Despite winning all season, your baseball team can disappoint in the playoffs. The Oakland Athletics played 162 games and won 60% of them, then lost the one-game wildcard playoff (again!). All probable outcomes suffer from statistical significance. If you're 60% sure your trade will succeed, that's a phenomenal edge. But if you make a single bet, you lose 4 times out of 10. And remember—you only care about the outcome realized. The Athletics went home. That's what happened. Life has a statistical sample size of 1.

So how do you capitalize on your edge? Make more bets. Play more games. In a 7-game series, a team with a 60% win rate wins the series 71% of the time. If you place 41 trades with that same edge, you come out ahead 90% of the time.

The math here is the binomial distribution. If your probability of winning a single game is p, the probability of winning at least k games out of n is:

P(X ≥ k) = Σ C(n, i) · pi · (1 − p)n − i, for i = k to n

For a 7-game series (first to 4 wins), we sum the probabilities of winning 4, 5, 6, or 7 games (n = 7, k = 4, 5, 6, 7)—which comes to 71%. But what if you want 90% confidence? Working backwards, we solve for the smallest n such that P(X > n/2) ≥ 0.90—i.e., more wins than losses. The answer is 41 trades.

Effective win rate vs number of attempts for a 60% base probability

This is the fundamental argument for diversification, for taking many small positions instead of one large one, for building systems that let you make the same bet repeatedly. A 60% edge sounds modest. But compounded across enough trials, it becomes nearly certain.

The catch: most people don't have enough at-bats. They make one or two big decisions—one job, one investment, one launch—and live with the outcome. The edge was real. The sample size wasn't.