Understanding Your Bot's Performance Through Equity Curve and P&L Distribution

Two numbers tell you how your bot is doing right now: total P&L and win rate. But they don't tell you the shape of how you got there, whether you've been steadily building, recovering from a drawdown, or riding a lucky streak that may not be representative of the bot's true profile.

Relay's Equity Curve and P&L Distribution modes give you that shape. Used together, they provide a much more complete picture of what your bot's performance actually looks like, and what you can realistically expect from it going forward.

Equity curve: the story of your P&L over time

What it shows

An equity curve is a line chart of your cumulative P&L over time. Each point on the curve represents the running total of all your closed trades up to that moment.

Rather than summarizing performance as a single number, the equity curve shows you the path, the ups, the downs, the recoveries, and the overall trajectory.

Triggering the equity curve

From the capability grid: Click "Equity Curve" in the Relay capability menu.

Through conversation:

"Show me the equity curve for my QQQ UT bot." "Plot my cumulative P&L over the last 90 days." "Show me the equity curve for all my SPY bots."

You can request curves for individual bots or aggregated across your portfolio. Specifying a time range gives you a more focused view:

"Show me the equity curve for QQQ UT (SHRT) for the last 60 days."

What to look for

Consistent upward slope The ideal equity curve trends upward with relatively small deviations. This indicates a bot that's producing consistent, repeatable results. The slope doesn't have to be steep, consistency matters more than speed.

Drawdown depth and duration Every bot will have periods where the equity curve declines, a series of losing trades or a stretch of stop-outs. What you're evaluating is how deep those drawdowns go and how quickly the curve recovers. A deep drawdown that recovers slowly is a signal worth paying attention to, even if the bot eventually came back positive.

Inflection points A sudden change in the slope of the equity curve, a sharp improvement or a sharp decline, often corresponds to a change in market conditions or (if you've made changes) a configuration update. If you see an inflection point, it's worth understanding what caused it.

Equity curve stall If the curve flattens out for an extended period, the bot is trading roughly breakeven. This isn't necessarily a problem. It may just reflect low-signal conditions. But a prolonged stall followed by a decline is a pattern that warrants investigation.

Late-period trajectory The most recent portion of the equity curve tells you the most about where things stand right now. A bot whose overall curve is positive but whose recent trajectory has turned negative is in a different situation than one whose curve has been steadily positive throughout.

P&L distribution: the spread of your outcomes

What P&L distribution shows

A P&L distribution is a histogram of your individual trade outcomes. Each bar represents a range of P&L values, and the height of the bar shows how many trades fell in that range.

Where the equity curve shows you the story over time, the distribution shows you the shape of your typical trade outcome. It answers the question: when trades close, where do they tend to land?

Triggering P&L distribution

From the capability grid: Click "P&L Distribution" in the Relay capability menu.

Through conversation:

"Show me the P&L distribution for QQQ UT." "What does the distribution of trade outcomes look like for my SPY bots?" "Show me a histogram of my trade P&L over the last 90 days."

What to look for in the distribution

Where the distribution is centered A well-calibrated bot should have most trades concentrated in a positive range. If the center of the distribution is near zero or slightly negative, the bot is producing marginally positive results that could swing negative easily.

The shape of the losing tail In options trading, stop losses create a floor on individual trade losses. The distribution should show a clear edge on the left side (the stop loss level) beyond which there should be very few or no trades. If you see losses beyond your stop level, it may indicate slippage or fill quality issues.

The shape of the winning tail Where your winners cluster tells you whether your take profit levels are well-positioned. If most winning trades cluster tightly at TP1 with very few reaching TP2 or TP3, either your TP2/TP3 levels are rarely achievable or exits are happening before those levels are reached for other reasons.

Outlier trades Very large wins or very large losses that fall far outside the main distribution are worth examining individually. Large wins can distort your average P&L upward in ways that overstate typical performance. Large losses can indicate rare conditions where stop loss protection didn't work as expected.

The win/loss ratio in context A distribution that shows many small wins and occasional larger losses is a different profile than one showing modest wins and consistently contained losses. Neither is inherently better — but understanding which profile your bot produces helps you evaluate it accurately.

Using equity curve and distribution together

The two views complement each other:

| What you're evaluating | Best view |
|---|---|
| Is my bot trending in the right direction? | Equity Curve |
| Was there a recent change in behavior? | Equity Curve inflection points |
| How deep can drawdowns get? | Equity Curve troughs |
| Are my TP levels capturing the right moves? | P&L Distribution winning tail |
| Is my stop loss working as intended? | P&L Distribution losing tail |
| Are outlier trades distorting my averages? | P&L Distribution extremes |

A bot with a smoothly rising equity curve and a tight, positive-centered distribution is performing consistently. A bot with a choppy equity curve and a wide, spread distribution has more variance, which may or may not be a problem depending on how you've configured it.

Practical Application

If the equity curve shows a sustained drawdown: Check the P&L distribution for the same period. If the distribution has shifted left (more trades landing in the losing range), the drawdown reflects a real change in performance. If the distribution looks similar to the positive period but you've had more trades than usual, the drawdown may reflect normal variance rather than a strategy problem.

If the distribution shows most winners clustering only at TP1: Use What-If Simulation to model what would happen if you moved TP2 closer to TP1. The distribution shows you the opportunity; the simulation helps you evaluate whether adjusting would improve outcomes.

If you see a large outlier loss in the distribution: Use Trade Replay to examine that specific trade. Understanding what happened in the worst-case trade helps you evaluate whether the risk profile of the bot is what you intended.