Skip to content
jarvispoker
Tools

Heads-up push/fold charts

Short-stacked heads up, the game is just jam or fold. Pick an effective stack to see which hands the small blind moves all in with, and which hands the big blind calls with.

Effective stack

Small blind jams

58.9% of hands

Big blind calls the jam

38.3% of hands
AKoat 10bb effective
In the small blind
Always jam.
In the big blind, facing a jam
Always call.

The rule behind the chart

When the big blind is calling every hand, the small blind's decision has a closed form. Jamming an effective stack of S big blinds with equity e wins S(2e − 1) on average; folding gives up the half blind already posted. Jamming is better whenever:

e > 0.5 − 0.25 / S

At 2bb that bar is 37.5 percent equity. The weakest starting hand, 32 offsuit, runs at 32.6 percent against a random hand by this site's own equity engine, so it stays below the bar and folds. That is why even at 2bb the small blind is not jamming literally any two cards: by this formula, that would only start below 1.44bb.

Jam and call frequencies by stack

Share of all 1,326 starting hand combinations played, at each effective stack.

Effective stackSB jamsBB calls
2bb90.3%100.0%
3bb78.6%92.8%
4bb74.4%74.7%
5bb72.0%63.5%
6bb68.3%54.9%
7bb66.9%49.0%
8bb62.3%45.5%
9bb61.1%41.5%
10bb58.9%38.3%
11bb55.7%35.7%
12bb53.8%33.4%
13bb51.9%31.2%
14bb48.9%30.1%
15bb46.3%28.8%

Chart snapshots

The small blind's jam range at three common depths, readable without JavaScript. Salmon marks a jam; everything unshaded is a fold. Use the chart at the top of the page for the big blind's calling ranges and for every stack in between.

5bb · jams 72.0%

AA
AKs
AQs
AJs
ATs
A9s
A8s
A7s
A6s
A5s
A4s
A3s
A2s
AKo
KK
KQs
KJs
KTs
K9s
K8s
K7s
K6s
K5s
K4s
K3s
K2s
AQo
KQo
QQ
QJs
QTs
Q9s
Q8s
Q7s
Q6s
Q5s
Q4s
Q3s
Q2s
AJo
KJo
QJo
JJ
JTs
J9s
J8s
J7s
J6s
J5s
J4s
J3s
J2s
ATo
KTo
QTo
JTo
TT
T9s
T8s
T7s
T6s
T5s
T4s
T3s
T2s
A9o
K9o
Q9o
J9o
T9o
99
98s
97s
96s
95s
94s
93s
92s
A8o
K8o
Q8o
J8o
T8o
98o
88
87s
86s
85s
84s
83s
82s
A7o
K7o
Q7o
J7o
T7o
97o
87o
77
76s
75s
74s
73s
72s
A6o
K6o
Q6o
J6o
T6o
96o
86o
76o
66
65s
64s
63s
62s
A5o
K5o
Q5o
J5o
T5o
95o
85o
75o
65o
55
54s
53s
52s
A4o
K4o
Q4o
J4o
T4o
94o
84o
74o
64o
54o
44
43s
42s
A3o
K3o
Q3o
J3o
T3o
93o
83o
73o
63o
53o
43o
33
32s
A2o
K2o
Q2o
J2o
T2o
92o
82o
72o
62o
52o
42o
32o
22

10bb · jams 58.9%

AA
AKs
AQs
AJs
ATs
A9s
A8s
A7s
A6s
A5s
A4s
A3s
A2s
AKo
KK
KQs
KJs
KTs
K9s
K8s
K7s
K6s
K5s
K4s
K3s
K2s
AQo
KQo
QQ
QJs
QTs
Q9s
Q8s
Q7s
Q6s
Q5s
Q4s
Q3s
Q2s
AJo
KJo
QJo
JJ
JTs
J9s
J8s
J7s
J6s
J5s
J4s
J3s
J2s
ATo
KTo
QTo
JTo
TT
T9s
T8s
T7s
T6s
T5s
T4s
T3s
T2s
A9o
K9o
Q9o
J9o
T9o
99
98s
97s
96s
95s
94s
93s
92s
A8o
K8o
Q8o
J8o
T8o
98o
88
87s
86s
85s
84s
83s
82s
A7o
K7o
Q7o
J7o
T7o
97o
87o
77
76s
75s
74s
73s
72s
A6o
K6o
Q6o
J6o
T6o
96o
86o
76o
66
65s
64s
63s
62s
A5o
K5o
Q5o
J5o
T5o
95o
85o
75o
65o
55
54s
53s
52s
A4o
K4o
Q4o
J4o
T4o
94o
84o
74o
64o
54o
44
43s
42s
A3o
K3o
Q3o
J3o
T3o
93o
83o
73o
63o
53o
43o
33
32s
A2o
K2o
Q2o
J2o
T2o
92o
82o
72o
62o
52o
42o
32o
22

15bb · jams 46.3%

AA
AKs
AQs
AJs
ATs
A9s
A8s
A7s
A6s
A5s
A4s
A3s
A2s
AKo
KK
KQs
KJs
KTs
K9s
K8s
K7s
K6s
K5s
K4s
K3s
K2s
AQo
KQo
QQ
QJs
QTs
Q9s
Q8s
Q7s
Q6s
Q5s
Q4s
Q3s
Q2s
AJo
KJo
QJo
JJ
JTs
J9s
J8s
J7s
J6s
J5s
J4s
J3s
J2s
ATo
KTo
QTo
JTo
TT
T9s
T8s
T7s
T6s
T5s
T4s
T3s
T2s
A9o
K9o
Q9o
J9o
T9o
99
98s
97s
96s
95s
94s
93s
92s
A8o
K8o
Q8o
J8o
T8o
98o
88
87s
86s
85s
84s
83s
82s
A7o
K7o
Q7o
J7o
T7o
97o
87o
77
76s
75s
74s
73s
72s
A6o
K6o
Q6o
J6o
T6o
96o
86o
76o
66
65s
64s
63s
62s
A5o
K5o
Q5o
J5o
T5o
95o
85o
75o
65o
55
54s
53s
52s
A4o
K4o
Q4o
J4o
T4o
94o
84o
74o
64o
54o
44
43s
42s
A3o
K3o
Q3o
J3o
T3o
93o
83o
73o
63o
53o
43o
33
32s
A2o
K2o
Q2o
J2o
T2o
92o
82o
72o
62o
52o
42o
32o
22

How these charts were solved

These ranges are not copied from a published chart. They were solved here, from this site's own equity engine: every one of the 14,365 distinct starting-hand matchups was evaluated, then a fictitious-play solver was run at each stack until both sides stopped being able to improve.

Convergence is measured, not assumed. A separate best-response check, built on a general game-tree engine that predates this feature and knows nothing about how the ranges were produced, scores every chart at under 0.001 big blinds of exploitability, and each of the 4,732 individual jam-or-call decisions is re-checked against its own expected value. Both bounds are enforced by the verification script rather than asserted here. The two checks do share the payoff arithmetic and the equity inputs with the solver, so they confirm the solve converged, not that the underlying model is a perfect picture of poker.

Which brings up the model's one real simplification: matchup equities use a single representative suit combination per hand pair rather than averaging over every suit layout. Measured across eight archetype matchups, that sits between 0.02 and 0.71 percentage points away from the true suit average. Equity errors of that size only change a decision for hands already sitting near the jam-or-fold boundary, so treat the fringes of these ranges as close calls rather than exact cutoffs.

Common questions

What is a push/fold chart?
A push/fold chart is the solved strategy for a heads-up pot where stacks are short enough that the only sensible options are moving all in or folding. It has two halves: the hands the small blind jams, and the hands the big blind calls that jam with. At these stack depths there is no postflop play to speak of, so the whole game collapses into those two decisions and can be solved exactly.
When should you use push/fold?
Push/fold applies once the effective stack is short enough that raising to anything less than all in leaves you committed anyway. The usual rule of thumb among players is about 15 big blinds and below, which is why these charts stop there; deeper than that, having room to raise and still fold gives up less than jamming, though that comparison is a matter of convention here rather than something these charts solve. At 15bb the small blind is jamming 46.3% of hands; by 5bb that widens to 72.0%, and by 2bb it reaches 90.3%.
Which hands does the small blind jam heads up?
It depends entirely on the stack. At 10bb effective the small blind jams 58.9% of all hands and the big blind calls with 38.3%. The ranges widen as stacks get shorter because the blinds are worth relatively more: winning them uncontested matters more, and the risk of being called costs less. Use the chart above to read any specific hand at any stack from 2bb to 15bb.
Why does the big blind call wider than the small blind jams at 2bb?
Because the big blind is getting overwhelming pot odds. Facing a 2bb jam it has already posted 1bb and is calling 1bb more to win a 4bb pot, so it only needs about 25 percent equity and every hand clears that bar. The small blind, risking its whole stack to win 1bb, needs more than 37.5 percent equity and so folds its worst hands. The two ranges cross over at 5bb, above which the big blind is the tighter of the two.

Push/fold tells you whether to move in. The equity calculator shows how any two hands run against each other, and the preflop charts cover the deeper-stacked opens.

Study and review tool. Not for use during live online play.