Splitting pots and tie payoffs

Distribute a tied pot, distinguish returned chips from net payoff, and include ties in an expected-value model.

12 minutes · beginner · Original Poker4Quant lesson

What you will learn

Divide a single eligible pot among tied winners and calculate the incremental payoff of a final call when ties are possible.

Prerequisites: Hand rankings and ties · The price of a call · Explaining your first decision

Equal hands share an eligible pot

A showdown compares each player’s best five-card hand. If eligible players have identical best hands, they share the pot; suit does not break the tie. A player using the board can tie another player even when their private cards differ. A stronger-looking unused sixth card does not break a best-five tie.

Start with one pot and equal eligibility. If two players tie for a 120-chip pot, each receives 60. Three tied winners sharing 150 receive 50 each. Dividing the pot does not create chips: the shares must add back to the original pot. These examples exclude rake, side pots and indivisible odd chips; those require additional rules.

Returned share and incremental payoff differ

For a final call, let P be the pot before your new call and C the cost of that call. The final pot is P+C. If m eligible players tie, your returned share is (P+C)/m. Relative to folding now, the incremental tie payoff is that share minus C. Earlier contributions are already sunk and should not be subtracted again in this comparison.

A tie is therefore not automatically a zero-payoff outcome. With P=80, C=20 and a two-way tie, the final pot is 100, your share is 50, and your incremental payoff is +30. This does not mean you made 30 over the whole hand: the comparison begins at the current call decision. Keep the baseline explicit.

Give each outcome its own payoff

In a no-fee final-call model, a win gains P incremental chips, a loss costs C, and a tie gains (P+C)/m−C. Weight these three payoffs by mutually exclusive win, tie and loss probabilities that sum to one. Do not count a tie as both a win and a loss or use a tie fraction without specifying the number of tied winners.

These are risk-neutral chip expectations under supplied probabilities, not forecasts or advice to wager. More complex pots can have several eligible subsets, and the number of winners may vary by outcome. Build those cash flows separately instead of applying one equal split to every chip on the table.

Worked example

A final call costs 20 into an existing pot of 80. Assume 20% win, 40% two-way tie and 40% loss. What is the expected incremental payoff?

The probabilities add to one: 0.20+0.40+0.40=1.
Winning gains the pre-call pot: +80. Losing costs only the new call: −20.
On a tie, the final pot is 100. Your returned half is 50, so the incremental tie payoff is 50−20=+30.
Weight all three outcomes: 0.20×80 + 0.40×30 + 0.40×(−20) =16+12−8=20.

Expected incremental payoff is +20 chips under this model. It is neither a guaranteed result nor the net profit of the entire hand.

Hands-on: Distribute a tied pot and audit the call

Move conserved fake chips into equal eligible shares, subtract the new call and calculate a win/tie/loss expected payoff.

Reading chip split…

Common mistakes

Using suit or an unused sixth card to break identical best-five hands.
Entering zero for every tie without reconstructing the returned share.
Subtracting the call twice, or subtracting sunk contributions from an incremental decision.
Splitting side pots as if every player were eligible for every chip.

Keep these ideas

Split a single eligible pot equally among tied best hands.
Conserve every chip and state any excluded rules.
Tie payoff relative to folding now is returned share minus the new call.
Weight win, tie and loss separately under explicit assumptions.