Why Win Rate Is the Wrong Metric to Optimize

Win rate, the percentage of trades that end in a profit, is the metric the trading-education industry advertises most, because a high number sounds impressive and intuitive: win 90% of the time and surely you are winning. It is also, on its own, nearly useless as a measure of whether a strategy makes money, and optimizing for it can actively lead you toward losing strategies. What actually determines profitability is expectancy, the average amount you make or lose per trade accounting for both how often you win and how much you win and lose. This piece shows the math, demonstrates how a high win rate can lose money and a low win rate can be highly profitable, and explains why the shiny percentage is not just incomplete but often a warning sign.

The Formula That Actually Matters

Expectancy is the average result you can expect per trade over many trades, and it is what determines whether a strategy makes money. The formula is straightforward: expectancy equals your win rate times your average win, minus your loss rate times your average loss. In words, it weighs how often you win against how much you win, and how often you lose against how much you lose, and nets them out. A strategy is profitable if and only if its expectancy is positive, which is to say the wins, sized and weighted by frequency, outweigh the losses, sized and weighted by frequency.

Notice what this formula contains and what win rate alone leaves out. Win rate is only one of the four inputs; the average win and the average loss, the sizes, matter just as much, and they are exactly what a win-rate figure discards. Two strategies with identical win rates can have opposite expectancies if their win and loss sizes differ, which means win rate by itself cannot tell you whether a strategy is profitable. It is one piece of a four-piece calculation, and quoting it alone is quoting a quarter of the answer as if it were the whole.

How a 90% Win Rate Loses Money

The demonstration that makes this concrete: a strategy can win the overwhelming majority of its trades and still lose money overall. Suppose a strategy wins 90% of the time, but each win is small and each of the rare losses is large. Say each win makes 1 unit and each loss costs 10 units. Over 100 trades, you win 90 times for 90 units, and you lose 10 times for 100 units. Your win rate is a spectacular 90%, and your net result is a loss of 10 units. The strategy wins nine times out of ten and bleeds money, because the size of the occasional loss overwhelms the many small wins.

This is not a contrived edge case; it is the natural profile of an entire class of strategies, and that is the important part. Strategies that sell premium, short options, credit spreads, condors, and similar structures, characteristically produce exactly this shape: a high win rate of small, reliable gains, punctuated by occasional large losses when the market moves against the short position. The high win rate is real and it is seductive, and it coexists with negative expectancy whenever the large losses are large enough. A trader optimizing for win rate would be delighted by this strategy right up until the accumulated small wins are erased by a run of large losses, which the win rate gave no warning of, because the win rate was never measuring the thing that would sink the account. This is the same short-gamma trap discussed in the pieces on IV crush and on theta decay: the reliable small win hides the rare large loss, and win rate is precisely the metric that hides it best.

How a 40% Win Rate Prints Money

The mirror case is just as important and even more counterintuitive to someone anchored on win rate. A strategy can lose the majority of its trades and be highly profitable. Suppose a strategy wins only 40% of the time, but each win is large and each loss is small, cut quickly. Say each win makes 10 units and each loss costs 2 units. Over 100 trades, you win 40 times for 400 units and lose 60 times for 120 units, netting a profit of 280 units on a win rate of just 40%. You are wrong more often than you are right, and you make excellent money, because the wins are far larger than the losses.

This is the profile of trend-following and many momentum strategies: frequent small losses as most attempts fail, punctuated by occasional large wins that more than pay for all of them. A trader optimizing for win rate would reject this strategy as a loser, six failures for every four successes, and would be rejecting a highly profitable approach because they were measuring the wrong thing. The low win rate is not a flaw; it is a feature of a strategy whose edge lives in the size of its wins, not their frequency, and win rate systematically makes such strategies look bad.

Why Win Rate Is Not Just Incomplete but a Warning Sign

Put the two cases together and win rate is revealed as not merely an incomplete metric but a misleading one, and its prominence in marketing is itself informative. A high win rate is the easiest performance number to advertise and the most emotionally compelling, win 90% of the time sounds like mastery, and it is exactly the number a marketer reaches for when selling a strategy or service, precisely because it impresses without informing. When a trading product leads with its win rate and is quiet about its average win versus average loss, or about its worst drawdown, that emphasis is a signal: the win rate is being used to distract from the sizes, which is where the profitability actually lives and where a high-win-rate strategy often hides its fatal flaw. A genuinely profitable strategy can be described honestly with its full expectancy profile; a strategy that must be sold on win rate alone is often one whose sizes would not survive disclosure.

So the practical posture toward any advertised win rate is skepticism. Ask what the average win and average loss are, ask what the worst losing streak and largest single loss looked like, and ask what the expectancy is once all of that is accounted for. A win rate with no win-and-loss sizes attached is not evidence of profitability; it is a number chosen because it sounds good, and the absence of the sizes is the tell.

What to Optimize Instead

The constructive answer is to optimize for expectancy, not win rate, and to hold win rate as one descriptive input rather than a goal. A positive expectancy is the necessary condition for profitability, and it can be achieved with a high win rate and small edge per trade, or a low win rate and large edge per trade, or anywhere in between; what matters is that the frequency-weighted wins exceed the frequency-weighted losses. Alongside expectancy, the metrics worth attending to are the ones win rate hides: the average win and average loss and their ratio, the largest single loss and the worst drawdown, which tell you whether the strategy is survivable, and the consistency of the expectancy across different conditions. These describe whether a strategy makes money and whether you can survive running it, which is the actual question. Win rate, by contrast, describes only how it feels to trade, frequent wins feel good, and feeling good is not the objective.

How This Connects to the Platform

StaxInvesting is a self-hosted platform for automating options strategies, and the expectancy-over-win-rate principle shapes how the platform presents information and how it should be evaluated. The backtester and paper-trading tools report the full picture, wins and losses and their sizes, drawdowns, the complete distribution of outcomes, rather than reducing a strategy to a flattering headline percentage, because a validation that showed only win rate would be hiding exactly what determines profitability. And in a deliberate design choice reflecting the same principle, the community leaderboard is ranked by percentage gain rather than by win rate or by raw dollar amount, so that what is surfaced is closer to actual results than to the shiny metric.

The honest framing is the one this whole piece argues, and it applies to the platform as much as to anyone. A high win rate produced by any strategy on the platform is not evidence the strategy is profitable; only its expectancy is, and expectancy depends on sizes the win rate conceals. Automation executes whatever strategy you give it, and a high-win-rate, negative-expectancy strategy automated is negative expectancy executed reliably, small wins accumulating until a run of large losses erases them, which is why the validation tools report the full distribution and why you should read it rather than the headline number. The platform can execute a positive-expectancy strategy with discipline; it cannot make a negative-expectancy strategy profitable, however good its win rate looks, a point developed in the piece on what automated options trading can and cannot do. The broader context is in the post-PDT market regime analysis, and the engineering behind the reporting and backtesting in the Node.js performance material and the worker thread pool reference.

The Short Version

Win rate, the percentage of trades that win, is the metric the industry advertises and nearly useless on its own, because profitability is determined by expectancy: win rate times average win, minus loss rate times average loss. A 90% win rate loses money if the rare losses are large enough to overwhelm the many small wins, which is the natural profile of premium-selling strategies and the exact shape that hides a fatal flaw behind an impressive percentage. A 40% win rate prints money if the wins are large and the losses are cut small, which is the profile of trend-following and would be wrongly rejected by anyone optimizing for win rate. Because a high win rate is the easiest number to advertise and the best at concealing the sizes that actually matter, a product that leads with win rate and is quiet about average win versus loss and worst drawdown is waving a warning flag. Optimize for expectancy, demand the sizes behind any win rate, and treat the shiny percentage as what it is: how it feels to trade, not whether you make money.

Past performance does not guarantee future results, and nothing on this page is financial, legal, or tax advice or a recommendation to buy or sell any security or options contract, or to pursue any strategy. The numerical examples are simplified illustrations of the expectancy relationship, not representations of any actual strategy or result. StaxInvesting LLC provides software tools and educational content; it is not a broker-dealer or a registered investment adviser, does not provide personalized investment advice, and never accesses member funds, credentials, accounts, or trades. Options trading involves substantial risk of loss and is not suitable for all investors; research indicates most retail options traders lose money, and losses can exceed deposits. A high win rate does not indicate profitability; automated execution acts on the strategy and settings you configure, executes a negative-expectancy strategy as faithfully as a positive one, does not create an edge, and does not guarantee a profitable outcome. Consult a licensed financial professional regarding your own circumstances.