In early March, Matt Berkey was sent a photo of a player who was looking at the charts on his phone right during the hand.
The situation happened at the Bellagio at the $5/$10 table. One of the players turned to the floor, but was told that there was no problem since the chart was “not interactive”.
Matt Burkey strongly disagrees with this approach. In his podcast, he explained that this is a huge problem in live play. There is software that gives solvers in 30 seconds. Online, due to timebanks and the speed of the game, this is not enough, but offline, no one will pay attention at all, even if you spend 3 minutes in an important pot.
One of the top managers of MGM (ed. – a casino chain, which includes Bellagio ) Sean McCormack praised Matt that he raised this issue. He also assured readers that in all casinos of their network, the use of any applications during the game is strictly prohibited. He promised to convey this information to all employees. Off the tables, players can use anything.
It would seem that this question was settled, but a few days later a new storm arose on Twitter.
The developers of the Odin Poker solver have sent letters to their customers that their software will no longer have a 20-second delay in calculations. Some players immediately concluded that this turns the application into a real clue. The creator of Odin Poker, Rory Young, was hit with a stream of public anger so strong that he had to make excuses:
– When we launched Odin in July 2021, I turned to the developers of GTOWizard, RunItOnce (ed. – Phil Galfond and partners developed Vision, a solver for PLO ) and other solvers, so that we agreed and made the same delay in all of our products. NO ONE supported me. GTOWizard never had a delay at all (it's strange that they are not attacked by Berkey 🤔).
Despite this, I still decided to make a 20-second delay, which we kept for 2 years. Although our employees also need to feed their families. This had a huge impact on sales and user experience.
I tried to create a responsible attitude towards solvers within the industry, but it turned out that no one was interested. Including poker rooms. Only one major room showed interest, and for two years we helped them analyze hands to identify bots and tipsters.
And, for example, in partypoker I wrote about Ali [Imsirovic] and Jake [Schindler] long before Foxen, but they simply ignored me.
I still believe that bots are the biggest threat to online poker. My team is developing software that can automatically detect the use of hints. But large rooms are obliged to invest money themselves in order to make the use of prohibited software impossible.
“My development team is ex-Google and Facebook,” Rory Yang added in an interview with PokerNews. – If I wanted to cheat, we would turn Odin into a full-fledged prompter in a couple of weeks and sell it to everyone. Our software would simply read the information on your screen and tell you exactly what to do in real-time. But we never even had that thought. We are going in the exact opposite direction.
At the time of launch, Odin Poker made a lot of noise. Fedor Holtz actively participated in the advertising campaign, and Dominik Nitsche criticized the software for large errors in calculations.
Immediately after the cancellation of the delay, the termination of cooperation with Odin Poker was announced by Khale Burns, who also advertised the solver. And Fedor Holtz left the project even earlier. According to Rory Young, they broke up a couple of months ago and this has nothing to do with the reform. At that moment, the delay in Odin Poker was still in effect. Now Fedor is advertising GTOWizard, which never had such a delay at all.
Dominik Nitsche, as the developer of the DTO Poker GTO simulator, was also asked to comment:
– I don’t really understand what the DTO has to do with this dispute. Our simulations, in principle, cannot be used as a clue. But in this dispute, I am entirely on the side of Rory. In this story, the bad guy is definitely not him. The real problem with other solvers is GTOWizard, Vision and Ruse. I agree that a general forced delay would be the ideal solution. But you need to direct your anger at the real violators. Nobody knows GTOW developers at all. But it is well known that Vision has been owned by RIO for many years. But for some reason, no one pours shit on Phil Galfond for this. I wonder how it happened?
A little later, Dominik Nitsche explained what is happening in the world of solvers in general:
– Applications that provide real-time solver solutions have been around for a long time. Almost all of them work without delay. They are advertised, and sometimes developed by very famous and respected players.
Even among the active participants in this dispute, expressing sincere indignation, there are those who have been using wizard/ruse or vision for a long time. And all these programs can easily be turned into hints.
This is hypocrisy. The community should have spoken out against these programs long ago. But instead, many of us are happy to pay them for a subscription every month. Yes, they are comfortable, they can easily and quickly evaluate how you played the hand (although, by the way, this is a bad way of learning).
Odin hasn't broken any rules that other major players in the market haven't already broken. Don't give in to the herd instinct. We found ourselves in this situation only because the demand for such software is very high.
– It seems that in the past you have strongly criticized Odin for inaccuracies in the calculations. Has something changed recently? – asked in the comments.
– The quality of their simulations has remained the same, in this matter my opinion has not changed in any way.
From the side of the software owners, only Phil Galfond spoke out, who is always open to public debate. He explained that the developers of Vision themselves are interested in not using their software as a hint. They partnered with a number of major poker rooms to jointly track when Vision launches at the same time as a client. Galfond also listed a few more reasons why using their solver as a hint is not possible:
– We released Vision at the beginning of 2020. Naturally, the potential use of it as a hint was of great concern to us, since at that time we were actively working on RIO Poker. I myself have tried using Vision while playing play money on a single 6-max PLO table. Nothing happened. Preflop and the river, Vision is basically impossible to use as a hint. For 6-max there are calculations for no more than 10% of situations on the flop and turn (my rough estimate). Only one sizing is used on the flop. It’s a so-so hint given these limitations.
“In Vision, you can calculate any spot preflop, so here it is also suitable as a hint,” Alex Livingston corrected Phil. – Yes, you will play slowly and take time, but in difficult situations, it is quite suitable.
“Yeah, that’s right,” Phil agreed. – In addition, recently our software has become even faster. But it will still take about 15 seconds for each action, and this is when playing at one table.
“The above limitations do not prevent using Vision as a hint,” Nitsche also objected to Galfond. – Even a solver with limitations and simplified solutions will play better than any person. It has long been no secret that someone used Vision during the game and broke PLO with a win rate of 20bb/100 over 100k hands.
Yes, Vision is not a perfect clue. But for a strong player to get a huge advantage, it is enough to look at the approximate solution. I am sure that, if desired, I could successfully cheat with the help of Vision. For obvious reasons, I will not go into details.
“I never understood why you always speak in hints, instead of directly giving the name of the scammer,” one of the commentators was indignant.
In response, Dominik gave a link to the 2+2 topic, which tells about the atrocities of Zhuang Ruan.
“100k hands in PLO is not that big of a deal,” said Dan Cates.
“It depends which way you look,” Justin Bonomo objected. – Winning 20,000 big blinds is a huge amount at any limits. Does this mean that the person is playing with a win rate of 18-22 bb/100? Naturally, no.
“We can say one thing for sure – against an average field, the bots definitely play +EV,” summed up GingePoker.
– The main danger is a hint in the hands of a strong player, – Darius Bucinskas reminded themabout the obvious. – Such a cheater is almost impossible to distinguish from a top reg who plays honestly. But you can find ways to make the use of hints unprofitable without damaging poker. True, their implementation will require significant effort from the rooms and will be very expensive.
“That’s right,” Nitsche agreed. – A strong player is able to win on his own. But even minimal outside help will make him the most profitable player in the world. That is why even “partial” hints are a huge problem. Limited and sub-optimal sizings, inaccurate boards – incapable hands, none of this matters. Vision cannot be attributed to full-fledged prompts, but this does not eliminate the problem. A bad hint still plays better than a human, especially when a strong regular uses it.