PokerToast · study tools for mixed-game poker
In Razz, should I complete a three-card 9 (9-5-2) against a King bring-in?
Heads-up, PokerToast's limit Razz solver completes with 9-5-2 against a King bring-in 99% of the time.
The spot
- Game: heads-up limit Razz, 3rd street.
- You: (5♠ 2♦) 9♣ (down cards in brackets): three cards to a 9-low (9-5-2), with 5-2 in the hole and the 9 showing.
- Your opponent: shows a King and brought in (up: K♥).
- Pot before this street: 0.3 small bets (the antes).
- Pot size and the numbers: the solver groups pots by size: its numbers here are averages over every pot of up to 1.5 small bets at this point, so in exactly this pot they would be a little different.
The solver's numbers
| Play | How often (strategy) | Worth from here on (EV estimate, small bets) |
|---|---|---|
| Complete | 98.6% | +0.48 |
| Call the bring-in | 1.4% | +0.37 |
| Fold | 0.0% | +0.00 |
Measured on 74,764 play-outs of this situation. The highlighted row is the play that measured best.
Why, in plain words
Complete is worth about 0.11 small bets more than call the bring-in on average from here on: putting more chips in now pays with this hand.
One limit of the model: here, when you just call, the street ends; under real rules the bring-in could still complete. That makes calling look a little better than it is, so where calling and completing are close, completing is at least as good.
What popular strategy articles say
In full-ring games experienced players fold three-card starting hands that contain any card higher than an 8.
What strategy articles found by web search said on 2026-10-08 (PokerNews, Card Player, PokerStrategy, pokerskill and similar; mostly written between about 2007 and 2015, and mostly for full-ring games with several players in the pot).
Heads-up, the solver disagrees: it practically never chooses the recommended play (fold); its main play is complete (99%).
Measured value from here on: complete +0.48 small bets, fold 0 (folding).
The 3-handed solver doesn't follow the common advice either (it completes 79%).
Heads-up vs multiway
The numbers above are for heads-up play (two players). Most published advice is written for full-ring games with several players in the pot, so part of any difference can come from that.
3-handed (the bring-in shows a King and a third player showing a Queen is still to act behind you), PokerToast's 3-handed solver completes 79% of the time and just calls the bring-in 21%. So the heads-up answer holds with three players.
| Play | How often (strategy) | Worth from here on (EV estimate, small bets) |
|---|---|---|
| Complete | 78.9% | +0.67 |
| Call the bring-in | 21.1% | +0.66 |
| Fold | 0.0% | +0.00 |
From the 3-handed limit Razz solver, trained 25,358,751 iterations, exported 2026-10-05 18:07:23, 172,805 situations (version R33-25358751-20261005180723); measured on 8,024 play-outs.
How these numbers were made
From PokerToast's own heads-up limit Razz solver, trained 3,690,500 iterations, exported 2026-10-05 09:56:29, 35,165 situations (version R-3690500-20261005095629). It learned by self-play (Monte Carlo counterfactual regret minimization) for limit betting, at most four bets a street.
- How often is the solver's average strategy for this situation.
- Worth (EV) is an estimate measured after training by playing the situation out many times against the solver's own strategy: small bets won or lost from this decision on, compared with folding (0). It assumes an opponent who plays like the solver.
- Hand groups: the solver groups similar hands, so the numbers are for the group (here:
0|D250|1-|S|i), not only these exact cards. For each decision it looks at the street; your hand (a made 8-or-better low and how good it is, or how many cards you have toward an 8-low and how many of the low cards you need your opponent already shows); your opponent's board (how many high cards, "bricks", and their highest low card); the size of the pot; and the betting so far on this street. - Stakes: ante 0.15 small bets per player, bring-in 0.25, a completion or bet on 3rd and 4th street is 1 small bet, 2 from 5th street.
- Looked up when this page was built (2026-10-19); a newly trained solver changes the page and its version number.
Practice it
Your equity with these cards Today's Razz Daily Spot Open PokerToast
Part of: Razz 3rd street: completing, stealing and defending the bring-in
More Razz spots: Razz: complete 8-6-3 against a King bring-in? · Razz: defend the bring-in with A-2 under a King when a 5 completes? · Razz: slowplay A-2-3 or complete against a King bring-in? · Razz: complete A-2-3 against a low (6) bring-in? · Razz 4th street: you catch a King, the opponent shows 5-2 and bets. Call or fold? · Razz 7th street: call with a 9-low against K-8-4-2? · Razz: steal with a 7 up (K-Q in the hole) against a King bring-in? · Razz 5th street: 7-5-2 with two bricks against 6-4-3. Fold? · Razz: brought in with an 8 (6-3 under), a 7 completes. Call or re-raise? · Razz: defend the bring-in with 3-2 under a Queen when an 8 completes? · Razz 4th street: you catch a 2, the opponent catches a King. Bet or check? · Razz 4th street: you pair your 3, the opponent shows 6-4 and bets. Call or fold? · Razz 5th street: a made 9-8 low against 6-4-3 showing. Call, raise or fold? · Razz 6th street: a made 8-7 low against 7-5-4-2 showing. Call, raise or fold? · Razz: a three-card 9 (9-5-2) against an 8 bring-in. Complete, call or fold? · Razz: a paired start (7-3-3) against a King bring-in. Complete? · Razz 7th street: an 8-7 low against 9-8-4-2, checked to you. Bet or check? · Razz: you complete 7-4-2, the 6 bring-in re-raises. Cap, call or fold? · Razz 5th street: four to a 7 with one brick against 6-4-3 showing. Call? · Razz 4th street: you both catch bricks and you act first. Bet or check? · Razz: a rough three-card 8 (8-7-6) against a 7 bring-in. Complete? · Razz: defend the bring-in with 8-5 under a King when a 7 completes?