Indicator Strategies Workflow

The Indicator Strategies tab is where you combine any indicators freely — no family constraints. If Basic Strategies is the guided-rails version, this is the open workshop. Same backtesting engine underneath, same output format, but you choose every part of the rule yourself.

What lives in this tab

• Regime Analysis — Splits an asset's history into statistical regimes (low / mid / high volatility, trending / range-bound) and shows how returns, drawdowns and indicator behaviour differ in each. Read this first so you know what market you are actually designing for.
• Overlay/Builder — The universal strategy builder: every major indicator family (momentum, mean reversion, crossover, breakout) in one place, three rule layers (Entry, Confirmation, Exit), long or short, stop-loss and take-profit, in-sample / out-of-sample split, plus a one-click standalone Python export that replicates your exact strategy offline.
• Backtester — The portfolio counterpart to Overlay/Builder. Loads up to 10 saved configs and runs them together with shared capital, producing a combined equity curve, per-config breakdown and correlation matrix.
• Regime Optimizer — Lets you pick different parameters per regime. You build a base rule in Overlay/Builder, then let Regime Optimizer find the best parameter set for each regime slice of history. Useful when a single parameter set clearly underperforms in one environment.

Canonical workflow

1. Understand the asset first. Open Regime Analysis on your ticker. Note whether it spends most of its time trending or range-bound, and where the volatility regimes sit. This is the cheapest way to avoid designing a mean-reversion rule for a trend-heavy asset.
2. Compose a rule in Overlay/Builder. Entry layer (e.g. RSI crosses below 30), optional Confirmation (e.g. Bollinger %B below 0.2 on the same bar), Exit (e.g. RSI crosses above 60 or 10 bars elapsed). Set direction, stops and targets. Set an in-sample / out-of-sample split — typically 70/30.
3. Backtest on in-sample only. Look at win rate, average gain vs average loss, max drawdown, Sharpe. Tune parameters here.
4. Reveal out-of-sample. Click the OOS reveal once you're done tuning. If OOS metrics collapse vs. in-sample, you overfit — go back to step 2 with simpler rules.
5. Save the config. It lands in the Execution ribbon and also becomes available to the Backtester and to Advanced Paper Trading.
6. Either forward-test or combine. Deploy to Advanced Paper Trading (PT004UBPT) to run it live against virtual capital, or load it into Backtester (UB002UBKT) alongside other configs to see portfolio-level behaviour.
7. If the rule is regime-sensitive, try Regime Optimizer. Load the same base config into Regime Optimizer (RG001RGMO). It sweeps parameters per regime and returns a per-regime parameter map that the Overlay/Builder can then run as a single regime-aware strategy.

Reading the outputs

• Trade table — Every entry and exit, signed P&L, hold duration. Filter by win/loss to study the bad trades first.
• Equity curve — Strategy vs. buy-and-hold. A curve that beats buy-and-hold but is far noisier is usually not worth the operational cost.
• Sharpe and max drawdown — Max drawdown is more decision-relevant than Sharpe for most retail setups. A 40% drawdown you wouldn't actually sit through invalidates the strategy regardless of Sharpe.
• OOS vs IS — The gap is the honest overfitting signal. A rule that looks great on IS and ordinary on OOS is usually the one you built, not the one the market has.

Common pitfalls

Deeper reading

• Trading Strategies → Mean Reversion Overview and Momentum Overview for the underlying theory.
• Trading Strategies → Composite Scoring and Market Regime Detection for rule-design patterns that compose well.
• Platform Guides → Stochastic Methods Workflow if you want a model-driven alternative to indicator-based rules.