Replying to @mackie_campbell @KmartPoker
@mackie_campbell yes, especially vs the shallow stack
Hold’em has about 1,300 starting hands. PLO has over 270,000.
That’s exactly why PLO is so soft: everything looks playable, so almost nobody folds preflop.
@tombos21 shows you how to judge any PLO hand by its parts (suits, connectedness, high cards, and harmony) and build a real preflop edge the smart way.
Full video 🔗
youtube.com/watch?v=uSWTUDln…
Live players use all kinds of sizings. Overbets, min raises, weird half pots. Most players react on instinct. The best players know exactly how to counter those sizings.
Join @scrimitzu as he builds your live poker betsize compendium, breaking down how to read, respond to and exploit the full range of sizings you will encounter at the live cash game table.
Jump in at 6 PM CEST: app.gtowizard.com
Replying to @NerdO_AllTrades
Great question! Our paper focuses on making ICM computations more efficient, but ICM itself is a model and doesn’t necessarily capture all the dynamics of real tournaments.
Our paper cites this study, for example, which investigates how well the ICM holds up empirically: arxiv.org/abs/2506.00180
This reply from @tombos21 is also worth checking out: xitter.cf/tombos21/status/211023…
Ultra Early Bird ends tomorrow.
Lock in $229/month on Ultra annual and save $720 over the year. From October 16, the price will increase to $289/month.
Every spot from your game, ICM included, solved in seconds. 7-day refund guarantee, for any reason.
app.gtowizard.com/subscripti…
Replying to @tombos21
@tombos21 explains our "Near-Exact Computation of Independent Chip Model Placement Probabilities for Thousands of Players" paper in simple terms 👇
Replying to @tombos21
The solution is brilliant.
Imagine a big poker tournament. Every player forms a line and exchanges their chips for raffle tickets.
More chips = more raffle tickets.
Whenever your ticket is drawn, you move to the front of the line. Your ticket goes back into the raffle drum, and they draw another ticket, repeating indefinitely.
Big stacks have more tickets, so they get called to the front more often. If you had twice as many chips as buddy, you are twice as likely to be ahead of them in line.
Remarkably, the probability of occupying any position in this line is the same as your ICM finishing probability. Your chance of being 5th in line is your chance of finishing 5th in the tournament.
Now, you could simulate this endless raffle purgatory to slowly approximate tournament equities. But we can do better.
Triton Jeju II. One of the most prestigious stops on the high stakes circuit and some of the most complex tournament spots you will ever see.
Kevin reviews the most interesting spots in this years Triton Jeju II, unpacking the current live table meta and the solver insights behind them.
Jump in at 6 PM CEST: app.gtowizard.com
Change the stacks or villain tendencies and the A9s answer changes too.
With Ultra, you can solve your own final table spots with the exact stacks and players you faced.
Ultra Early Bird ends in 48 hours, get it while it’s still available: app.gtowizard.com/subscripti…
Shoving A9s isn’t a punt, but taking the less agressive line can often give us an extra edge over opponents.
A raise keeps our options open against both blinds:
🔹 vs the SB (covers us): we can easily fold A9s to a jam and stay in the tournament.
🔹 vs the BB (we cover): we keep the pressure on and can call off a jam.
Today we are revealing how GTO Wizard AI computes ICM for large tournament fields, and publishing our proprietary algorithm, free, for everyone.
Traditional ICM calculators top out at a few dozen players. In 2024, we showed that ICM for thousands of players can be computed in seconds, a problem academia still considered intractable.
We are sharing it because we believe in contributing to research, and because we want others to build on it, in poker and well beyond it. The same math applies to ranking problems in many other industries.
Behind GTO Wizard is a team of researchers and scientists who built the state-of-the-art poker AI and work closely with @MIT and other leading universities around the world, from guest lectures to joint research.
This problem was cracked by the brilliant Wataru Inariba (@b_inary). Wataru, thank you for your outstanding work.
Read the full paper on arXiv 👇 arxiv.org/abs/2610.06028
GTOWizard retweeted
Would you fire three barrels with Queen-high playing for $10M? With a $4M pay jump on the line heads-up in the WSOP Main Event, Lauri Saaskilahti put the heat on Lucas Jumalon with Q-9o across an A-T-7-K-A board.
We ran the spot through @GTOWizard.👇
pokernews.com/strategy/would…
Replying to @runjep
Exactly! The pay jumps matter a bit more than the bounties here!
With our latest Ultra tier release, you can adjust any part of your opponents’ ranges or your own and see how those changes affect the solution.
Alternatively, you could go for a cheaper tier with presolved solutions in our Premium tier to find similar spots to the one you’re playing!










