Positive after costs, and not resting on two lucky trades

A good expectancy per trade is one that stays above zero after every cost, across enough trades to trust. Above zero means the rules made money on average. Below zero means they lost it.

People want a benchmark, like a set number of dollars or a fraction of risk. There isn't a fair one. A scalper with small winners and a swing trader with large ones can both be fine on very different numbers.

What makes a number good is whether it survives. It has to survive commissions and slippage. And it has to survive losing its best couple of trades, because those may not come round again.

The expectancy entry in our glossary covers the definition in full. This piece is about judging your own number, using a journal export and a calculator. It also covers the one comparison worth more than any benchmark: your number this month against your number last month.

The formula: expectancy equals win rate times average win, minus loss rate times average loss.

Export your last 50 closed trades. If the export lists each fill separately, combine partial fills into one trade first. A position closed in three pieces otherwise counts as three winners and flatters the win rate.

Suppose 21 were winners averaging 180 dollars and 29 were losers averaging 95 dollars. The win rate is 21 out of 50, or 42 percent. The loss rate is 58 percent.

Now the arithmetic. 0.42 times 180 is 75.60. 0.58 times 95 is 55.10. Subtract and you get 20.50 dollars per trade.

As a check, total profit is 21 times 180, or 3,780, minus 29 times 95, or 2,755. That's 1,025 dollars over 50 trades, the same 20.50 per trade.

A 42 percent win rate feels mediocre. The expectancy says the rules still made money, because the average winner is almost twice the average loser.

Check whether your export's profit column is gross or net. Some broker reports show results before commissions and fees, and a few show both.

If it's gross, take the costs off. Say you pay 3 dollars a round trip in commission and fees. The 20.50 drops to 17.50 per trade. That's still positive, but the cushion is thinner than it looked. Options traders feel this most, because commission is often charged per contract on both sides. Count every contract when you total it.

Slippage is the cost an export hides. Live fills already include it, so a real export accounts for it automatically. A paper or replay record doesn't, and needs a realistic deduction on each side.

The smaller your average trade, the more this matters. If your expectancy is a few dollars and your costs are a few dollars, you're mostly paying your broker. Fix the costs or find bigger moves before judging the rules.

A positive average can rest on one or two outliers. Sort the 50 trades by profit and look at the top of the list. Then look at the bottom, because outsized losers distort the number the other way.

In the example, total profit was 1,025 dollars. Suppose the two biggest winners made 600 and 450. Together that's 1,050. Remove them and the other 48 trades lost 25 dollars between them. Expectancy drops from 20.50 to about minus 0.52 per trade.

That doesn't make the edge fake. Big winners are part of plenty of sound approaches. It does mean the result depends on catching those rare trades, and on never skipping them.

So ask two questions. Were those outliers taken under your normal rules? Would you have held them the same way again? If either answer is no, your real expectancy sits closer to the negative figure. If both are yes, keep the rules and protect the habit that caught them.

Dollars depend on size, and size drifts over time. Divide each trade's result by the amount you risked on it to get the result in R.

If you risked about 100 dollars a trade in the example, 17.50 dollars after costs is 0.175R. Each trade adds, on average, a bit under a fifth of what you put at risk.

In R you can compare this quarter with last quarter, even if your size changed in between. The breakeven arithmetic gets plain too. When your target is twice your risk, you break even by winning one trade in three. That's about 33 percent.

To check your own number fast, the free expectancy calculator does the arithmetic from a win rate and two averages. Run it once with your raw export and once with costs and outliers removed. The gap between the two answers is the real lesson. A small gap means the number is sturdy. A large one tells you which leak to fix first.

Do this once a month with a fresh export.