Trading Expectancy: Calculate Your Real Edge

Most traders focus on win rate. It feels intuitive — winning more often than you lose should mean you’re profitable. But win rate tells you nothing without knowing how much you win versus how much you lose, and that relationship is what trading expectancy measures. According to a 2000 study by Brad Barber and Terrance Odean in the Journal of Finance, 80% of active day traders lose money over a two-year period — most of them trading systems with negative expectancy without ever realizing it.

Why Win Rate Alone Is Meaningless

Consider two traders. Trader A wins 65% of their trades, taking $100 profits and $250 losses. Trader B wins only 40% of the time, taking $200 profits and $100 losses.

Most people would bet on Trader A. They’d lose that bet.

Applying the expectancy formula — E = (Win% × Avg Win) − (Loss% × Avg Loss) — the math is unambiguous:

Trader A: (0.65 × $100) − (0.35 × $250) = $65 − $87.50 = −$22.50 per trade
Trader B: (0.40 × $200) − (0.60 × $100) = $80 − $60 = +$20.00 per trade

Trader A loses $22.50 on average every time they click the button, despite winning 2 out of 3 trades. Trader B profits $20 per trade while losing more often than not. This counterintuitive reality is why expectancy — not win rate — is the metric that actually tells you if a strategy works.

A system with a 45% win rate, $300 average win, and $150 average loss has expectancy of +$52.50 per trade. That system is viable. A system with a 65% win rate, $100 average win, and $250 average loss has expectancy of −$22.50 per trade. That system is a slow bleed.

How to Calculate Expectancy from Your Journal

You need four numbers from your trade log: win count, average winning trade, loss count, average losing trade. Pull them from any period with at least 50–100 trades.

Here’s a worked example using 50 trades:

22 winners, averaging $312 each
28 losers, averaging $198 each
Win rate: 22/50 = 44%
Loss rate: 28/50 = 56%

Expectancy = (0.44 × $312) − (0.56 × $198) = $137.28 − $110.88 = +$26.40 per trade

At 3 trades per day across 20 trading days, that’s 60 trades per month and approximately $1,584 in expected monthly profit on a $25K account — a 6.3% monthly edge. That’s the number that matters when deciding whether to continue trading a strategy or scale it up.

The formula is simple. What most traders skip is doing this calculation at all, let alone doing it regularly across a meaningful sample size. Expectancy from 20 trades is noise. A handful of outsized wins can make a losing system look profitable, and a bad week can make a solid system look broken. The calculation only becomes reliable past 50 trades and statistically stable past 100.

Gross vs. Net Expectancy: Where Marginal Systems Die

Gross expectancy ignores trading costs. Net expectancy is what you actually keep.

For a concrete example: a trader runs 60 SPY trades over one month — 27 winners averaging $285 and 33 losers averaging $175. Gross expectancy = (0.45 × $285) − (0.55 × $175) = $128.25 − $96.25 = +$32 per trade. That looks viable.

But after accounting for $8 per trade in commissions and slippage (realistic for active retail trading on equities or ES futures, where round-trip costs typically run $4–8 on futures and $2–10 on equities depending on broker and share size), net expectancy drops to $24 per trade. Projected forward: 60 trades × $24 = $1,440 expected monthly profit on a $30K account — a solid 4.8% monthly edge.

Now change the scenario slightly. If gross expectancy were +$10 instead of +$32, $8 in costs would reduce net expectancy to $2 per trade. At that level, a single bad week in execution — slightly worse fills, a few extra commissions — erases the edge entirely. This is why the minimum viable threshold is net expectancy above 1.5× your average cost per trade, not just above zero.

Benchmark thresholds to evaluate your system:

Negative expectancy: Losing system. Stop trading it.
$0 to 1.5× costs: Marginal. Costs will frequently eat the edge in practice.
1.5× to 3× costs: Healthy. The system has room to absorb variance and execution drag.
Above 3× costs: Excellent. The edge is durable across market conditions.

For more on tracking these metrics, see net expectancy per trade in the JournalPlus metrics library.

Expectancy as a Strategy Segmentation Tool

Overall expectancy is useful. Segmented expectancy is where the real insight lives.

Returning to the SPY trader from the example above: overall net expectancy is +$24. But when they break it down by session, the picture changes:

Regular-hours trades: +$41 net expectancy per trade
Pre-market trades: −$19 net expectancy per trade

The pre-market trades — likely wider spreads, thinner liquidity, higher slippage — are destroying edge built during regular hours. Cutting pre-market entirely raises overall net expectancy from $24 to $41 per trade. That’s a 71% improvement without changing a single entry signal.

This is why expectancy should be calculated across at least three dimensions:

Setup type — Does your breakout setup have positive expectancy? Your mean-reversion setup? A strategy that works on one pattern may have negative expectancy on another.
Time of day — Open-range trades, midday chop, and close setups often behave differently. Segment them.
Instrument — If you trade SPY, QQQ, and individual stocks, calculate expectancy per instrument. One of them may be dragging the overall number negative.

A trader with +$18 overall expectancy who discovers their tech stock trades have −$45 expectancy per trade has actionable information. Removing those trades doesn’t require changing the strategy — it requires reading the data.

This kind of segmentation is detailed in the guide on how to build a trading edge and in trading journal data analysis.

Using Expectancy to Make Sizing Decisions

Once you have a reliable expectancy figure — based on at least 50–100 trades — it becomes the decision framework for position sizing and strategy management.

The logic is straightforward:

Positive expectancy, large sample: The system works. Scale position size within your risk limits.
Positive expectancy, small sample (under 50 trades): Too early to conclude. Trade at minimum size until the sample grows.
Negative expectancy, any sample: Stop trading the strategy. No amount of position sizing saves a negative-expectancy system.
Marginal expectancy (near zero net): Reduce size and focus on cost reduction — better execution, lower-commission broker, tighter entries — before scaling.

For day traders, this also connects directly to time of day trading analysis, where expectancy often varies significantly by session. For risk-based sizing frameworks, position sizing journal guide covers the mechanics in detail.

The underlying principle: expectancy tells you the expected value of each trade. Combined with your trade frequency, it projects monthly expected profit. If that projection doesn’t justify the risk and time involved, the system isn’t ready to trade at size — and no amount of confidence in your setup changes that math.

Key Takeaways

Expectancy = (Win% × Avg Win) − (Loss% × Avg Loss). This single number tells you more about a system than win rate alone ever could.
A 40% win rate system with 2:1 reward-to-risk outperforms a 65% win rate system with 0.5:1 R:R — the math is unambiguous.
Always use net expectancy (after commissions and slippage). A gross-positive, net-negative system is a losing system.
Segment expectancy by setup type, time of day, and instrument. Overall positive expectancy can hide sub-strategies with deeply negative expectancy that drag down the whole account.
Minimum sample for a reliable expectancy calculation is 50 trades. Under that threshold, the number is noise.

JournalPlus automatically calculates net expectancy per trade after costs — the metric most journal tools omit — and lets you segment it by setup, session, and instrument directly from your trade log. If you’re trading without knowing your real edge, JournalPlus gives you that number for a one-time $159.