When we posed the question “Is poker dead?” in a podcast episode, one commenter wryly noted that a poker training site would inevitably answer that question in the negative. A very fair point. So why did we pose this question initially, and then elaborate on it in this article?
There are multiple reasons. The first is simply the doom-posting one sees on some social media platforms describing the alleged terminal state of poker in 2025. Bot rings, RTAs, collusion, and the antipathy of online sites and B&M cardrooms towards poker players are all invoked.
We chose to address these concerns because, to varying degrees, they are very much real. The poker environment has changed over the last decade, and in the U.S. at least is vastly different from the golden age before Black Friday.
In this article, we will examine how poker players can both protect themselves and profit in the current poker environment. To do so, we will first provide a brief historical context. If you are not interested in poker history, you can skip the following section, or check out this video for the contemporary state of play.
While the origins of poker remain controversial, the game established itself in the U.S. in the 1820s, having crossed the Atlantic with European settlers. Poker then steamed up the Mississippi on riverboats while spreading west by land into saloons and parlors.
The ubiquitous presence of poker in Hollywood westerns cemented its image as an American game, whereas in the Old World poker had stiffer competition from other card games such as bridge. In its early American days, poker was known as the cheating game, to the extent that card sense often came second to the ability to cheat more effectively than one’s opponents.
John Scarne’s 1949 classic Scarne On Cards injected some legitimacy into poker as a true strategy game, while also including detailed information on how to spot poker cheats. It was not until the 1970s, however, that the first flurry of poker strategy books appeared, including the late Doyle Brunson’s classic Super System. Such texts are the forerunners of modern training sites such as Red Chip Poker.
The 1970s also saw the annual World Series Of Poker introduced to Las Vegas, and the decision that no-limit hold’em would be the game that determined the annual poker world champion. By the late 1990s, however, poker in both North America and Europe had suffered something of a decline. While still a popular home game, casino cardrooms experienced a contraction. In the Midwest heartland of U.S. poker, the depressing sight of poker tables being replaced by slot machines was common.
Within five years, this trend had completely reversed. Public poker rooms began springing up like mushrooms after rain, while online poker exploded into a multi-billion dollar industry.
We have argued elsewhere that attributing this explosion to the “Moneymaker Effect” is little more than a fairytale. Hole card cameras, televised poker, and the advent of easily-accessible online poker had a far greater impact than a single tournament. What is clear is that the 2000s constituted a golden age of poker, particularly in the U.S.
And in the same country in 2010, it all came crashing down. Poker was dealt a body blow thanks to Black Friday.
The point of this rather lengthy preamble is that, like many other hobbies and competitive activities, the health of poker rises and falls due to capricious external influences. It will not die in the foreseeable future, and in 2025 poker is far from dead.
However, the modern game does present novel challenges, both because of the increasing skill of players, and the sophisticated cheating methods available to the unscrupulous, both of which have been accelerated by mathematical advances in poker theory. It is to this topic that we now turn.
Much of the current concern about the demise of poker stems from the advent of poker solvers. The first shockwave came in 2015, when the Cepheus Poker Project announced that they had solved heads-up, limit hold’em. Things calmed down once the community remembered that basically nobody plays heads-up LHE.
The fear that the success of Cepheus would extend to no-limit hold’em gained some traction in 2017 when the poker bot Libratus won a match against four human players. It was argued that further development would only make such bots into bigger crushers, and poker was once again read the last rites.
Concerns about the impact of bots and solvers are usefully separated into two categories.
The first is that the availability of solvers that provide “the answer” to any poker situation means that players will become so strong that nobody can make any money. Apparently the idea is that we will all sit around reproducing perfect GTO lines, so that everyone loses to the poker sites and cardrooms.
If you have played any public poker at all, you will realize at once that this position is rather silly. A knowledge of poker fundamentals at the level provided by our CORE program, will enable you to sit in any low-stakes game and witness multiple mistakes made by your opponents in a single hand.
The simple fact is that the vast majority of poker players enjoy the game as a recreational hobby. Their “study” may be limited to watching the occasional free videos or live stream on YouTube. If you choose to study poker seriously, you will beat these guys because you know more than them.
The second concern mostly relates to online poker. An opponent playing from their computer can potentially use RTAs (Real Time Assistance) or other pre-solved GTO solutions in order to play “perfect” poker. What chance does an honest player have against such unscrupulous devils?
This is certainly not a problem that we wish to minimize. Within the online poker tournament community, for example, there is considerable concern that players might use RTAs to solve preflop situations with significant ICM effects. Most dedicated tournament players have a good grasp of baseline push-fold and 3-bet/shove ranges, but modifying those ranges with variable stack depths and ICM effects requires subtle interpolations. Someone armed with a ICM calculator would have a significant advantage in late-game spots, while also eradicating the skill expression of those who have spent thousands of hours studying such scenarios.
That said, and as pointed out in the video of the previous section, there are also technical reasons to have hope that even RTAs and bots are far from infallible.
While heads-up LHE has been solved, NLHE has not. The fact one can bet anything up to the effective stack makes the game far more complex to solve, so that for the foreseeable future no general solution to the game is anticipated. In multiway pots, it is possible that no such general solution even exists.
Another consideration is that, once a poker hand has deviated from GTO, the most profit that one can extract from the hand is not a node-locked equilibrium solution. In other words, a solver is very good at equilibrium calculations, but fails when tasked with finding lines of interest to human players: those that make the most money.
What about bots, though? Surely their presence in an online game will turn any human player into a loser? Oh you sweet, summer child. Bots can be a source of human profit if you know how to handle them, as we now explain.
At first glance, the poker bot appears to be a formidable opponent. Potentially incorporating GTO solutions, it does not need to eat nor sleep. It never goes on tilt; it never misclicks. The bot, as the name suggests, is a perfect, robotic poker player. And yet with the right approach, a poker bot can be crushed like any other player. Here is coach w34z3l to explain how.
The first stage in your anti-bot campaign is to learn how to recognize them. As coach w34z3l notes, not all bots are created equal. And given that the use of bots is a ToS violation, your best approach when you find an effective silicon opponent might be to report them to the site’s game integrity team.
Note that identifying bots successfully requires more than running their hand histories through a “fair play” algorithm that looks for deviations from GTO play. Many bots do not rely on GTO as their primary strategy. Profile bots, for example, use a reactive algorithm based on the preflop stats of the opponent. Pool exploit bots employ the results of mass data analysis on a site’s player pool, and take lines that exploit those population tendencies.
The RTA bot uses a GTO-based algorithm, and is thus the one that is easily outed by comparing its play to GTO solutions. All of these bot types can become massively more powerful if they are coordinated in bot rings. The ability to share hole card information alone means a human sitting at a table with five communicating bots is on a hiding to nothing. This is a situation to avoid at all costs.
We noted above that one advantage a bot has over a human is that it can play twenty-four hours a day. For the greedy bot operator, this is also the bot’s Achilles’ heel. The easiest way to detect a bot is to demonstrate that it plays sessions that are beyond the realm of possibility for a human player.
Determining what kind of a bot one is facing, and thus how to exploit it, follows the same principles as one uses for profiling human opponents. Indeed profiling bots is easier, since their algorithmic nature produces a far more consistent playstyle than most humans achieve. And just like profiling humans, once we know a bot’s tendencies, we can craft counter-exploits to profit from them.
But what about the RTA bots and their GTO lines? Is not GTO poker by its very nature unexploitable? How can we make money from them?
We exploit GTO bots in the same way we exploit humans attempting to reproduce GTO lines.
With bots safely packed back in their boxes, let us turn to the final supposed sign that poker is on its death bed. We have already dismissed the notion that solvers will produce a poker population entirely comprised of GTO-perfect players, but surely at the highest stakes such players dominate. And because GTO is “unbeatable,” this will cap what any human can achieve in the game.
More good news incoming. The idea that GTO is unbeatable reflects a common misunderstanding. Like any other strategy, a human player can develop effective counter-measures.
We noted earlier in this article that a full GTO solution to no-limit hold’em does not exist, and that one key reason for this is that NLHE by definition allows a huge number of bet sizes in any given scenario. This unavoidable fact provides one of our sharpest weapons in fighting “GTO opponents,” whether they be humans or bots.
The first opportunity we have to implement this general tactic is preflop. If you are familiar with our preflop range charts featured on the GTO ranges app, you will be aware that the open-raise sizing is specified. We use 4x for live play and 2.5x for online 6-max. This is a common choice and is a likely default for both bots and players who have memorized preflop ranges.
By moving our opening size away from these canonical values, we immediately take our opponents out of their comfort zone into a realm where they may make a mistake. This is particularly the case if we have solid ranges for our chosen raise size. One common scenario in which this approach can yield dividends in messing up the big blind’s defense frequency.
Postflop play provides even more possibilities for setting the cat amongst the pigeons. As emphasized in our PRO course “Mix It Up,” there are many low-frequency GTO lines that are strategically sound, but that will be outside the scope of our opponent’s knowledge or programming.
Using such lines, particularly against skilled opponents and RTA bots, will place them in spots where they simply do not know what to do. Again, if we have analyzed these lines carefully, we will generate a significant profit against opponents who are unfamiliar with them.
In addition to being profitable, these examples illustrate that poker is still a thinking game, where creativity and experimentation continue to play key roles. Notably, we can create winning situations against serious students of GTO play and their silicon cousins. Poker in 2025 is very much alive.
Cheating has been a constant feature of poker throughout its recorded history. We would suggest that the current state of play is far less troubling than back in the 19th century when poker was “the cheating game.” That said, it is important to realize that the modern-day cheat has an array of technological tools they can employ, particularly in the online realm.
If you have played a couple of hundred hours of poker, it is likely that you have been cheated. That may be someone shorting the pot, a couple signaling to each other, or an online player taking advantage of pre-solved solutions. The above videos and discussion provide tips on how you can best protect yourself. Even with anti-cheat measures enforced in live and online play, ultimately this responsibility falls on you.
In terms of the impact that solvers might have on the health of poker, we feel the game of chess provides an instructive parallel.
It has been over two decades since a human has been able to beat the best chess engines and self-learning AIs in match-play. If the current world chess champion played a hundred games against the latest iteration of Stockfish or Leela, it is unlikely they would manage a single draw. Has this led to the death of chess?
Quite the reverse. Chess has been enriched by the insights provided by The Machines. They have produced new approaches to the game, and proven to be a remarkably useful learning tool for players at all levels. Rather than sounding the death knell of chess, engines and AIs have reinvigorated it.
Whatever your current interest in poker may be, from weekend recreational to professional, there is no reason to suppose the great game is breathing its last. Poker has survived hostile legislation, cheating scandals, corrupt online sites, and competition from online games and other pursuits.
Ultimately, it seems inevitable that poker will survive, simply because it is such an engaging and enjoyable game.
This website uses cookies.