On this page (5 sections)
Content reviewed: 2026-10-13. Examples are hypothetical.
A losing streak can result from ordinary variation, poor probability estimates, execution errors or a change in the market. The appropriate response depends on the cause. Increasing stakes solely to recover a previous loss does not establish an edge.
Measure what actually happened
Reconcile filled trades and fees before diagnosing the strategy. Separate realized losses from the marked value of unresolved positions. Compare each result with the probability written down before entering, not a revised explanation after the event.
For illustration, if five independent contracts each truly have a 60% chance of winning, the probability of losing all five is 0.4^5 = 1.024%. That is the probability for that specified block of five. It is not the probability of ever encountering a five-loss run across hundreds of trades, and independence may be a poor assumption for related events.
Quantify the drawdown
If each loss consumes 5% of the remaining bankroll, five successive losses leave 0.95^5, about 77.38%, of the starting bankroll. Recovering from that 22.62% drawdown requires approximately 29.24% growth from the remaining capital. Percentage losses and recovery gains use different denominators.
This arithmetic is a stress test, not a recommendation to risk 5%. Kelly's growth framework also depends on the assumed probability and payoff; using an overstated edge can produce an oversized stake.[1]
Audit process separately from outcomes
| Question | Record to inspect |
|---|---|
| Was the forecast well defined? | Original question and probability |
| Did the contract match the forecast? | Rules and resolution source |
| Was the intended price obtained? | Fills, spread and fees |
| Were losses concentrated? | Shared event and platform exposures |
| Did behavior change after a loss? | Stake sizes, frequency and written limits |
Score a consistent set of resolved forecasts instead of counting only wins. Proper scoring rules provide a way to evaluate stated probabilities.[2] Keep unsuccessful trades in the sample and compare with the same baseline used before the streak.
Reduce exposure while investigating
A pause gives you time to review without adding new risk. Resume only when the records support a specific explanation and your budget still permits trading. Paper-tracking the next set of forecasts can test whether the process is consistent, although it does not reproduce real execution costs or the experience of losing money.
Kalshi offers trading breaks, self-exclusion and deposit limits through its responsible-trading controls.[3] If you feel unable to follow a stop decision, use those controls and seek appropriate support rather than designing a larger recovery bet. A previous loss creates no obligation to make another trade.
References
- J. L. Kelly Jr. A New Interpretation of Information Rate. Bell System Technical Journal 35(4), 917-926, July 1956. DOI: 10.1002/j.1538-7305.1956.tb03809.x.
- Tilmann Gneiting and Adrian E. Raftery. Strictly Proper Scoring Rules, Prediction, and Estimation. Journal of the American Statistical Association 102(477), 359-378, March 2007. DOI: 10.1198/016214506000001437.
- Kalshi. Responsible Trading Tools. Accessed October 13, 2026.