Real World Maths Busters

The Maths Behind Everyday Life

The story, the formula and a worked example — see exactly where this maths shows up in the world around you.

Back to all stories
Games & Chance 4 min read

Is the "hot hand" in basketball real or a myth?

For decades, statisticians said no. Newer maths says they were wrong.

Players and fans swear a shooter "gets hot" — that makes seem more likely to hit the next shot after making several. Statisticians called it a fallacy in 1985. But in 2015, new analysis showed the original study had a subtle statistical bias.

The conditional probability puzzle

The question is: is P(hit∣previous 3 hits)>P(hit)P(\text{hit} \mid \text{previous 3 hits}) > P(\text{hit})? A player shooting 50% overall might shoot 55% after three consecutive makes.

The 1985 study looked at sequences and found no effect, but the method was biased: restricting to streaks of hits selects for lucky runs, which makes the NEXT shot look average even if the player is genuinely hot.

The bias in streak selection

Consider a fair coin flipped 4 times. What is the probability of heads right after a heads?

Selection bias in streaks

  1. 1

    All 4-flip sequences

    There are 16 possible sequences of 4 flips (HHHH, HHHT, HHTH, ...).

  2. 2

    True P(H)

    P(H)=0.50P(H) = 0.50
  3. 3

    P(H after H) in finite sequences

    E[P(H∣prev H)]≈0.40E[P(H \mid \text{prev H})] \approx 0.40
  4. 4

    The bias

    In short sequences, conditioning on a streak being present biases the sample — P(H after H) comes out BELOW 50% even for a fair coin. The original "hot hand" study fell for this.

Answer

The "hot hand fallacy" was itself partly a fallacy — newer analysis correcting for sequence-selection bias suggests the hot hand is real, albeit small (2–3 percentage points).

The takeaway

The hot-hand debate shows how subtle conditional probability can be. Even professional statisticians got it wrong for 30 years because of a sampling bias hiding in plain sight.

Back to all stories