Thinking In Probabilities, Not Certainties

A setup shifts the odds, it does not decide the outcome

Thinking in probabilities means accepting that no trade can be known in advance to work, and that a well-researched setup changes the distribution of outcomes without determining any single one. This is not a consolation. It is the operating assumption that everything else in a trading plan is built on.

What follows from it is uncomfortable but clarifying. A loss on a well-reasoned trade is not automatically a mistake. A win on a poorly-reasoned trade is not validation. Decisions have to be judged by the quality of the reasoning, because the single outcome that followed is mostly noise.

What 100 trades of the same method can look like

The scale of that noise is worth seeing in numbers, because most people underestimate it badly. Take an approach that resolves in your favour four times in ten, risking 1R and making 2R, so the expectation is a positive 0.2R per trade. Over 100 trades, the expected result is plus 20R. Here is what the actual result is, depending on how the hundred falls.

The ordinary statistical spread for 100 trades at those odds covers roughly 30 to 50, which means an unchanged method with a genuine positive expectation can produce anything from a losing hundred to a spectacular one. Same rules, same instrument, same trader. That range is why a quarter tells you far less than it feels like it does, and why judging an approach on ten trades is measuring almost nothing at all.

Why this makes sizing the first decision

If any individual trade can lose regardless of how well it was prepared, then no individual trade can be allowed to risk an amount that would be damaging on its own. That is the whole argument for position sizing, and it is a consequence of probability rather than a matter of temperament.

The consequence compounds in the wrong direction as risk per trade rises. A positive expectation only pays out across a sample, so the account has to survive long enough to reach one. Raise per-trade risk far enough and the probability of hitting an unrecoverable drawdown before the sample completes becomes the dominant term, which is exactly what the risk of ruin calculator is showing you. A method with an edge and reckless sizing is still a method that ends. Set the number with the position size calculator and leave it alone for a block of trades.