Trading Expectancy and Risk of Ruin: The Math Every Chart Trader Ignores

You know your win rate from memory. You cannot state your expectancy. You have never estimated risk of ruin.

So you size up after a green week and size emotionally after a red one. The math did not change. Your mood did. The account still obeys arithmetic.

Trading expectancy answers: "How much do I make per trade on average?" Risk of ruin answers: "How likely is this risk level to wipe me out if my edge is real (or fake)?"

Chart traders ignore both because spreadsheets feel boring. This guide makes them usable: formulas in plain English, worked examples with round numbers, when to reduce size, and links to the 1% risk position sizing workflow you should run before every entry.

For structured entries, stops, and targets from screenshots, see the pillar guide on how to analyze a trading chart screenshot with AI.

Who this is for

This fits you if:

• You trade discretionary setups and track wins/losses but not edge per trade.
• You use a fixed percent risk rule (or want to) and need to know if it is safe.
• You blew up or came close and suspect size, not just psychology.
• You finished a manual backtest and have win rate and average R data ready.

If you refuse to log results in R (risk units), read the trading journal with screenshots first. Expectancy without honest logs is fiction.

What trading expectancy means (no jargon wall)

Expectancy is the average amount you expect to make or lose per trade over a long series, if your past statistics stay similar.

Think of each trade as a weighted coin:

• Sometimes you lose about 1R (one unit of planned risk)
• Sometimes you win +1.5R, +2R, or more
• Sometimes you scratch at 0R

Expectancy combines how often you win with how big wins are versus losses.

If expectancy is positive, the system pays you over many trades. If negative, you are slowly donating to the market even with a "decent" win rate.

Important: expectancy describes the plan executed, not your best daydream. Slippage, early exits, and moved stops belong in the numbers you feed the formula.

Expectancy formula in plain English

Let:

• W = win rate (decimal). 50% = 0.50
• L = loss rate = 1 minus W
• AvgWin = average winner in R (risk units)
• AvgLoss = average loser in R (usually 1 if stops are consistent)

Expectancy (in R per trade) = (W × AvgWin) minus (L × AvgLoss)

That is it. No calculus. You are averaging the R outcomes.

Worked example A: balanced trend trader

Suppose after 100 logged trades:

• Win rate 45% → W = 0.45, L = 0.55
• Average winner +2.0R
• Average loser -1.0R (you honor stops)

Expectancy = (0.45 × 2.0) minus (0.55 × 1.0)
= 0.90 minus 0.55
= +0.35R per trade

On a $50,000 account risking 1% per R ($500 per R), expected value per trade is about $175 before costs. Over 200 trades, that is rough long-run drift if statistics hold (they wobble in real life).

Worked example B: high win rate, small winners

Many scalpers live here:

• Win rate 62% → W = 0.62, L = 0.38
• Average winner +0.8R
• Average loser -1.0R

Expectancy = (0.62 × 0.8) minus (0.38 × 1.0)
= 0.496 minus 0.38
= +0.116R per trade

Positive, but thin. A small rise in losers or a slip in average win erases the edge. Size should stay modest. Costs matter more.

Worked example C: looks good, is not

• Win rate 55%
• Average winner +1.0R
• Average loser -1.2R (you often give back extra before stop)

Expectancy = (0.55 × 1.0) minus (0.45 × 1.2)
= 0.55 minus 0.54
= +0.01R per trade

Technically positive. Practically noise. One regime shift and you are negative. This is where traders say "I win more than I lose" while the account flatlines.

From R expectancy to dollars (bridge to position sizing)

Expectancy in R is portable. Dollars depend on risk per R.

Expected dollars per trade ≈ Expectancy (R) × Risk dollars per R

If you risk 1% of $40,000 ($400) per trade and expectancy is +0.35R:

Expected dollars ≈ 0.35 × $400 = $140 per trade (long-run average, not a guarantee)

Position sizing sets Risk dollars per R. Expectancy tells you if repeating that risk is rational. Use the full calculator workflow in position size from stop loss on your chart so R stays consistent trade to trade.

Risk of ruin: what it measures

Risk of ruin is the approximate probability that a series of losses (given your win rate and risk per trade) will draw your account down to zero or below your pain threshold before you recover.

You do not need exact perfection. You need directional truth:

• Bigger risk per trade → ruin risk rises fast
• Lower win rate or worse payoff ratio → ruin risk rises
• Positive expectancy → ruin risk lower, but never zero if size is reckless

Chart traders should treat ruin math as a speed limit, not a prophecy. Markets change. Expectancy wobbles. Ruin formulas assume your inputs are stable enough to matter.

A usable risk of ruin approximation

Exact formulas vary. This version is common in trader education and good enough for decisions:

Let:

• E = edge per trade in "units" where one unit = one full risk loss (here, use expectancy in R if positive; if negative, ruin is eventual unless you stop trading)
• U = maximum drawdown you will tolerate, in the same units (e.g. "I will stop at 20R drawdown" → U = 20)

For a simplified intuition when risking fixed fractional f of account per trade (as percent risk), ruin risk rises sharply as f grows.

Practical shortcut many desks use:

1. Compute expectancy in R (above).
2. Estimate your largest realistic losing streak in trades (from journal or backtest).
3. Multiply streak by risk percent per trade. If that product threatens your max drawdown cap, size is too large.

Worked example D: losing streak vs account

Account: $30,000
Risk: 1% per trade ($300)
Expectancy: +0.3R (healthy)
Historical worst streak: 8 losses in a row (from journal)

8 × $300 = $2,400 drawdown ≈ 8% of account from streak alone

If you had been risking 3% per trade ($900):

8 × $900 = $7,200 ≈ 24% drawdown from streak alone

Same edge. Different survival. Ruin is often a sizing story.

Worked example E: negative expectancy

Win rate 40%, avg win +1.2R, avg loss -1.0R

Expectancy = (0.40 × 1.2) minus (0.60 × 1.0) = 0.48 minus 0.60 = -0.12R

No risk-of-ruin formula saves you long term. You will grind down. Fix the playbook or stop trading live size.

Win rate vs payoff ratio (the tradeoff chart traders miss)

Define payoff ratio (reward-to-risk on winners):

Payoff ratio = AvgWin ÷ AvgLoss (in R)

Break-even win rate needed:

Breakeven W = AvgLoss ÷ (AvgWin + AvgLoss)

Round numbers: if average win is +2R and average loss is 1R:

Breakeven W = 1 ÷ (2 + 1) = 33.3%

You can lose two-thirds of trades and still break even if winners truly average +2R and losses stay at -1R. Most traders never verify averages. They remember the +3R hero.

Run this after every weekly journal review. One month of +2R winners and -1.3R losers changes everything.

When to reduce position size (clear rules)

Cut risk before you negotiate with willpower. Use these triggers:

1. Live expectancy turns negative over last 30 trades
Not one bad week. Thirty plan-following trades with negative average R. Drop from 1% to 0.5% or paper trade until forward test recovers.

2. Drawdown hits your pre-written cap
Example cap: 10R or 6% of account, whichever you defined in writing. At cap, halve risk or stop for the week.

3. Payoff ratio collapses while win rate looks fine
You still "win often" but average winner shrinks (early exits, fear). Investigate behavior and structure tags from market structure misreads before you size back up.

4. Correlated exposure stacks
Three USD-long equity trades are one macro bet. Either count them as one R bucket or cut per-trade risk to 0.33% each. See sizing notes in the 1% risk guide.

5. Regime change without data
Volatility doubles, your stops widen, same dollar risk means smaller size in shares but larger swing in account terms. Recompute risk dollars after volatility shifts.

6. Forward test after backtest fails
If holdout backtest expectancy is negative, live size is not a psychology problem. It is a no-edge problem.

When to increase size (slowly)
Only after 30 to 50 forward trades at reduced risk show positive expectancy and you followed stops. Increase by 25% steps (0.5% → 0.625% → 0.75%), not doubles.

Expectancy, ruin, and your chart workflow

Math does not replace screenshots. It judges them.

1. Capture pre-trade screenshot with entry, stop, target.
2. Size so loss ≈ 1R using structure stop (workflow).
3. Log result in R.
4. Monthly: recompute W, AvgWin, AvgLoss, expectancy.
5. Quarterly: stress-test worst streak against risk percent.

AI chart tools (including Bullsights) help draft plans faster. They do not compute your long-run expectancy. Your journal does.

Quick spreadsheet layout (copy this)

At the bottom:

• W = COUNTIF(C,">0") / COUNT(C) for plan-following rows only
• AvgWin = AVERAGEIF(C,">0")
• AvgLoss = ABS(AVERAGEIF(C,"<0"))
• Expectancy = W*AvgWin - (1-W)*AvgLoss

You can build the same logic in Notion or on paper. The point is repeatability, not software.

Rolling windows: do not trade on all-time hero stats

Split expectancy into last 30 and last 100 trades. If the short window goes negative while the long window stays positive, you are in a rough patch. Reduce size. If both windows are negative, the playbook or your execution broke. Fix before you fund ego with account equity.

This pairs with pre-market planning: write the day’s risk cap first, then let rolling math tell you if the cap should be tighter this month.

Common mistakes

Mistake: using dollar P and L only.
Dollars mix sizing luck with edge. Fix: R multiples.

Mistake: counting plan violations as system losses.
That is a discipline problem, not expectancy. Tag separately or ruin math lies.

Mistake: ignoring partial wins.
+0.5R and +2R average together. Be honest.

Mistake: assuming today's edge forever.
Recalculate rolling 30 and 100 trade windows.

Mistake: 1% risk with negative expectancy.
You will lose faster. Fix edge or stop.

Mistake: no daily loss cap.
Caps limit ruin speed on tilt days. Three -1R stops and done is a valid rule.

Pre-trade and monthly math checklist

Before increasing size

• Last 30 trades logged in R with setup tags
• Expectancy positive on last 30 and last 100 (or you stay small)
• Worst streak × risk% fits drawdown cap
• Payoff ratio and breakeven win rate computed
• Correlated positions counted as one bucket
• Stop placement still structure-based on screenshot

Monthly review (15 minutes)

• Update W, AvgWin, AvgLoss, expectancy
• Note regime tags (trend vs chop weeks)
• Compare live vs backtest expectancy drift
• Adjust risk percent only by pre-written rules

FAQ

What is a "good" trading expectancy?

For discretionary intraday traders, +0.2R to +0.5R per trade after costs is a strong zone if sample size is 100+ trades. Below +0.1R is fragile. Negative is stop-live-size territory.

How many trades do I need before expectancy is reliable?

Directionally useful at 50+ plan-following trades. Comfortable near 100+. Below 30, treat numbers as hints only.

Does high win rate mean low ruin risk?

No. High win rate with tiny winners can have worse ruin profile than moderate win rate with +2R winners. Use full expectancy, not win rate alone.

Should I use Kelly criterion?

Full Kelly is aggressive for discretionary traders with unstable inputs. If you explore Kelly, use fractional Kelly (quarter Kelly or less) and compare to simple 0.5% to 1% rules. Most chart traders do better with fixed fractional risk and strict caps.

How does risk of ruin relate to prop firm drawdown rules?

Prop rules are hard ceilings. Set personal caps inside firm caps (e.g. stop at 4% if firm allows 6%). Streak math still applies.

Can Bullsights calculate expectancy for me?

Bullsights produces structured trade plans from screenshots (entries, stops, targets, scenarios, macro context). It does not replace your trade log or expectancy math. Export R results from your journal monthly.

Bottom line

Trading expectancy tells you if the playbook pays per trade. Risk of ruin tells you if your size lets you survive the losing streaks that always show up.

Compute expectancy in R: (win rate × average win) minus (loss rate × average loss). Stress-test streaks against your risk percent. Cut size on rules, not feelings. Link sizing to chart stops with the 1% risk calculator workflow.

When you want cleaner plans from the screenshots you already log, try Bullsights. Structure the trade first. Let the math decide if you deserve full size.