Risk-Reward, Win Rate and Expectancy: The Complete Math Guide for Funded Crypto Traders

Most traders who fail a funded account do not fail because their chart reading is bad. They fail because they never did the arithmetic. They have a vague sense that they "win more than they lose," and a vague sense that their stops are "reasonable," and that vagueness is exactly what a drawdown rule punishes.

This guide is the arithmetic. Risk-reward ratio, win rate, expectancy, profit factor, losing streaks, costs and sample size. None of it needs more than a calculator. All of it is the difference between a trader who passes an evaluation once by luck and a trader who can hold a funded account for years. It is written for crypto perps on a simulated funded account, but the numbers do not care what market you trade.

Start With R: The Unit That Makes Everything Comparable

Before any ratio makes sense, you need one unit. That unit is R. One R is the amount you risk on a single trade, measured from your entry to your stop loss. If you enter BTC at 80,000 with a stop at 79,200 and your size means that stop costs you 100 USDT, then 1R equals 100 USDT on that trade.

Every result is then expressed as a multiple of R. A trade that closes for a 250 USDT gain is a +2.5R trade. A trade that hits the stop is a -1R trade. A trade you cut early for a 40 USDT loss is -0.4R. A stop that slips and costs 130 USDT is -1.3R.

Why bother? Because dollar P&L hides the truth. A 500 USDT win on a trade where you risked 1,000 USDT is a bad trade dressed as a good one. A 50 USDT win on a 20 USDT risk is excellent. R strips out size and lets you compare a quiet Tuesday scalp with a Sunday-night swing on equal terms. On a funded account, where your risk per trade should be a fixed fraction of the account, R is also what your drawdown rule is actually counting. Nine losing trades at 1% risk each is -9R and 9% of the account. The rule does not care which coins you traded.

From here on, every number in this guide is in R.

Risk-Reward Ratio: What It Really Measures

Risk-reward ratio is the simplest of the metrics and the most abused. It is the size of your target divided by the size of your stop, measured before you enter. A 1,600-point target against an 800-point stop is a 2:1 trade, or 2R.

Three things traders get wrong about it.

First, the planned ratio is not the realised ratio. You plan 3R. Price goes to 2.2R and reverses. You take profit at 1.5R. Your realised multiple is 1.5R, and that is the number that matters. Most traders' average winner is far smaller than their planned target because they cut winners early, and most traders' average loser is larger than 1R because they widen stops or let slippage eat them. Track the realised number, not the planned one.

Second, risk-reward in isolation says nothing about profitability. A 5:1 setup that hits its target 10% of the time loses money. A 0.8:1 setup that hits 70% of the time makes money. The ratio only has meaning when it is paired with win rate. That pairing is called expectancy, and it is the only number that actually tells you whether you have an edge.

Third, the ratio is a function of where the stop sits, and the stop should sit where the trade idea is wrong, not where the ratio looks nice. Tightening a stop to turn a 1.5R trade into a 3R trade on paper does not create a better trade. It creates a trade that gets stopped out by noise and then goes to your target without you. On a 24/7 crypto perp book, where wicks through obvious levels are routine, the stop has to live outside the noise. Accept the ratio that gives you, or skip the trade.

Win Rate: Honest Definition, Honest Sample

Win rate is the percentage of your trades that close in profit. It sounds trivial to measure. Two traps make it the most misreported number in trading.

The first trap is definition. A trade closed at +0.1R after being up 2R is technically a win. A trade closed at breakeven is often logged as a win. If your "win rate" is 65% but a third of those wins are scratches, your real win rate on meaningful trades might be 45%. Define a win as a trade that closed at or above a minimum threshold, say +0.5R, and log scratches separately.

The second trap is sample size, and it is the one that destroys evaluations. Suppose your true win rate is 50%. Over 20 trades, pure chance will put your measured win rate anywhere between roughly 28% and 72% about 95% of the time. Over 100 trades, the band narrows to roughly 40% to 60%. You need around 400 trades before the band tightens to about 45% to 55%. Twenty trades tells you almost nothing. A trader who passes a challenge in fifteen trades and concludes "I have a 73% win rate" has not measured a win rate. They have measured a coin flip streak.

Practical rule: do not trust any win-rate figure built on fewer than 100 trades, and treat anything built on fewer than 50 as noise. This is one of the quieter reasons a no-time-limit evaluation matters. It lets you accumulate a real sample at normal size instead of forcing 20 oversized trades into 30 days and calling the result a system.

Expectancy: The Only Number That Says If You Have an Edge

Expectancy is the average amount you expect to make or lose per trade, in R, over a large number of trades. The formula is short:

Expectancy = (Win Rate × Average Win in R) − (Loss Rate × Average Loss in R)

If your average loss is exactly 1R, which is the goal, the formula simplifies to: Expectancy = (Win Rate × Average Win) − Loss Rate.

Three worked examples.

Trader A wins 40% of the time with an average win of 2R and an average loss of 1R. Expectancy is (0.40 × 2) − (0.60 × 1) = 0.80 − 0.60 = +0.20R. Every trade is worth a fifth of an R on average. Over 100 trades at 1% risk, that is roughly +20% before costs. Trader A loses more often than they win and is comfortably profitable.

Trader B wins 65% of the time with an average win of 0.6R and an average loss of 1.2R, because they let losers run a little. Expectancy is (0.65 × 0.6) − (0.35 × 1.2) = 0.39 − 0.42 = −0.03R. Trader B wins almost two trades in three, feels like a good trader, posts green screenshots, and slowly bleeds. This is the most common profile among traders who blow funded accounts: high win rate, negative expectancy.

Trader C wins 30% of the time with an average win of 3R and an average loss of 1R. Expectancy is (0.30 × 3) − (0.70 × 1) = 0.90 − 0.70 = +0.20R. Same edge as Trader A, but reached by a completely different route. Trader C loses seven trades in ten and must be psychologically built for it.

The lesson is that there is no "right" win rate and no "right" risk-reward. There is only the product. A positive expectancy is an edge. A negative one is a slow or fast exit from the account. Everything else, the indicators, the timeframes, the narrative, is scaffolding around this one number.

The Breakeven Table: How Often You Must Win at Each Ratio

Because expectancy is the product of two numbers, you can solve for the win rate that makes a given risk-reward exactly breakeven, ignoring costs. The formula is 1 divided by (1 + R).

• 0.5R average win: you need to win 66.7% of the time just to breakeven.
• 1R average win: 50%.
• 1.5R average win: 40%.
• 2R average win: 33.3%.
• 3R average win: 25%.
• 4R average win: 20%.

Read this table against your own journal. If your realised average winner is 1.2R and your win rate is 44%, you are losing money, and no amount of screen time will fix it until one of those two numbers moves. If your realised winner is 2R and you win 42%, you have a real edge and your only job is to not interfere with it.

The table also explains why scalping is hard on a funded account. Scalp setups tend to produce average winners below 1R after slippage. That demands win rates above 55% to 60%, sustained, on a book that wicks through stops every hour. Possible, but the margin for error is thin, and thin margins and fixed drawdown rules are a poor combination.

Costs: The Expectancy Tax Nobody Models

Every expectancy calculation above ignores costs. On a crypto perp account, costs come from three places: trading fees on entry and exit, funding payments every eight hours while a position is open, and slippage between where you wanted to fill and where you did. All three are paid out of equity, and on a funded account equity is what the drawdown rule measures.

Work a realistic example. Your stop is 2% from entry and you risk 1% of the account, which means your notional is half the account. If round-trip fees and slippage together cost around 0.1% of notional as a working figure, that is 0.05% of the account per trade, or 0.05R. Add a few funding windows on a held trade and the total cost per trade can reach 0.1R. That sounds trivial. It is not.

Trader A had an expectancy of +0.20R. Subtract 0.1R in costs and the edge is +0.10R. Costs took half the edge. A trader with an expectancy of +0.08R before costs is a losing trader after them. This is why high-frequency, tight-stop styles struggle on funded accounts: the tighter your stop, the larger your notional for the same risk, and the larger the notional, the bigger the cost in R terms. Wide stops with small size are cheaper in R than tight stops with big size.

Two practical consequences. Use limit orders where the setup allows, because maker fills and better prices directly raise your post-cost expectancy. And measure your costs in R per trade in your journal. If you do not know that number, you do not know your edge. On Bybit-powered execution, where fills come against a live order book rather than a synthetic feed, the cost you measure in evaluation is the cost you will face later, which is the point.

Profit Factor: The Second Opinion

Profit factor is gross profit divided by gross loss over a period. If your winners added up to 80R and your losers added up to 60R, your profit factor is 1.33. Below 1.0 you are losing. Between 1.0 and 1.3 you have a fragile edge that costs and a bad month can erase. Above 1.5 over a large sample is strong. Above 2.0 on more than 200 trades is rare and should be checked for a survivorship error before you believe it.

Profit factor and expectancy say the same thing in different units, but profit factor is useful because it is hard to fake. A trader can quote a win rate from a good fortnight. Gross wins divided by gross losses over six months is a blunter instrument. If your profit factor is 1.1 and your account is up, you are up on variance, not edge.

Losing Streaks: Probable, Not Possible

Here is where the math meets the drawdown rule. Every positive-expectancy system loses in streaks, and the length of those streaks is a function of win rate alone, not of skill.

At a 50% win rate, the chance of any given trade starting a run of five straight losses is about 3%. Six straight is about 1.6%. Over a few hundred trades, a six-loss streak is not a possibility, it is an expectation. At a 40% win rate, like Trader A, five straight losses happen about 8% of the time, eight straight about 1.7%. Over 300 trades, you should expect to see an eight-loss streak roughly twice. Trader C, at 30%, should plan for streaks of ten or more as a normal feature of the system.

Now connect that to a funded account. Suppose your maximum drawdown is 9% and you keep a 2% safety margin because stops slip and funding accrues. Your usable budget is 7%. If you risk 1% per trade, an eight-loss streak that the math says is coming puts you 1% from breach with a 40% system, and a single slipped stop finishes it. At 0.7% per trade, the same streak costs 5.6% and leaves room. At 0.5%, it costs 4% and you barely notice.

This is the real answer to "how much should I risk per trade." Not a rule of thumb. Take your usable drawdown, divide by the longest losing streak your win rate makes probable over a few hundred trades, and that is your maximum per-trade risk. For most systems that lands between 0.5% and 1%. The full budgeting method is in our drawdown guide and the sizing mechanics in the position sizing guide.

The Target in R: How Many Trades Does Passing Actually Take?

Traders often look at a profit target as a percentage and a drawdown limit as a percentage and never translate either into trades. Do the translation.

Say the target is 10% and you risk 1% per trade. The target is +10R net. With Trader A's post-cost expectancy of +0.10R, the expected number of trades to reach +10R is about 100. With a pre-cost expectancy of +0.20R, it is about 50. Variance will make the real number anywhere from 30 to 200. If you are sitting on a 30-day deadline, 50 to 200 trades of a system designed for two or three setups a day is not going to fit, and the only lever left is size, which is exactly the lever that breaks the account on the streak you just read about.

This is the quiet math behind why challenge structures matter. A positive-expectancy trader at sane size on a no-time-limit evaluation will reach the target. The same trader on a deadline either waits for an abnormally good month or sizes up into a probable streak. The edge is the same in both cases. The structure decides which one gets to express it.

Why High Win Rate Systems Blow Funded Accounts

It deserves its own section because it is the most seductive failure mode in trading.

A trader notices that if they do not use a stop, most trades eventually come back. Their win rate climbs to 80% or 90%. The winners are small, maybe 0.5R, but there are so many of them. The equity curve is a smooth upward line for weeks. Then one trade does not come back. Price goes 8R against them, and the drawdown rule ends the account in a single afternoon. Expectancy: (0.90 × 0.5) − (0.10 × 8) = 0.45 − 0.80 = −0.35R. Deeply negative, hidden behind a beautiful win rate.

The same logic applies in softer forms. Widening a stop "just this once" turns a -1R into a -2.5R. Adding to a loser to "improve the average" doubles the loss on the trades that do not recover. Each of these raises win rate and lowers expectancy. If your win rate is climbing while your account is not, this is why.

The fix is mechanical: average loss must be capped at 1R, always, on every trade, with a stop-market order placed at entry. Then your win rate and average win can be whatever your setups honestly produce, and the expectancy number you compute will be the truth.

Building the Numbers Into a Journal

None of this is useful if it lives in your head. Every trade needs seven fields: date and time in UTC, pair, planned R-multiple at entry, realised R-multiple at exit, cost in R, setup type and a one-line note on whether you followed the plan. From those seven fields you can compute win rate, average win, average loss, expectancy, profit factor and longest losing streak, by setup type, in a spreadsheet, every weekend.

Review by setup, not just overall. Most traders find that one or two setups carry all the expectancy and the rest are negative. Cutting the negative setups is the single fastest way to improve an account, and you cannot find them without the numbers. The trading plan guide covers how this review loop fits into a weekly routine.

And review your discipline rate, meaning the share of trades where you followed the plan. If your expectancy on plan-following trades is +0.25R and on improvised trades is −0.4R, you do not have a strategy problem. You have a behaviour problem, and the math has just shown you precisely what it costs.

Five Mistakes With the Math

Quoting planned risk-reward instead of realised. The plan was 3R; the journal says 1.4R. Only the journal counts.

Trusting a win rate from 20 trades. You have measured variance, not skill. Keep trading at small size until the sample is real.

Ignoring costs. Fees, funding and slippage are paid in R and can take half a thin edge. Measure them.

Sizing for the average and not the streak. The streak your win rate makes probable is the one that meets the drawdown rule. Size for that.

Letting average loss drift above 1R. One uncapped loser can erase forty disciplined trades. Stop-market at entry, every time.

How This Fits a FundedXYZ Account

FundedXYZ evaluations are single-phase with no time limit, a static maximum drawdown and no daily loss cap, on simulated capital with Bybit-powered execution. In the language of this guide, that structure removes the two things that most distort a trader's math: the deadline that forces size, and the daily cap that turns a normal two-loss morning into a rule event. What is left is your expectancy, your streak length and your sizing, which is exactly what should decide whether a trader is funded.

Nothing here guarantees a result. A positive expectancy is a statistical edge over many trades, not a promise about the next one, and a trader with a real edge can still breach through poor sizing or poor discipline. But a trader who knows their numbers knows where the edge is, how much variance it carries and what size survives it. That is the whole game.

If you want to run your numbers on a live order book without risking your own capital, an evaluation starts from $20 and the clock is not running. Start when the math says you are ready, not when a deadline says you must.