Expectancy: the only math that decides whether you make money

Updated 3 September 2026 · Belphor Research · the data behind this guide

One number decides whether a strategy earns or bleeds: expectancy, the average outcome per trade once wins, losses, their sizes, and all costs are netted. It is simple arithmetic with an inconvenient personality, because it routinely contradicts the two things traders love most: high win rates and vivid recent memories.

The formula, and what it hides

Multiply your win rate by your average win; subtract your loss rate times your average loss; subtract the round-trip cost of trading. What remains is your earnings per trade, on average, and multiplied by trade count it is your whole result. Everything else in trading, entries, indicators, narratives, is upstream decoration: it matters only insofar as it moves one of these four numbers. The formula's inconvenient property is that the terms trade against each other, so improving the one you can see often degrades the one you cannot.

The high win rate trap

A strategy that takes profits at plus five percent and stops out at minus fifty can win nine trades in ten and still lose money: nine times five is forty-five, one fifty-point loss erases it, and costs finish the job. Yet that strategy feels wonderful, an unbroken green streak with occasional bad luck. The inverse feels awful and pays: lose seven small, win three huge. Win rate is an emotional number; expectancy is a financial one, and in this market they usually point in opposite directions.

What memecoin data actually shows

When Belphor replayed thousands of its own historical signals against tight take-profit and stop grids, every small-target combination came out negative once realistic fees and slippage were charged; the best still lost nearly two percent per trade. The same data showed the winning shape: cohorts where most trades lost modestly and rare outliers gained hundreds of percent. The conclusion is structural, not stylistic: this market pays from the right tail, so the profitable design holds losses small, lets winners run behind a trailing floor, and treats capping the outlier as the one fatal error.

Costs: the silent sign-flipper

Because costs subtract from every trade, they punish frequency and small targets hardest. A two percent round trip is trivia to a strategy hunting three-hundred percent moves and a death sentence to one scalping fives. Before believing any backtest or screenshot, ask one question: are fees and slippage inside these numbers? Most published records that look brilliant are answering no. The full cost stack is in crypto trading fees explained.

Living with the math

Positive expectancy still loses often; that is what a thirty percent win rate means, and it is why sample size, sizing, and the stomach to keep executing through streaks are part of the edge. The practical routine: track every closed trade, compute expectancy per strategy rather than in aggregate, judge nothing on fewer than dozens of trades, and let the number, not the streak, decide what keeps running. This is precisely why Belphor publishes every close and audits per-preset returns nightly: expectancy only disciplines a strategy if it is computed where everyone can see it.