Suppose a strategy has a 60% win rate: when right, it gains 50%; when wrong, it loses 50%. Sounds good, right?
If you go all-in every time, calculate the long-term result: 60% × 50% - 40% × 50% = +10% expected per trade,
looks positive. But note the path: two consecutive losses mean -75%, and you need +300% just to break even.
The expected strategy + full-position execution = a high probability of bankruptcy. This is the gambler's ruin theorem playing out in the crypto world
every day.
With the same strategy, if you change it to only using 20% of your position each time: two consecutive losses = -18%, with no real damage,
and you still collect the long-term expectancy. The strategy hasn't changed; only the position size has. The outcome is completely different.
Remember this inequality: your long-term returns ≈ strategy expectancy × how long you can stay in the game. Position sizing determines
the latter. Only those who last can talk about expectancy.
What do you think? Share your experiences in the comments.