Track Strategy Performance Over Time (Trends & Capital)

A strategy that looked brilliant in January can bleed quietly through March while lifetime stats still show green. Snapshot metrics — one month’s win rate, one quarter’s profit factor — hide drift until capital or confidence is already damaged. Tracking strategy performance over time means watching rolling windows, per-strategy equity curves, and capital allocation against pre-written rules. For Indian traders juggling Nifty scalps, Bank Nifty expiry plays, and separate cash swing books, time-series discipline separates scaling decisions from revenge sizing after a red Tuesday. This guide covers how to track playbooks across weeks and months inside TradeLyser and when to scale, hold, or pause.

Snapshots versus trends

A single month’s P&L answers “How did I do?” It does not answer “Is this edge stable?” Rolling metrics — last 20 trades, last 30 calendar days, last 60 sessions — reveal direction: improving, flat, or decaying. Two strategies with identical lifetime net P&L can have opposite rolling trends; only one deserves increased size. The methodology strategies pillar (Strategies pillar) assumes you compare playbooks on equal windows, not whichever period flatters each setup.

Treat snapshots as photographs and rolling series as video. Photographs lie easily when sample size is small or one expiry week dominates. Video is slower but shows regime change before the account shocks you.

Per-strategy equity curves

Account-level equity curves blend everything — impulsive trades, experimental tags, and your core edge. Per-strategy curves in Strategy Board (Strategy Board) isolate each playbook so you see drift early. A flat or declining curve on a “core” setup while the account stays green means something else is subsidising the book — investigate before size-up.

• Plot net P&L cumulatively by closed trade sequence, not calendar alone — trade count matters for frequency.
• Mark rule-change dates on the chart mentally or in journal notes.
• Compare curve shape: smooth grind versus cliff — cliffs often mean tail risk or size violations.
• Overlay max drawdown bands — depth and duration both matter for rupee-based risk limits.

Rolling metrics that catch decay early

Configure a fixed review set and run it every cycle without adding new indicators. Suggested rolling panel for active NSE day traders:

• Rolling 20-trade expectancy — direction matters more than the absolute level week to week.
• Rolling profit factor — sustained drop below your floor triggers investigation, not instant deletion.
• Rolling win rate with average win/loss — catches early profit-taking or stop widening.
• Rolling max adverse excursion if you log it — execution quality separate from outcome.
• Trade frequency — overtrading spikes often precede metric decay.

Swing traders may prefer rolling 30-day or last-15-closed-trades windows. Pick one convention per strategy and keep it for at least two review cycles before changing — otherwise you optimise noise.

Market regime and Indian session context

Low VIX, high VIX, trend, chop — edges rotate on the same Nifty chart. Expiry Thursdays on index derivatives compress time and raise gamma risk; many intraday edges behave differently that day. Tag session context or filter performance by week type so rolling metrics are interpreted against regime, not blamed on “bad luck”.

When rolling expectancy turns negative while lifetime stats look fine, treat it as a regime alert: reduce size, paper trade updates, or pause until a fresh sample rebuilds. The rules pillar (Rules pillar) should already define pause thresholds in rupees — tracking over time is how you hit those thresholds objectively.

Capital buckets and scaling rules

Think in buckets: core (proven expectancy, stable drawdown), research (promising but insufficient sample), and paused (negative rolling metrics or rule chaos). Never promote a strategy to core because of one hot streak after Budget day or a trending Bank Nifty fortnight.

Example policy: core strategies share 80% of risk budget weighted by rolling expectancy; research shares 15%; 5% cash buffer for mental clarity. Numbers are personal — write yours on paper and follow them when the screen is red at 10:45.

When to scale, hold, or pause

Scale up only after stability

Increase size only when rolling expectancy is positive across two consecutive review cycles, drawdown stayed within policy, and sample size meets your minimum. One viral week is not stability — verify symbol concentration and session dependence first.

Hold when signal is mixed

Lifetime positive but rolling flat often means “do nothing dramatic”. Fix tagging hygiene, note regime in journal, and wait for the next full cycle. Mixed signal is the default state — resist impulsive rule surgery.

Pause with a written protocol

Two consecutive review cycles below expectancy floor → pause live size → 20-trade paper sample with at most one rule change → resume only if paper metrics recover. Emotional attachment to a “bread and butter” setup is why written policies exist. Pausing is data hygiene, not failure.

Tracking multiple strategies without blur

If you run opening range breakout, midday mean reversion, and swing pullbacks, each needs its own curve and rolling panel. Overlap tagging without rules splits metrics and inflates trade counts. Document primary-tag rules in your strategy spec; Strategy Board comparisons only work when tags are honest.

For options, never merge structure types. Short premium and long directional books have different curve shapes — blending them produces a curve nobody can act on.

Weekly tracking workflow

• Fix the comparison window (e.g. last 90 days or last 200 trades per strategy).
• Open Strategy Board (Strategy Board) — export or note side-by-side stats.
• Update rolling 20-trade expectancy trend: improving, flat, decaying.
• Check per-strategy equity curve shape and drawdown vs policy.
• Slice symbol and session — flag if one ticker or one hour drives results.
• Write allocation decision (scale/hold/pause) before Monday open.
• Close the loop in weekly review (Weekly review) with one journal paragraph on why.

Link performance to journal context

Numbers without narrative mislead future you. When rolling metrics shift, log what changed: new slippage on market orders, skipped pre-market plan, family distraction week, or genuine volatility regime. The journals methodology (Journals pillar) treats context as data — performance tracking should point back to those notes when curves bend.

Insights layer — verify, do not react

Charts and reports in the insights pillar (Insights pillar) can highlight deteriorating expectancy or concentration faster than manual spreadsheets. Use them as triage: open the underlying trades, confirm the pattern, then decide. Colour on a dashboard is not permission to double size on Friday expiry.

Document every allocation decision

Tracking without written decisions is entertainment. Each Friday, log scale/hold/pause per strategy tag in one sentence: “ORB-nifty: hold — rolling expectancy flat, drawdown within ₹12k cap.” Next month, read those sentences before changing size — they expose whether you followed policy or traded mood. Mentors can review decision logs without debating individual tick entries; the habit builds institutional memory for solo traders too.

Indian market examples of drift

A Nifty opening-range breakout tag can show lifetime green while rolling twenty-trade expectancy turns negative after VIX compression — ranges fake breakouts for three weeks. A weekly iron condor tag can show smooth equity until one budget gap erases a quarter; per-strategy curves make tail risk visible before you add lots. Bank Nifty scalpers often see frequency drift: trade count rises in chop while A-setup percentage falls — rolling panels catch that before margin stress.

Cash swing tags on NSE large-caps may look stable until earnings season concentrates wins in two reporting weeks — symbol breakdown plus rolling expectancy prevents promoting a “stock picker” narrative when the edge was event variance.

Separate sim curves from live curves

Paper and sim tags must not merge into live equity curves until promotion rules are met — usually minimum sim sample, stable rolling metrics, and documented rule adherence. Retail traders promote sim heroes after one good week; per-strategy tracking keeps sim experiments visible without contaminating capital decisions on core tags.

When rolling metrics disagree with lifetime stats for more than two review cycles, default to rolling — markets change faster than lifetime denominators update. Lifetime stats are legacy; rolling stats are steering. Write that sentence in your strategy notebook so Friday reviews stay consistent when ego prefers lifetime green.

Closing: time is the honest judge

Tracking strategy performance over time is how you earn the right to scale. Indian markets offer the same candles to everyone; edge differences show up in rolling expectancy, not in hindsight stories. Build the weekly habit in Strategy Board, tie decisions to written rules, and let trends — not Tuesday’s mood — govern your capital.

Open your core tag’s rolling twenty-trade panel before next Monday: note direction, write scale/hold/pause in one line, and file it beside last week’s line. Twelve consecutive lines of honest decisions beat any new scanner subscription for capital preservation on NSE books.

If you manage multiple tags, colour-code decision log lines by bucket — core, research, paused — so monthly audit takes five minutes. Visual consistency beats another dashboard widget you will ignore by Wednesday.