Quant Memo
Core

Reviewing P&L Per Trade

A book's total P&L can look healthy while individual trades are a mess underneath it — a handful of big winners masking a majority of small losers. Reviewing every trade separately is how you find out which one you actually run.

Prerequisites: Reading Your Daily P&L

The book is up $400,000 for the month. Good month, on its face. But that single number is an average over dozens of individual trades, and averages hide the shape of what's underneath. Two very different books can produce the exact same monthly total: one where most trades are small winners and losses are rare, and one where two-thirds of trades lose money and the total is rescued by a couple of outsized winners. Those are different businesses to be running, with different risks going forward, and the only way to tell them apart is to review P&L trade by trade, not month by month.

The two numbers that matter together

Hit rate: the fraction of trades that were profitable. Average win / average loss ratio: how big the winners are relative to the losers. Neither one alone tells you whether the strategy is sound — a low hit rate can still be a great strategy if winners are large enough, and a high hit rate can still be a bad strategy if the rare losers are catastrophic.

Expectancy=(hit rate×avg win)((1hit rate)×avg loss)\text{Expectancy} = (\text{hit rate} \times \text{avg win}) - ((1 - \text{hit rate}) \times \text{avg loss})

In words: expected profit per trade is the chance of winning times what you win on average, minus the chance of losing times what you lose on average.

Worked example

Thirty trades this month, total P&L +$400,000. Reviewed individually: 11 were winners (hit rate 37 percent), averaging +$58,000 each; 19 were losers, averaging -$14,700 each.

Expectancy=(0.37×58,000)(0.63×14,700)=21,4609,261=12,199\text{Expectancy} = (0.37 \times 58{,}000) - (0.63 \times 14{,}700) = 21{,}460 - 9{,}261 = 12{,}199

About $12,200 of positive expectancy per trade, consistent with the $400,000 total over roughly 30 trades once you allow for size variation. The picture: this is a low-hit-rate strategy that works because winners run much larger than losers — closer to a trend-following profile than a mean-reversion one. That has a real implication: this strategy will have long stretches, plausibly several weeks, where nearly every trade loses a little, and that's not a sign anything is broken, it's the normal texture of a strategy that makes its money on the tail of a distribution rather than on being right most of the time.

A team that didn't look trade-by-trade and only watched the monthly total might panic three weeks into a losing streak and cut the strategy right before the next big winner arrived.

30 trades, in order
Nineteen small losers below the line, eleven winners above it — most of them modest, a handful large enough to carry the whole month.

What the review changes

Once you know the shape, you manage differently. A low-hit-rate, big-winner strategy needs discipline to hold winners and cut losers fast — cutting winners early destroys the very thing that makes the expectancy positive. A high-hit-rate, small-edge strategy needs the opposite worry: a single outsized loss can erase months of small wins, so tail risk on individual trades matters far more than hit rate.

Hit rate and average win/loss size only mean something together. Review both, trade by trade, before drawing any conclusion from the monthly total — the same total P&L can come from opposite strategies.

Don't let a string of small losses inside a low-hit-rate strategy trigger a size cut or a strategy review on its own — that pattern is expected, not a warning sign, as long as expectancy per trade is still positive.

Related concepts

Practice in interviews

Further reading

  • Van Tharp, Trade Your Way to Financial Freedom (ch. 9)
ShareTwitterLinkedIn