Skip to content

Trading metrics

Win Rate in Trading: Formula, Break-Even Point and Limits

Win rate in trading is the number of winning trades divided by the number of closed trades, shown as a percentage: 45 wins from 100 closed trades is a 45% win rate. TradingView reports it as Percent profitable and MetaTrader 5 as Profit Trades (% of total).[1][2] Win rate measures how often trades win, not how much they win or lose. Before costs and with no breakevens, the win rate needed to break even is 1 ÷ (1 + payoff ratio).

A closer target or a wider stop can move the win rate without changing what a rule earns on average. Read win rate next to average win, average loss and expectancy; the trading metrics library covers each of them.

Illustrative cover for win rate in trading: a ring split into wins, losses and breakevens beside small wins and larger losses, with the official PineConnector logo.
Win rate counts how often trades win. The sizes of the wins and losses decide what that count is worth.

Win rate at a glance

  • Definition: winning trades as a share of all closed trades. Open positions are excluded.
  • Formula: win rate = winning trades ÷ closed trades × 100%.
  • Breakeven trades: TradingView keeps them in the total but not in the wins. State the convention you use.
  • Loss rate: losing trades ÷ closed trades. It equals 1 − win rate only when there are no breakevens.
  • Where to find it: Percent profitable in TradingView's strategy report; Profit Trades (% of total) in the MetaTrader 5 Strategy Tester report; win rate in PineConnector Portal Analytics for connected MT5 accounts.
  • Break-even win rate: 1 ÷ (1 + payoff ratio), before costs, with wins counted against wins plus losses.

How do you calculate win rate?

Win rate (%) = winning trades ÷ all closed trades × 100

Each term needs a definition before two reports can be compared:

  • Winning trade: a closed trade whose final result is above zero. TradingView counts only trades with a final profit greater than zero as winners.[3]
  • Closed trades: every trade opened and closed in the period. TradingView's Percent profitable considers closed trades only and leaves open positions out.[1]
  • Breakeven (even) trade: TradingView defines an even trade as one whose profit is exactly zero after commissions and slippage.[4] Percent profitable excludes even trades from the winning count but keeps them in the total.[1]

This page uses TradingView's convention. Some traders divide by wins plus losses instead, which leaves breakevens out entirely. Both forms are legitimate, but they give different numbers from the same trade list, so name the convention whenever you quote a win rate. MetaTrader 5's report help does not say how a deal with a result of exactly zero is classified.[2]

Worked example: wins, losses and breakevens

Illustrative example, not a recommendation or a strategy result. A trade list has 100 closed trades: 45 wins, 50 losses and 5 trades that closed at exactly zero after costs. The average win is 200 USD and the average loss is 100 USD, both after costs.

  1. Win rate: 45 ÷ 100 = 45%.
  2. Win rate excluding breakevens: 45 ÷ 95 = 47.4%, rounded.
  3. Loss rate: 50 ÷ 100 = 50%. Win rate plus loss rate is 95%; the other 5% are breakevens.
  4. Gross profit and gross loss: 45 × 200 = 9,000 USD and 50 × 100 = 5,000 USD, so net profit is 4,000 USD.
  5. Expectancy per trade: (0.45 × 200) − (0.50 × 100) = 90 − 50 = 40 USD, the same as 4,000 ÷ 100. TradingView reports this average as Expected payoff: total net P&L divided by the number of closed trades.[5]
  6. Payoff ratio and break-even: 200 ÷ 100 = 2.0, so the break-even win rate is 1 ÷ (1 + 2) = 33.3%, with costs already inside these averages. Breakevens add nothing to profit or loss, so compare that threshold with wins as a share of wins plus losses: 47.4%.

Now keep the same 45 wins, 50 losses and 5 breakevens, but swap the sizes: an average win of 100 USD and an average loss of 200 USD. The win rate is still 45%. Expectancy becomes (0.45 × 100) − (0.50 × 200) = 45 − 100 = −55 USD per trade. A count of wins alone cannot tell these two trade lists apart.

Win rate vs risk-reward: how do the two combine?

Expectancy joins them: E = (W × AW) − (L × AL), where W is the win rate, L the loss rate, AW the average win and AL the average loss. The payoff ratio is AW ÷ AL, which TradingView reports as Average profit / average loss.[6] Setting E to zero with no breakevens gives the break-even win rate.

Break-even win rate = 1 ÷ (1 + payoff ratio) = AL ÷ (AW + AL), before costs

Enter your average win, average loss and cost per trade in the break-even win rate calculator to find the threshold for your trades.

PineConnector's Analytics guide gives the same relationship for a simplified case in which every win is the same size and every loss is the same size: 50% at 1:1 and 33.3% at 2:1, before costs.[7] Costs move the threshold up. With a 1R stop, a target of T R and a cost of c R per trade, the break-even win rate becomes (1 + c) ÷ (T + 1). The cost column below assumes that 1R stop, so the payoff equals the target in R before costs.

Scroll horizontally to read every column.

Payoff (average win ÷ average loss) Break-even win rate, before costs With a 0.1R cost per trade
0.5 66.7% 73.3%
1.0 50.0% 55.0%
2.0 33.3% 36.7%
3.0 25.0% 27.5%
Illustrative chart of the break-even win rate against the payoff ratio, before costs and with a 0.1R cost per trade, with the official PineConnector logo.
Illustrative arithmetic, not a target. A win rate above the blue line is above break-even for that payoff before costs; the orange line shows how a 0.1R cost per trade moves the threshold.

The three illustrative trade sets below have no breakevens, so each loss rate is 1 − win rate. They show why a win rate needs its payoff beside it.

  • Set A: 80% win rate, 50 USD average win, 200 USD average loss (payoff 0.25). Break-even win rate 80%; expectancy (0.80 × 50) − (0.20 × 200) = 0 USD per trade.
  • Set B: 50% win rate, 150 USD average win, 100 USD average loss (payoff 1.5). Break-even win rate 40%; expectancy (0.50 × 150) − (0.50 × 100) = 25 USD per trade.
  • Set C: 30% win rate, 300 USD average win, 50 USD average loss (payoff 6). Break-even win rate 14.3%; expectancy (0.30 × 300) − (0.70 × 50) = 55 USD per trade.

Set A has the highest win rate and zero expectancy, because its win rate sits exactly on its break-even point. Set C has the lowest win rate and the largest expectancy in this arithmetic. The break-even win rate guide derives the thresholds, and the payoff ratio page covers the other half of the pair. Planned risk-reward can differ from the payoff measured afterwards, as how to calculate risk-reward ratio explains.

How do exits change win rate without changing the edge?

Exit placement decides what counts as a win, so exit placement moves the win rate directly. A simple probability model shows how far that can go when there is no edge at all.

Take a symmetric random walk: price moves one equal step up or down, each with probability 0.5, with no drift, gaps or costs. Place the stop S steps below entry and the target T steps above. The classic gambler's ruin result gives the probability of reaching the target first.[8]

Share of trades reaching the target first = S ÷ (S + T)

With the stop fixed at 1R, a 0.5R target wins 66.7% of the time and a 3R target wins 25% of the time. Expectancy is zero in every case. For the 0.5R target, (2/3 × 0.5R) − (1/3 × 1R) = 0. Each layout's win rate lands exactly on its own break-even win rate. Add a cost of 0.1R per trade and every layout expects −0.1R.

Illustrative bars of model win rates of 66.7%, 50%, 33.3% and 25% for targets from 0.5R to 3R, each with zero expectancy before costs, with the official PineConnector logo.
Idealised no-edge model, not a forecast or a recommended exit. Closer targets produce more, smaller wins; the expected result per trade stays at zero before costs and at minus the cost after them.

Real prices are not a fair random walk, and gaps and costs change the exact figures. The lesson survives: a closer target, a wider stop or a breakeven stop changes the mix of wins and losses. Only a change in the average result per trade shows a change in edge. Adjusting exits until a backtest shows a high win rate risks overfitting to that sample.

Where do TradingView, MT5 and PineConnector report win rate?

As of 25 September 2026, each platform reports win rate under its own name and counting rules.

  • TradingView: the strategy report, formerly called the Strategy Tester, shows Percent profitable.[9][1] In Pine Script v6, the strategy.wintrades, strategy.losstrades, strategy.eventrades and strategy.closedtrades variables hold the counts. TradingView's own example divides strategy.wintrades by strategy.closedtrades to display the win percentage.[10] The strategy report metrics guide maps the other fields.
  • MetaTrader 5: MetaQuotes defines Profit Trades (% of total) as the number of profitable trades and their percentage of Total Trades, which counts deals that fixed a profit or loss. The Strategy Tester report also shows Loss Trades (% of total), Short Trades (won %), Long Trades (won %) and the longest runs of consecutive wins and losses.[2] The MT5 report guide explains each field.
  • PineConnector: Analytics in the PineConnector Portal shows a win rate for connected MT5 accounts, described as winning trades as a proportion of the trades counted. The published Analytics guide requires MT5 EA v3.41 or later and states that Analytics is not available for MT4.[7] How Analytics counts breakevens and partial closes is not documented; test on demo and compare it with the account's MT5 history.

How do alert stops and targets shape the win rate measured in MT5?

In an automated workflow, the idea becomes a measurable rule, the rule becomes a TradingView alert and the alert becomes an MT5 order. When the rule's stop and target travel in the alert message as a broker-side stop loss and take profit, they are the exits that decide which trades count as wins in the MT5 account.

Verify any message on a demo account before relying on it; PineConnector's setup test follows one demo order from TradingView to the broker. Choose your own symbol, size and rules; the example below is not an instruction to place a live trade.

Illustrative message, not a recommendation. A 20-pip stop and a 40-pip target give a planned risk-reward of 2. If every trade ends at one of those two levels, the break-even win rate is 33.3% before costs. In the Message field of a TradingView alert, the PineConnector message could read:

LicenseID,buy,EURUSD,vol_lots=0.01,sl_pips=20,tp_pips=40

At an illustrative entry of 1.10000, PineConnector's syntax guide shows a requested stop loss at 1.09800 and take profit at 1.10400, levels the broker must still accept.[11] A stop that exists only inside the Pine strategy closes the MT5 position only if an exit alert triggers and is delivered, processed and filled.

The MT5 account's win rate counts broker results, not chart results. MT5 executes buys at the Ask price and sells at the Bid.[12] TradingView and MetaTrader can also use different data feeds, quotes, spreads and timestamps.[13] A trade that closes a fraction above zero in the strategy report can close below zero in the account.

Breakeven stops show the counting problem clearly. The parameters betrigger=10,beoffset=2 request a stop 2 pips above a buy entry after a 10-pip favourable move, and the guide notes that a breakeven level does not ensure zero net loss after costs.[11] Costs and execution differences then decide whether such a trade counts as a small win, an even trade or a small loss. The exit automation guide covers these parameters.

The alert condition becoming true, the alert triggering, webhook delivery, PineConnector processing, the EA's order request, and broker acceptance and the fill are separate events. A trade that fails at any step stays in the strategy report but never becomes a deal in the MT5 account history. PineConnector's setup test follows one demo order through that chain, from TradingView to the broker.[14] Compare the trade lists line by line before comparing percentages; why TradingView and MT5 results differ covers the usual causes.

What is a good win rate?

There is no universal good win rate. A win rate is only good or bad relative to its payoff, its costs and the number of trades behind it. Three reference points are commonly used, each with a caveat.

  • The break-even win rate for the measured payoff. PineConnector's Analytics guide uses 1 ÷ (1 + reward-to-risk) as the breakeven win rate before costs.[7] With the average win and loss both measured after costs, a win rate counted as wins ÷ (wins + losses) above 1 ÷ (1 + payoff) means that sample had positive expectancy. Caveat: average win and loss sizes shift from one sample to the next, so clearing the threshold in one trade list says nothing certain about the next.
  • The sampling margin. The standard error of a win rate W over N trades is √(W × (1 − W) ÷ N), and a 95% margin is about 1.96 standard errors.[15] At a 50% win rate, 100 trades give about ±9.8 percentage points and 400 trades about ±4.9. Caveat: NIST's handbook uses the Wilson interval instead, which Agresti and Coull recommended, and notes that its lower limit cannot be negative.[15]
  • The shape of the losses. TradingView's help article warns that a 90% win rate can still lose money if the few losses are large.[1] Caveat: a short sample may not contain the large loss yet; the backtest sample size guide covers how many trades a comparison needs.

Automation does not change any of these relationships. PineConnector executes the rule that produced the number; any edge, or the lack of one, belongs to the rule.

What are the common mistakes when reading win rate?

  • Reading it alone. PineConnector's Analytics guide warns that a high win rate alone does not show whether a strategy made money.[7] Pair it with expectancy and profit factor, and with maximum drawdown for how far equity fell from a peak.
  • Mixing conventions. Breakevens in or out of the total, costs included or not, and partial closes all change the figure. MT5 counts deals that fixed a profit or loss, so a position closed in parts can add more than one result.[2]
  • Ignoring streaks. A 45% win rate with no breakevens means a 55% loss rate. If trades are independent, the chance that the five trades starting at any given point are all losses is 0.55 × 0.55 × 0.55 × 0.55 × 0.55 = 0.0503, about 5.0%. See the losing streak guide.
  • Tuning exits for the number. Moving a target closer changes the win rate far more easily than it changes expectancy, as the exit model above shows.

Frequently asked questions

What is win rate in trading?

Win rate in trading is the percentage of closed trades that finish with a result above zero: 45 winning trades out of 100 closed trades is a 45% win rate. TradingView calls it Percent profitable and MetaTrader 5 calls it Profit Trades (% of total). Win rate shows how often trades win, not how much.

What is the win rate formula?

The win rate formula is winning trades divided by all closed trades, multiplied by 100, with open positions excluded. In TradingView's convention, breakeven trades stay in the total but are not wins, so 45 wins, 50 losses and 5 breakevens give 45%. Leaving breakevens out gives 45 ÷ 95, about 47.4%.

What is a good win rate?

There is no universal good win rate, because a win rate only has meaning next to the payoff ratio, average win divided by average loss. Before costs and with no breakeven trades, the break-even win rate is 1 ÷ (1 + payoff): 50% at a payoff of 1 and 33.3% at a payoff of 2. Costs move that threshold up.

Win rate vs risk-reward: which matters more?

Neither matters on its own, because expectancy per trade is (win rate × average win) − (loss rate × average loss). In illustrative arithmetic, an 80% win rate with a 50 USD average win and 200 USD average loss has zero expectancy. A 30% win rate with 300 USD and 50 USD averages has an expectancy of 55 USD per trade.

Does a breakeven trade count as a win?

A breakeven trade does not count as a win in TradingView's strategy report. TradingView treats a trade with profit of exactly zero after commissions and slippage as an even trade, kept in the total but not in the wins. MetaTrader 5's report help does not say how a zero-result deal is classified.

Reviewed 25 September 2026. Facts were checked against the linked sources on that date. Nothing in this article was tested on a trading account and no code was compiled.

Sources

  1. TradingView – Percent profitable, accessed 25 September 2026.
  2. MetaQuotes – Testing Report (MetaTrader 5 Help), accessed 25 September 2026.
  3. TradingView – Total winners, accessed 25 September 2026.
  4. TradingView – Even trades, accessed 25 September 2026.
  5. TradingView – Expected payoff, accessed 25 September 2026.
  6. TradingView – Average profit / average loss, accessed 25 September 2026.
  7. PineConnector – Trading analytics, accessed 25 September 2026.
  8. Karl Sigman, Columbia University – Gambler's Ruin Problem (lecture notes, 2009), accessed 25 September 2026.
  9. TradingView – Pine Script release notes (July 2026), accessed 25 September 2026.
  10. TradingView – Strategies (Pine Script v6 User Manual), accessed 25 September 2026.
  11. PineConnector – PineConnector Syntax: take profit and breakeven, accessed 25 September 2026.
  12. MetaQuotes – Basic Principles (MetaTrader 5 Help, Trading Operations), accessed 25 September 2026.
  13. PineConnector – Frequently asked questions: price differences, accessed 25 September 2026.
  14. PineConnector – Test your setup, accessed 25 September 2026.
  15. NIST/SEMATECH – e-Handbook of Statistical Methods: confidence intervals for proportions, accessed 25 September 2026.

PineConnector executes the instructions you send it. It does not select trades, manage money, or hold funds. Trading carries risk, and past performance of any strategy does not indicate future results.

1 comment

Back to PiCo Blog

Run your strategies on our managed infrastructure

Bridge routes TradingView alerts to MetaTrader. Edge hosts and runs MT5 for you.

Core Trial

Then $59/month unless you cancel.

$7for 7 days

  • 3 account connections
  • 1 hosted MT5
  • Unlimited alerts, routed by Bridge
Try Core for $7

Trusted by 66,000+ traders in 156 countries and territories · <1s typical latency

Your Edge MT5 runs in our cloud · nothing to install · built on Microsoft Azure