Kelly Sizing, Martingale Risk, and the Limits of Fixed Trading Odds
Summary
The author reflects on losses in leveraged forex trading and explains two staking approaches using a fair coin game. A martingale raises stakes after losses and depends on effectively unlimited capital; a reverse martingale risks a fixed fraction of current wealth, reducing exposure as losses accumulate and increasing it as gains compound. The article presents Kelly sizing formulas that relate the stake fraction to win probability and win-to-loss payoff, and argues that positive expected value is necessary for growth.
The discussion connects sizing to discipline, drawdown risk, and the gambler’s ruin problem. The author initially claims that proportional staking can turn even fair odds into eventual wealth, but a later contribution challenges applying casino formulas to markets: trading odds are estimated from past data, may change, and can be distorted by fat tails and extreme losses. The article’s own claimed trading gains are anecdotal, not a controlled test. Its formulas rely on known, stable probabilities and payoffs; estimation error, changing regimes, leverage, and ruin risk limit their direct use in trading.
Key ideas
- Martingale staking increases bets after losses and can fail when capital is finite.
- Proportional staking adjusts each bet to current wealth, so losses shrink the next stake while gains enlarge it.
- Kelly sizing uses estimated win probability and payoff to choose a growth-oriented stake fraction.
- Kelly assumptions may not hold in markets, where probabilities and payoffs shift and tail losses can exceed historical estimates.
- The author's reported trading experience is anecdotal and does not establish that the approach reliably produces profits.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.