Recompia
Pokémon Pokopia (Switch 2) recompiled to run natively on Windows with Linux/Handheld support. Multiplayer included, modding, feature enhancements and more! Join the discord community as well
翻訳待ち
Hookah Pookah poker calculator with exact enumeration and honest Monte Carlo error bars. Hold'em, Short Deck, PLO4 and PLO5 equities, weighted ranges, hand chances and every next card, all in your bro
翻訳待ち
ゲームTypeScript · CSSGitHub @KarthikSubramanian07オンライン所有者未確認
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Exact-enumeration and Monte Carlo poker odds engine for Hold'em, Short Deck and PLO: equities, ranges, hand-type chances and next-card outs, all in your browser.
翻訳待ち
Loading the full README. The saved excerpt is shown below.
What are the chances? Poker odds that tell you exactly, or tell you exactly how sure they are.
Hookah Pookah is a poker equity calculator for Texas Hold'em, Short Deck (Triton and classic rules), PLO4 and PLO5. Give it hands, partial hands, random hands or weighted ranges, a board and dead cards, and it answers:Read the full README ↗What are the chances? Poker odds that tell you exactly, or tell you exactly how sure they are.
poker · equity-calculator · texas-holdem · short-deck · omaha · monte-carlo · web-worker · cloudflare-pages
Hookah Pookah is a poker equity calculator for Texas Hold'em, Short Deck (Triton and classic rules), PLO4 and PLO5. Give it hands, partial hands, random hands or weighted ranges, a board and dead cards, and it answers:
It is exact whenever the math allows. Heads-up preflop enumerates all 1,712,304 boards in about 35 ms. Range against range runs exactly in around 100 ms. When a spot is too big (say, five random opponents preflop), it switches to Monte Carlo on every CPU core and keeps sampling until the 95% confidence interval is as tight as you asked for. That interval is always on screen, and the result never shows more digits than it can defend.
Everything runs in your browser. No server, no account, no ads.
| Route | Game |
|---|---|
/holdem | Texas Hold'em |
/short-deck | Short Deck, Triton or classic rules |
/plo | Pot Limit Omaha (4 cards) |
/plo5 | 5 Card Omaha |
Spots live in the URL, so every calculation is shareable:
/holdem?p=AhKh&p=r:QQ+,AKs&b=Qh7h2c
The answer updates as you go: your chance to win, whether you are a favourite or an underdog, what hand you are likely to finish with, and (once there is a flop) which next cards help you. Open Should I call? to compare the price of a call with your chances.
Switch on Advanced for opponent ranges, the 13x13 range grid, dead cards, precision settings and the full hand table. Shared links that use those features open in Advanced automatically.
AhKh vs QQ+, AKs vs random on Ks7h2d dead 2ca h. Backspace clears it.| Notation | Meaning |
|---|---|
AA AKs AKo AK | a hand class (AK is suited and offsuit) |
QQ+ ATs+ KTo+ | pairs upward, or kickers up to one below the top card |
22-55 A2s-A5s K9o-KJo | pair ladders and shared top card ladders |
T9s-54s | constant gap ladders |
AhKh | one exact combo |
15% 10%-25% | top hands by preflop all-in equity (vs 3 random hands by default, vs 1 selectable) |
random any | every combo |
AA:0.5 KK:25% [50]AQs, AJs[/50] | weights |
When a combo appears twice, the later token wins, so QQ+, AA:0.5 means what it says.
Accuracy is the whole product, so it is checked against independent implementations rather than against itself.
| Check | Result |
|---|---|
| All 133,784,560 seven-card hands, category counts | exact match to the published table; 4,824 distinct ranks |
| Short deck seven-card counts, Triton and classic | exact match to an independent C enumerator; 762 and 752 ranks |
| PokerHandEvaluator test corpus: 5, 6, 7 card, PLO4, PLO5 (6.6M hands) | 0 inconsistencies, 0 ordering violations |
| Heads-up and 3-way preflop equities vs OMPEval exhaustive enumeration | identical to 4 decimals |
| 35 random range vs range matchups vs OMPEval | max difference 5e-7 percentage points |
| Randomised spots (ranges, weights, partial hands, dead cards, all variants) vs a naive brute-force oracle | identical to 1e-14 |
| Monte Carlo 95% intervals over thousands of seeded runs | cover the exact answer 94.8% to 95.6% of the time |
A finding worth passing on: short deck equities quoted around the web (AsAh vs KsKh at 75.02%) use classic rules. Under Triton rules, where trips beat a straight, it is 74.96%. Hookah Pookah supports both and says which one it is using.
The unit suite runs these oracles on every push; pnpm test:stress runs the exhaustive and long-running versions (weekly in CI).
This is the beginning of the README. Read the rest at the original. Read the full README ↗