ATI Learning / Performance Metrics
Profit Factor Explained: What Is a Good Profit Factor?
Profit factor compares total winning and losing trades. The useful question is not merely whether the ratio is above one, but whether its profits, losses, costs and evidence can withstand scrutiny.
The direct answer
What does profit factor mean in trading?
Profit factor is the total gross profit from winning closed trades divided by the absolute total gross loss from losing closed trades. A profit factor of 1.70 means that the measured sample recorded $1.70 of gross winning P&L for each $1.00 of gross losing P&L.
A value above 1.0 indicates that gross profits exceeded gross losses within that sample and under that report’s accounting rules. It does not mean the account returned 70%, that only $1 was risked, or that the relationship will continue.
A good profit factor is comfortably above 1.0 after realistic costs—and remains credible when you test its concentration, sample and stability.
There is no universal number that makes an automated system good or safe. A stable, live profit factor of 1.40 across varied conditions may be stronger evidence than a backtested 3.00 produced by a short, optimized period. Read the ratio alongside maximum drawdown, current equity, trade count and the gap between backtest and live results.
Interpretation
What is a good profit factor?
For screening, 1.50 or higher is a strong headline result when it is net of realistic costs and supported by a substantial, stable record. But the threshold is not a law. The evidence behind the number matters more than the label attached to it.
Losses exceeded profits
The measured closed trades lost money under the report’s assumptions. Check whether the period is representative before drawing a broader conclusion.
Gross break-even
Gross winning and losing totals are equal. Any omitted costs would push the result below break-even.
Thin margin
The measured edge has little room for worse spreads, slippage, fees, outliers or a change in market conditions.
Worth investigating
Positive evidence, but not standalone approval. Cost sensitivity and stability become decisive.
Strong headline result
Potentially meaningful if it comes from complete, cost-adjusted and sufficiently varied evidence.
Impressive—so audit it
Check for a small sample, a few dominant winners, cherry-picked dates, overfitting or missing open losses.
Calculation
How to calculate profit factor
MetaTrader 5 and TradingView define profit factor as gross profit divided by gross loss.12 Use the absolute value of gross loss so the displayed ratio is positive.
Gross Profit ÷ |Gross Loss|
Gross profit
Add the positive P&L from every winning closed trade in the stated period.
Gross loss
Add the negative P&L from every losing closed trade, then use its absolute value.
Net profit ÷ losses
Net profit is gross profit minus gross loss. Substituting it changes the meaning of the ratio.
Hypothetical trade summary
$15,300 ÷ $9,000
1.70 profit factorThe sample’s gross losses were greater than its gross profits.
Gross profit equalled gross loss before any costs omitted from the report.
The denominator is zero, so the ratio is mathematically undefined. Software may display infinity.
Original ATI framework
The six-part ATI Profit Factor Audit
A headline ratio becomes useful evidence only after these six questions have defensible answers.
Definition
Confirm the exact formula, dates, currency, trade grouping and treatment of breakeven trades. Do not assume two platforms calculate from identical inputs.
Evidence:Report methodology and export settingsCompleteness
Obtain the closed-trade list and inspect open positions, equity, deposits and withdrawals. Profit factor commonly uses realized closed trades only.
Evidence:Full history plus current open P&LCosts
Recalculate every trade after realistic commissions, spreads, slippage, swaps and relevant account or provider charges.
Evidence:Net fills and itemized cost assumptionsConcentration
Measure how much gross profit came from the largest one, five and ten winners. Recalculate without them as a sensitivity test.
Evidence:Ranked trade outcomesEvidence
Label the result backtested, demo, forward-tested or live. Record its duration, trade count, instruments and market coverage.
Evidence:Verifiable record and sample descriptionStability
Compare rolling periods, long and short sides, instruments, regimes and out-of-sample or live data. Look for persistence, not one peak number.
Evidence:Segmented and rolling analysisA reported profit factor can pass one check and fail another. For example, the formula may be correct while its costs are unrealistic, or a live record may be genuine while a single trade supplies most of its profit.
Use the complete ATI system-evaluation frameworkHidden weaknesses
Four ways an attractive profit factor can mislead
None of these weaknesses proves misconduct or guarantees future failure. Each identifies evidence that the ratio does not contain.
Open losses may be outside the calculation
TradingView explicitly notes that its profit-factor calculation uses realized figures and excludes currently open positions.2 A system can therefore show an attractive closed-trade ratio while holding a substantial floating loss.
This is especially important when a strategy delays closing losing positions, averages down, or realizes small winners while adverse positions remain open. The reported ratio may be accurate for the closed ledger and still be incomplete as an account-risk description.
Request current equity, lowest historical equity, open-position history and both balance and equity curves. Missing open-position data is incomplete evidence; it is not by itself proof that a provider is concealing losses.
The same ratio can describe very different strategies
Profit factor combines the frequency and size of wins and losses. It compresses those ingredients, so different trade distributions can produce the same result.
In the hypothetical comparison below, both systems complete 100 trades, generate $6,000 of gross profit and $4,000 of gross loss, and finish with a 1.50 profit factor and $20 expectancy per trade. Yet the experience of trading them would be very different.
| Measure | System Frequent | System Selective |
|---|---|---|
| Winning trades | 60 × $100 | 20 × $300 |
| Losing trades | 40 × $100 | 80 × $50 |
| Win rate | 60% | 20% |
| Average win ÷ average loss | 1.00 | 6.00 |
| Gross profit | $6,000 | $6,000 |
| Gross loss | $4,000 | $4,000 |
| Profit factor | 1.50 | 1.50 |
| Expectancy per trade | $20 | $20 |
One exceptional trade can dominate the numerator
Suppose a record reports $12,000 of gross profit and $6,000 of gross loss: a profit factor of 2.00. If its largest winner contributed $5,000, that single trade supplied 41.67% of gross profit. Removing it leaves $7,000 ÷ $6,000, or a profit factor of 1.17.
The sensitivity result does not prove the winning trade was luck or should be deleted from the official record. Trend-following strategies may legitimately depend on rare, large winners. The test reveals concentration: the reader can then ask whether the strategy is designed to capture such events and whether the sample includes enough opportunities to evaluate that claim.
Costs can consume a thin margin
Commissions, spreads, slippage and financing affect individual trades. They can shrink winners, enlarge losses and turn small winners into losing trades. Investor.gov advises readers to examine which fees and expenses are included in performance calculations because costs reduce returns.57
For this reason, do not subtract one total-cost estimate from net profit and assume the old profit factor still applies. Recalculate the classification and net result of each trade after its relevant costs.
| Hypothetical calculation | Gross profit | Gross loss | PF | Net result |
|---|---|---|---|---|
| Optimistic report | $18,000 | $12,000 | 1.50 | $6,000 |
| After realistic trade-level costs | $16,500 | $13,500 | 1.22 | $3,000 |
Context
Read profit factor with—not instead of—other metrics
No single performance statistic can describe return, reliability and capital risk at once.
| Metric | Question it answers | What it can miss |
|---|---|---|
| Profit factor | How much gross profit was recorded per unit of gross loss? | Sequence, drawdown, open P&L and reliability |
| Win rate | What share of closed trades won? | The size of wins and losses |
| Average win/loss | How large was the typical winner relative to the typical loser? | Frequency, outliers and sequence |
| Expectancy | What was the average outcome per trade in currency or R? | Capital path and uncertainty |
| Maximum drawdown | How deep was the largest observed peak-to-trough decline? | Future losses and trade-level profitability |
| Trade count and duration | How much and what kind of history supports the statistics? | Independence and market coverage |
When trades are wins or losses
Profit factor’s underlying relationship
(Win Rate × Average Win) ÷ (Loss Rate × Average Loss)
This explains why a low-win-rate strategy can have a strong profit factor when its average winner is much larger than its average loser—and why a high win rate can coexist with a weak ratio when occasional losses are very large. TradingView likewise cautions that win frequency does not reflect the magnitude of outcomes.3
Reliability
The number is only as credible as the record behind it
Separate mathematical correctness from evidential strength. A ratio can be calculated correctly from an unrepresentative sample.
Historical simulation
Useful for understanding rules and testing many conditions, but sensitive to selection, assumptions, hindsight and overfitting.
Current-market observation
Useful for checking signal behavior and timing, but may not reproduce the financial and execution pressures of a funded account.
Actual executed trading
Stronger execution evidence, but still historical and potentially short, concentrated or regime-specific.
Stability examination
Rolling, instrument, direction and regime breakdowns reveal whether one aggregate ratio hides weak periods.
Investor.gov says backtested performance is hypothetical, past performance cannot predict future results, and cherry-picked periods should be questioned.5 NFA similarly notes that hypothetical trading may benefit from hindsight and may not fully reflect liquidity, slippage or the financial pressure of actual losses.6
The solution is not to dismiss every backtest. It is to identify its role, preserve a genuinely unseen test period, compare it with live behavior, and examine the strategy under varied and stressed conditions. The FCA emphasizes robust simulation testing, including historical and theoretical stress scenarios, while FINRA highlights testing, limited pilot deployment and post-implementation monitoring.89
CFA Institute also describes rolling-window tests, scenario analysis and sensitivity analysis as complementary tools, noting structural breaks and the difficulty of representing downside tails from history alone.10
See how to compare backtest, out-of-sample and live evidencePractical use
How to compare two profit factors responsibly
Do not automatically prefer the larger number. First make the underlying evidence comparable.
- 01
Match the evidence type
Compare live with live or backtest with backtest. Do not treat a simulated 2.20 and a live 1.45 as measurements of equal evidential strength.
- 02
Match the accounting
Use the same currency basis, dates, cost treatment, position grouping and closed-trade convention.
- 03
Check scale and coverage
Record trade count, duration, instruments, direction, market sessions and regimes. More trades do not automatically mean more independent information.
- 04
Stress concentration and costs
Remove top winners as a sensitivity test, increase execution assumptions and recalculate from individual trades.
- 05
Read the capital path
Compare maximum equity drawdown, time under water, worst open loss, leverage and losing-streak behavior.
- 06
Prefer stable evidence
A slightly lower ratio that persists across unseen and live periods may be more decision-useful than a spectacular in-sample peak.
Due diligence
Ten questions to ask about a reported profit factor
These questions turn a marketing statistic into an evidence request.
- 01
Is the value calculated from backtested, demo, forward-tested or live trades?
- 02
What exact dates, instruments and number of trades does it cover?
- 03
Are commissions, spread, slippage, swaps and provider fees included?
- 04
Can I inspect the complete trade-level export used in the calculation?
- 05
Does the figure exclude current or historical floating losses?
- 06
What percentage of gross profit came from the largest one, five and ten winners?
- 07
How does profit factor change across rolling periods and market regimes?
- 08
What were the out-of-sample and live values compared with the optimized backtest?
- 09
What maximum equity drawdown and leverage produced the result?
- 10
Were the strategy rules, position sizing or risk limits changed during the record?
Evidence warnings
When the headline ratio deserves extra scrutiny
A warning sign is a reason to investigate—not automatic proof that a provider or system is fraudulent.
Learn how to assess unsupported trading-bot claims- !
A very high value from only a few trades or a short period
- !
No trade list, calculation method or cost assumptions
- !
A backtested ratio presented as though it were live performance
- !
Large differences between optimized, out-of-sample and live values
- !
Closed-trade statistics shown without equity or open-position evidence
- !
One extraordinary winner responsible for much of gross profit
- !
A fixed benchmark used as the sole reason to buy or allocate capital
- !
Claims that the ratio proves low risk, safety or future profitability
Frequently asked questions
Profit factor questions, answered
Is a profit factor of 1.5 good?
A value of 1.5 is a strong screening result because the sample recorded $1.50 of gross profit for each $1.00 of gross loss. It becomes meaningful only if the calculation includes realistic costs and is supported by enough varied, preferably out-of-sample or live, evidence. It does not prove future profitability or low risk.
Can profit factor be too high?
A high value is mathematically possible, but it deserves more scrutiny when it comes from a small or optimized sample. Check whether one or two winners dominate gross profit, whether losing positions remain open, and whether the value falls sharply outside the development period. High is not inherently bad; unsupported high is weak evidence.
What does a profit factor below 1 mean?
It means gross losses exceeded gross profits in the measured closed trades. The sample was unprofitable under those accounting assumptions. You should still examine the period, costs and strategy changes before deciding whether the result represents the system more broadly.
What does an infinite profit factor mean?
The calculation has no losing closed trades, leaving a zero denominator. The mathematical ratio is undefined, although software may display infinity. This often reflects a very small sample or a record that has not realized its losses. Inspect trade count, open P&L and the full observation period.
Is profit factor better than win rate?
It answers a more complete profitability question because it reflects both win/loss frequency and outcome size. But it still does not show sequencing, drawdown, open losses or reliability. Use profit factor and win rate alongside average win/loss, expectancy, drawdown and sample evidence.
Does profit factor include open trades?
Common platform calculations use realized results from closed trades. TradingView explicitly excludes currently open positions from its profit-factor metric. Other reports may use different conventions, so confirm the definition and inspect current equity and open P&L separately.
How many trades are needed for a reliable profit factor?
There is no universal trade-count cutoff. Reliability depends on the strategy’s frequency, independence of trades, market coverage, outcome distribution and whether parameters were selected using the same data. State the trade count, but also test rolling periods, unseen data, regimes and live results.
Should profit factor include commissions and slippage?
Yes, if the goal is to estimate tradable performance. Apply realistic commissions, spreads, slippage, swaps and other relevant costs to each trade, then recalculate. A pre-cost ratio can be useful for diagnosing the raw signal, but it must be labelled clearly and should not be presented as the investor’s likely result.
The ATI principle
Audit the dollars behind the ratio
Profit factor is a useful compression of realized wins and losses. Its simplicity makes it easy to compare—and easy to overinterpret.
Start with the formula, then examine completeness, costs, concentration, evidence and stability. If you cannot see the trade list, current equity and calculation assumptions, the correct conclusion is not that the system is bad. It is that the evidence is incomplete.