A strategy with a monster edge that only triggers twice a year will lose a payout race to a modest one that fires every session. Per-trade expectancy is the number everyone quotes, and on its own it’s misleading — because it says nothing about how often you actually get to use it.
Per-Trade Expectancy Ignores Frequency
Expectancy is the average result you expect from a single trade, usually expressed in R (multiples of the amount you risk). A system with 40% winners at +2R and 60% losers at −1R has an expectancy of (0.40 × 2) + (0.60 × −1) = +0.20R per trade. Solid.
But +0.20R tells you nothing about your account’s trajectory. Two traders can both run a +0.20R system and end the year in completely different places, because one took 30 trades and the other took 600. If you need a refresher on the underlying number before scaling it, the trading expectancy primer walks through the formula and the common ways it’s miscalculated. And if you’re not yet expressing results in R, start with R-multiple — the rest of this article assumes normalized units.
Expectancy × Trades = Expectancy Per Day
The fix is one multiplication:
Expectancy per day = per-trade expectancy × trades per day
That +0.20R edge at 5 trades a day is +1.0R/day. The same edge at 0.2 trades a day (once a week) is +0.04R/day — twenty-five times slower to compound, twenty-five times slower to clear a profit target. Same edge, radically different businesses.
You can roll the same logic up to whatever horizon matters for your firm:
- Per week — expectancy × trades per week. Good for swing systems.
- Per month — the horizon most prop targets are actually measured over.
Compute per-trade expectancy from your real log with an expectancy calculator, then multiply by your genuine trade frequency — not your best week, your median week.
Comparing a Rare, Fat Edge to a Frequent, Thin One
Here’s where the reframe earns its keep. Consider two systems:
| System | Per-trade edge | Trades/week | Expectancy per week |
|---|---|---|---|
| A — rare, fat | +0.80R | 2 | +1.6R |
| B — frequent, thin | +0.15R | 12 | +1.8R |
System A looks four times better on the headline number. But B produces more R per week because it fires six times as often. On throughput, the “worse” strategy wins. This is the single most common mistake in strategy selection: chasing the highest per-trade number instead of the highest per-unit-time number.
Opportunity Cost and Capital Efficiency
Frequency also decides how hard your capital works. A prop account tied up in a strategy that trades twice a month is idle most of the time — that idle capital has an opportunity cost equal to whatever else it could have been running.
For funded traders this compounds two ways:
- Time to target — a frequent thin edge clears a profit target sooner in calendar time, which means faster payouts and faster scaling.
- Cost drag — but every trade pays spread and commission, so a frequent system needs its edge to clear costs on every fire. A rare system amortizes almost no cost per unit of time but wastes the account’s availability.
The right question isn’t “which edge is bigger” — it’s “which edge produces the most R per unit of capital per unit of time, after costs.”
Frequency’s Effect on Variance and Drawdown
More trades isn’t only faster compounding — it’s also smoother. The law of large numbers works on trade count, not calendar time. A +0.20R edge over 600 trades a year has a far tighter distribution of outcomes than the same edge over 30 trades, where a single cold streak dominates the whole year.
This matters enormously on a trailing drawdown. A low-frequency system spends long stretches with a tiny sample, and its equity path is lumpy — a cluster of three losses is a meaningful chunk of the year. A high-frequency system averages out faster, so its worst drawdown as a fraction of annual profit tends to be shallower. More reps, less luck.
The caveat: frequency only reduces variance if the trades are genuinely independent. If you fire ten correlated positions on the same catalyst, you don’t have ten trades’ worth of diversification — you have one trade in ten costumes. Copying a setup across several prop accounts is the same risk wearing different names: it multiplies size, not diversification.
Choosing Systems by Throughput, Not Just Edge
Rank your strategies by expectancy per day or per week, not per trade. The reframe changes real decisions:
- Allocation — give more capital and attention to the higher-throughput edge, even if its per-trade number is smaller.
- Which to retire — a fat edge that almost never fires may not be worth the platform slot it occupies.
- Where to look for improvement — sometimes the fastest way to grow the account isn’t a better edge, it’s more clean reps of the edge you already have.
This is why a live view beats a static backtest number. Shibiki tracks edge health per strategy and pairs expectancy with a Wilson confidence interval, so a low-frequency system is honestly flagged as low-confidence until it has the sample to earn trust — and its trades are auto-journaled, so the frequency figure you multiply by is your real one, not an estimate. Throughput is what actually pays the account; measure it directly.
Related: Trading expectancy · Expectancy calculator · R-multiple