What Consistent Profitability Actually Means
Consistent profitability is the stated goal of almost every trader and is rarely defined. Defined loosely it means making money most months, which is a target that reliably produces the behaviour that prevents it.
Defined properly it is a statistical property of a long sequence of trades. This page sets out what that means and why the difference between the two definitions matters so much in practice.
Expectancy Is the Underlying Quantity
Expectancy combines how often a method wins, the average size of a win, and the average size of a loss, after costs. It is the figure that determines whether an approach makes money over a sequence.
Everything else — win rate, monthly results, the last ten trades — is a sample drawn from it and tells you about the sample as much as about the method.
Accuracy Alone Says Almost Nothing
A method winning most of the time with small gains and occasional large losses has negative expectancy. One winning less than half the time with larger gains than losses has positive expectancy.
Which is why accuracy is the most quoted and least informative statistic available, as covered in evaluating trading strategies.
Variance Is Not a Flaw
Even a positive-expectancy method produces losing runs. Outcomes are drawn from a distribution, and sequences of losses occur naturally within one.
Treating every losing run as evidence the method has broken is the single most expensive misunderstanding in trading, because it causes working approaches to be abandoned.
Sample Size Determines What You Can Conclude
A dozen trades tell you nothing. Short sequences are dominated by variance, and both sound and poor methods produce almost any short-run result.
Judging requires enough trades for variance to average out, across varied conditions rather than a single favourable stretch.
Monthly Consistency Is the Wrong Target
Requiring every month to be profitable forces trading in conditions that do not suit the method, because the alternative is a flat month.
That produces overtrading in quiet periods and oversizing in poor ones — the precise behaviours that turn positive expectancy negative.
Consistency of Process, Not of Result
What can genuinely be made consistent is preparation, contract selection, sizing, exits and review. Results vary regardless.
Measuring yourself on the controllable inputs makes improvement possible; measuring on outcomes produces conclusions drawn from variance.
Drawdown Is the Binding Constraint
A method’s worst losing sequence matters more than its average result, because that sequence determines whether it can be followed to completion.
A method abandoned partway delivers whatever result it had at the point of abandonment, which makes tolerance the real limit rather than expectancy.
Size to the Drawdown You Can Actually Survive
Estimate the worst run the method has produced and ask honestly whether you would continue through it. If not, reduce size until the answer becomes yes.
A smaller position in a method you can follow beats a larger one in a method you will abandon, which is the practical meaning of the constraint.
Costs Are Deducted From Expectancy, Always
Brokerage, exchange charges, levies and the spread apply to every round trip regardless of outcome. In options the spread alone is proportionally wide.
An expectancy calculated on price movement rather than on net results is not an expectancy, and it is why methods sound in analysis fail in practice.
Frequency Interacts With Cost
Costs scale with activity while the edge does not. A method with a modest edge traded selectively can be viable; the same method traded constantly frequently is not.
This is why frequency deserves as much scrutiny as accuracy, and why more trading is rarely the route to more consistency.
Options Add Non-Directional Ways to Lose
Premium responds to direction, magnitude, elapsed time and volatility expectations. A correct view can lose to decay or to volatility collapsing after an event.
Any expectancy calculation for options must account for these, since they produce losses on trades where the analysis was sound, as set out in options intraday tips.
Consistency Assumes Consistent Execution
Recorded expectancy assumes the method was applied identically each time. Where stops are occasionally widened and sizing varies, the record describes a mixture of approaches.
Recording whether the plan was followed, and reviewing compliant trades separately, is what makes the figure mean anything.
Regime Dependence Limits the Claim
Every method needs particular conditions. Breakout logic needs directional sessions, range logic needs boundaries that hold, premium selling needs quiet markets.
A method that earns in one regime and gives it back in another is not consistently profitable; it is a bet on conditions persisting.
Beware Consistency Produced by Hidden Risk
Some approaches produce long sequences of small gains punctuated by rare large losses. Premium selling is the obvious example.
Such a method looks consistent for a long time and is not, because the distribution simply has not delivered its tail yet. Position limits matter more here than anywhere.
Published Consistency Claims Deserve Four Questions
Over what period? All trades or a selection? At what assumed execution prices? Net of what costs?
A claim lacking these cannot be interpreted, and selection effects — discontinued strategies, favourable start dates — are the usual reason it looks smooth.
Compare Against the Right Benchmark
The relevant comparison is what the same capital could have done elsewhere at comparable risk, less the time and stress consumed.
A method producing modest returns for constant attention may be underperforming a far simpler alternative, as the framework in investment advisory sets out.
Define Your Own Test in Advance
Commit to a number of trades, a maximum tolerable drawdown, and the expectancy that would justify continuing, before the sequence begins.
Setting the test afterwards guarantees it is set to whatever the results happened to be, which is not a test at all.
Then Let the Sequence Run
Consistency is demonstrated by a long run of consistent behaviour producing a positive expectancy net of costs, not by a good month.
The routine that supports that is set out in the intraday trading guide, and the practical starting sequence in intraday tips for beginners.
Consistency Requires Consistent Sizing
Risking a consistent small fraction of capital means position size falls automatically as the account declines and rises as it grows.
Fixed quantities do the opposite, holding the loss constant while the capital supporting it shrinks, which converts a survivable drawdown into an unrecoverable one.
Instrument Choice Changes the Distribution
A concentrated benchmark produces wider outcomes than a broad one, and options produce wider outcomes again than linear exposure on the same underlying.
An expectancy measured on one instrument does not transfer to another, so the sequence has to be rebuilt when the instrument changes, as the differences in index intraday tips describe.
Do Not Confuse a Good Run With an Edge
A favourable sequence produces confidence, larger positions and the belief that the method has been proven. Frequently it has produced a sample.
Increasing size on a winning run is the most common route from a promising start to a serious loss, because the size arrives just as the sequence reverts.
FAQs
What does consistent profitability actually mean?
A positive expectancy net of costs demonstrated over a sequence long enough for variance to average out — not making money every month.
Why is monthly consistency the wrong target?
Because it forces trading in conditions that do not suit the method, producing overtrading in quiet periods and oversizing in poor ones.
Is a losing run evidence the method has broken?
Usually not. Positive-expectancy methods produce losing sequences naturally, and treating each one as failure causes working approaches to be abandoned.
Why does drawdown matter more than average return?
Because it determines whether the method can be followed to completion. A method abandoned partway delivers whatever it had at that point.
Can a method look consistent and not be?
Yes. Approaches producing many small gains and rare large losses appear stable until the tail arrives, which is why position limits matter most there.
Must costs be inside the calculation?
Always. An expectancy computed on price movement rather than net results is not an expectancy, and costs scale with frequency while the edge does not.
How should I test my own method?
Define the number of trades, the tolerable drawdown and the expectancy that would justify continuing, before the sequence begins rather than after.
Does expectancy transfer between instruments?
No. A concentrated benchmark produces wider outcomes than a broad one, and options wider again, so the sequence has to be rebuilt when the instrument changes.
Is a good run evidence of an edge?
Frequently it is evidence of a sample. Increasing size on a winning run is a common route to a serious loss, because the larger size arrives as the sequence reverts.