Guide
Law of Large Numbers in Trading
The law of large numbers explains why a trading edge becomes clearer over larger samples, while short samples can look random or misleading.
The simple idea
When the number of independent observations grows, the observed average tends to move closer to the expected average. In trading, this means a strategy's edge needs enough trades before the observed result becomes more reliable.
It does not mean every short sample must match the expected win rate or expectancy.
Small samples can look wrong
A system with a 55% assumed win rate can show 40% wins over a short sample, or 70% wins after a lucky start. Neither result proves the true edge by itself.
This is why sample size and variance need to be considered before changing a strategy after a small group of trades.
Trades need to be comparable
The law of large numbers is most useful when the observations are reasonably comparable. In trading, that means the trades should come from the same setup logic, similar execution rules and a market environment where the strategy is meant to operate.
If a sample mixes unrelated strategies, discretionary changes, different instruments and different risk rules, the average can become harder to interpret. More trades help only when the sample is measuring something consistent.
How sample size changes confidence
Each trade has less influence as the sample grows. That does not make the path smooth, but it makes the observed average less fragile.
| Sample size | One trade changes win rate by | Interpretation |
|---|---|---|
| 10 trades | 10 percentage points | Very noisy |
| 50 trades | 2 percentage points | Still streak-sensitive |
| 100 trades | 1 percentage point | More useful for review |
| 200 trades | 0.5 percentage points | More stable, still uncertain |
What the law does not promise
It does not promise a smooth equity curve. It does not prevent losing streaks. It does not guarantee that a strategy's edge still exists if the market changes.
It only explains why larger samples are usually more informative than smaller samples when the assumptions remain stable.
Do not use it as an excuse to ignore feedback
A trader can misuse the law of large numbers by saying every bad result just needs more trades. That is not good enough. If execution has changed, costs are higher, market conditions are different or losses are larger than planned, the system needs review.
The practical balance is simple: do not overreact to ten trades, but do not blindly defend a system when a meaningful sample shows that the assumptions no longer match reality.
How to apply it practically
Use the trading probability simulator to compare 20-trade, 100-trade and 200-trade samples with the same inputs. The averages may become more stable, but the path can still include drawdown.
Then use sample size in trading and how many trades do you need to decide how much evidence you have.
Frequently asked questions
Does the law of large numbers guarantee profits?
No. It only describes how averages behave over larger samples. The strategy still needs a real edge, stable execution and risk control.
Can a good system lose over 20 trades?
Yes. Twenty trades can be dominated by outcome order and short-term variance.
Does more trades remove drawdown?
No. Larger samples can make averages clearer, but losing streaks and drawdowns can still appear.
Does the sample need to use the same strategy?
Yes, as much as possible. Mixing unrelated strategies or changing execution rules makes the average harder to interpret.
Why do traders overreact to small samples?
Because short-term results feel immediate and personal, even when the sample is too small to prove much statistically.