06-01 | -6.4007% | 2.7221% | 13.3751% | 5.8131% | yes | 2.0137% | 0 / 21 |
| 2024-06-01 → 2024-07-01 | -6.6168% | 1.9369% | 11.8205% | -0.7263% | yes | 2.0154% | 0 / 18 |
| 2024-07-01 → 2024-08-01 | -6.7852% | 2.2637% | 12.8264% | -1.3695% | yes | 1.8803% | 3 / 21 |
| 2024-08-01 → 2024-09-01 | -5.7526% | 2.0542% | 11.045% | 5.4513% | yes | 1.6837% | 1 / 21 |
| 2024-09-01 → 2024-10-01 | -5.3489% | 2.3458% | 10.9961% | 3.3997% | yes | 1.639% | 0 / 19 |
| 2024-10-01 → 2024-11-01 | -5.8944% | 2.373% | 11.0583% | -1.0269% | yes | 1.6895% | 0 / 22 |
| 2024-11-01 → 2024-12-01 | -6.323% | 2.2343% | 11.9455% | 10.3966% | yes | 2.0671% | 0 / 19 |
| 2024-12-01 → window end | -5.633% | 3.5961% | 13.944% | -6.2871% | no | 1.9643% | 2 / 20 |
B: without the market rule
Portfolio book, rebalanced monthly · 12 constructions · 97 names held · selection: reselect · 0.4% in cash · turnover 10.5× · cost drag 1.05%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 2.5% of 240 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2024-01-01 → 2024-02-01 | -8.1607% | 1.4942% | 12.3926% | 4.063% | yes | 2.0748% | 0 / 20 |
| 2024-02-01 → 2024-03-01 | -9.7281% | 1.5926% | 14.8489% | 3.1487% | yes | 2.5912% | 0 / 19 |
| 2024-03-01 → 2024-04-01 | -10.7784% | 1.5229% | 16.0902% | 1.8547% | yes | 2.9029% | 0 / 19 |
| 2024-04-01 → 2024-05-01 | -8.6643% | 2.6307% | 16.1452% | -4.8786% | yes | 2.5161% | 0 / 21 |
| 2024-05-01 → 2024-06-01 | -6.4007% | 2.7221% | 13.3751% | 5.8131% | yes | 2.0137% | 0 / 21 |
| 2024-06-01 → 2024-07-01 | -6.6168% | 1.9369% | 11.8205% | -0.7263% | yes | 2.0154% | 0 / 18 |
| 2024-07-01 → 2024-08-01 | -6.7852% | 2.2637% | 12.8264% | -1.3695% | yes | 1.8803% | 3 / 21 |
| 2024-08-01 → 2024-09-01 | -5.7526% | 2.0542% | 11.045% | 5.4513% | yes | 1.6837% | 1 / 21 |
| 2024-09-01 → 2024-10-01 | -5.3489% | 2.3458% | 10.9961% | 3.3997% | yes | 1.639% | 0 / 19 |
| 2024-10-01 → 2024-11-01 | -5.8944% | 2.373% | 11.0583% | -1.0269% | yes | 1.6895% | 0 / 22 |
| 2024-11-01 → 2024-12-01 | -6.323% | 2.2343% | 11.9455% | 10.3966% | yes | 2.0671% | 0 / 19 |
| 2024-12-01 → window end | -5.633% | 3.5961% | 13.944% | -6.2871% | no | 1.9643% | 2 / 20 |
Step 20 · 2025-01-02 → 2025-12-31
A: the book's rules
Portfolio book, rebalanced monthly · 13 constructions · 84 names held · selection: reselect · 0.4% in cash · turnover 8.3× · cost drag 0.83% · regime gate active
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 10.92% of 238 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2025-01-01 → 2025-02-01 | -6.1061% | 2.9464% | 13.2683% | 7.2127% | yes | 1.9646% | 1 / 19 |
| 2025-02-01 → 2025-03-01 | -5.2216% | 2.9264% | 12.2913% | -4.0518% | yes | 1.8326% | 3 / 18 |
| 2025-03-01 → 2025-04-01 | -3.5439% | 3.1049% | 10.3617% | -5.4102% | no | 1.3792% | 5 / 20 |
| 2025-04-01 → 2025-05-01 | -3.7949% | 2.3512% | 9.0273% | -0.432% | yes | 1.3284% | 5 / 20 |
| 2025-05-01 → 2025-06-01 | -5.0194% | 2.8976% | 11.6562% | 5.0349% | yes | 1.5903% | 0 / 20 |
| 2025-06-01 → 2025-07-01 | -3.2299% | 1.7108% | 7.1067% | 1.0272% | yes | 1.0101% | 1 / 19 |
| 2025-07-01 → 2025-08-01 | -2.8881% | 2.2044% | 7.8996% | -2.8115% | yes | 0.9557% | 4 / 21 |
| 2025-08-01 → 2025-09-01 | -5.4967% | 2.8953% | 12.2266% | -0.165% | yes | 1.6193% | 1 / 20 |
| 2025-09-01 → 2025-10-01 | -7.3057% | 2.511% | 13.5998% | 8.516% | yes | 2.056% | 0 / 20 |
| 2025-10-01 → 2025-11-01 | -6.494% | 3.7065% | 14.6281% | 4.1984% | yes | 1.9339% | 1 / 22 |
| 2025-11-01 → 2025-12-01 | -6.3006% | 3.1739% | 14.2197% | 2.4201% | yes | 2.074% | 3 / 18 |
| 2025-12-01 → 2026-01-01 | -5.8825% | 3.9624% | 15.5356% | 3.0889% | yes | 2.0464% | 2 / 21 |
| 2026-01-01 | no segment follows this rebalance, not scored | ||||||
B: without the market rule
Portfolio book, rebalanced monthly · 13 constructions · 90 names held · selection: reselect · 0.4% in cash · turnover 9.0× · cost drag 0.9%
Projection accuracy, realized outcome fell inside the P5–P95 cone in 91.7% of 12 scored rebalances (a well-calibrated 90% band contains ~90.0%) · 95% VaR breached on 10.92% of 238 days (expected ~5.0%)
| Rebalance | P5 | Median | P95 | Realized | In band | VaR 95 (1d) | Breaches |
|---|---|---|---|---|---|---|---|
| 2025-01-01 → 2025-02-01 | -6.1061% | 2.9464% | 13.2683% | 7.2127% | yes | 1.9646% | 1 / 19 |
| 2025-02-01 → 2025-03-01 | -5.2216% | 2.9264% | 12.2913% | -4.0518% | yes | 1.8326% | 3 / 18 |
| 2025-03-01 → 2025-04-01 | -3.5439% | 3.1049% | 10.3617% | -5.4102% | no | 1.3792% | 5 / 20 |
| 2025-04-01 → 2025-05-01 | -3.0312% | 1.9577% | 7.3128% | -0.6426% | yes | 1.0679% | 5 / 20 |
| 2025-05-01 → 2025-06-01 | -3.519% | 1.5981% | 7.0991% | 0.8332% | yes | 1.1222% | 3 / 20 |
| 2025-06-01 → 2025-07-01 | -3.6024% | 1.4861% | 7.0529% | 2.2695% | yes | 1.1772% | 1 / 19 |
| 2025-07-01 → 2025-08-01 | -3.2276% | 2.5885% | 9.1418% | -0.6319% | yes | 1.0898% | 1 / 21 |
| 2025-08-01 → 2025-09-01 | -5.456% | 2.8689% | 12.1194% | -1.2832% | yes | 1.5792% | 1 / 20 |
| 2025-09-01 → 2025-10-01 | -7.3862% | 2.6128% | 13.9288% | 7.7222% | yes | 2.1427% | 0 / 20 |
| 2025-10-01 → 2025-11-01 | -6.5916% | 3.7912% | 14.9283% | 3.6471% | yes | 1.936% | 1 / 22 |
| 2025-11-01 → 2025-12-01 | -6.5436% | 3.2581% | 14.7255% | 2.7763% | yes | 2.1641% | 3 / 18 |
| 2025-12-01 → 2026-01-01 | -5.7219% | 4.1259% | 15.701% | 2.6745% | yes | 2.0588% | 2 / 21 |
| 2026-01-01 | no segment follows this rebalance, not scored | ||||||