Skip to content
← All guides

Guide · 5 min read · Updated 17 November 2026

Is your edge just luck? The random-entry test

A profitable backtest does not prove your entry signal is doing anything. In a rising market almost any long-only strategy makes money. The random-entry test asks a sharper question: if you had entered at random instead, would you have done just as well?

How the test works

Keep everything about the strategy the same except the entry signal: the same market, the same period, the same holding time and the same costs. Replace the entries with random ones, run it many times, and look at the spread of results. That spread is the luck zone.

Reading the result

  • Well above the luck zone: the entries seem to add something beyond chance. Evidence, not proof.
  • Inside the luck zone: you cannot tell skill from noise. The profit may simply be the market's drift.
  • Below the luck zone: random entries did better, which suggests the signal is hurting.

Why it matters

Many popular strategies only look good because they were tested during a bull market. The random-entry test separates the market's return from the strategy's.

A quicker version: the coin-toss check

A related check needs no simulation. Every strategy has a break-even win rate set by the size of its average win against its average loss. If winners are twice the size of losers, you only need to win about a third of the time to break even.

The coin-toss check compares the actual win rate with that break-even rate and asks whether the gap is larger than luck alone would produce on that many trades. We use roughly two standard errors as the line. With 20 trades, a win rate a few points above break-even is well within luck. With 300 trades, the same gap starts to mean something.

This is the version we run on every AlphaEdgeCheck result today, together with comparing against buy-and-hold. A full random-entry simulation is not part of our results yet. See all our checks.

Common questions

Is the random-entry test the same as a Monte Carlo simulation?
It is one kind of Monte Carlo test. Monte Carlo just means repeating an experiment many times with random inputs. Here the random input is the entry timing.
What is a break-even win rate?
The win rate at which a strategy neither makes nor loses money, given its average win and average loss. It equals average loss divided by the sum of average win and average loss, before costs.

Research, not advice. This guide is general education. It does not recommend buying or selling anything. Read the risk disclosure.