Sizing amplifies edge β it doesnβt create it
Compounding a real edge is how small accounts grow; compounding a negative edge is how they blow up. This tool shows both curves side by side from your own numbers. Find your optimal risk fraction on the Kelly criterion calculator.
Percent sizing or fixed dollars: the real difference
Flat sizing risks the same dollars every trade; percent sizing risks the same share of a moving account. Same win rate, same strategy β different account curves. Percent sizing compounds gains and self-throttles in drawdowns (risking 1% of a shrinking account shrinks the bets); flat sizing grows linearly and hits drawdowns at full nominal weight.
The numbers over 100 trades at +0.3R expectancy, 1% risk: percent sizing lands near +34% with drawdowns softened by the automatic de-risking; flat sizing yields +30% with every loss hurting the same regardless of account state. The gap widens with time and expectancy β compounding is exponential, flat is arithmetic.
Why anyone chooses flat anyway: psychological stability (each trade means the same dollars, no creeping stake anxiety) and immunity to the "compounding into a losing streak" pattern, where percent sizing after a hot run risks the largest-ever dollar amounts exactly as the edge cools. Serious operations often blend: percent sizing, rebased monthly rather than per-trade.
FAQ
Why does percent-of-equity sizing compound faster?
Because your bet size scales with the account. After wins you risk more dollars, after losses less β geometric growth. With a genuine positive edge this outpaces a fixed dollar bet, but the same mechanism deepens drawdowns, so most pros cap risk at 1-2% per trade.
Can position sizing turn a losing system into a winner?
No. If your expectancy (win rate Γ reward β loss rate) is negative, every sizing scheme still loses β compounding just loses faster. Sizing optimizes a positive edge; it cannot manufacture one. Validate the edge before tuning size.