Feedback based on my 77%+ vipip Sunday Session yesterday
my 2nd Video
25 million Hands sample on bots GG poker ACR mostly
I am still awaiting some form of an answer to my first attempt btw
Expanded Analysis my 1st vid from a ~month ago
The transcript reveals a deep frustration with the structural and cultural issues plaguing online poker platforms, especially Coin Poker. The speaker’s discussion underscores the critical role that trust and transparency play in sustaining an online gaming community. Being banned without explanation from key community channels not only silences honest voices but also signals a lack of accountability from platform moderators.
The financial model of these poker platforms comes under scrutiny, highlighting the excessive rake and the misleading promotional structures that deceive players. The speaker’s reference to losing approximately 17-19 big blinds per 100 hands just to rake at lower stakes reveals how unprofitable the environment is for the average player. This is compounded by the exploitation of bonuses through bot accounts, which unfairly siphon value away from genuine players.
Collusion and bot use at these stakes reflect a broader problem of cheating and unfair advantage that is difficult to police. The speaker’s observation that regular players behave uniformly and often sit out together suggests coordinated behavior that further tips the scales against honest competition.
The speaker’s proposal for a rake-free environment is both practical and visionary. Removing rake at low stakes would lower barriers to entry, increase player retention, and foster a healthy competitive scene akin to esports. This could also incentivize community-driven staking models, where players collectively invest in talent, thus creating deeper engagement and loyalty.
The lament over the loss of poker’s esports spirit points to a missed opportunity for the industry. Poker once had a vibrant crossover with gaming and esports culture, but outdated management, greed, and poor promotion have severed this connection. The mention of Warcraft and Starcraft commentators highlights how esports models could be adapted to poker, enriching the spectator experience and attracting new audiences.
Finally, the speaker’s suggestion to leverage AI for marketing and game management introduces an important avenue for innovation. Intelligent systems could not only detect cheating more effectively but also design better promotions and personalize player experiences. This could revitalize poker platforms struggling with player retention and trust issues.
In sum, the video transcript offers a comprehensive critique of the current online poker ecosystem’s failings and lays out a roadmap for reform centered on fairness, transparency, community engagement, and innovation. The speaker’s personal experiences bring authenticity and urgency to the call for change, emphasizing that without structural reforms, the game risks continued decline.
Conclusion
The speaker’s detailed critique of Coin Poker and the broader online poker ecosystem reveals systemic issues around fairness, transparency, and community management. Excessive rake, bot exploitation, unfair promotions, and poor moderation have alienated honest players and damaged trust. To revive online poker’s appeal, the speaker advocates for a rake-free environment at low stakes, better community-driven incentives, esports-style competition, and smarter use of AI in marketing and game management. Without addressing these fundamental problems, online poker risks losing both its player base and its cultural relevance.
charon — 09:06
BRUCHWASSERLÄUFER¿ in pages reached
[09:14, 05/11/2025] Urs-Li: Excellent question that gets to the heart of the evolution of poker in the 21st century.
The answer is a nuanced "yes and no," depending on the format of the game. Let's break it down.
Part 1: Has Poker Been "Solved" by AI?
The term "solved" in game theory means finding a perfect, unbeatable strategy. For poker, this is called Game Theory Optimal (GTO). A GTO strategy is one that cannot be exploited; against any opponent's strategy, it will, at worst, break even over the long run.
Here's how AI has fared in different poker formats:
1. The "Solved" Part: Heads-Up No-Limit Hold'em (HUNLHE)
Yes, for all practical purposes, two-player No-Limit Hold'em has been solved.
The Breakthrough: The major milestone was an AI named Libratus, developed at Carnegie Mellon University. In 2017, it played against four of the world's top heads-up specialists over 120,000 hands and beat them decisively.
How it Works: Libratus developed a GTO strategy that is so well-balanced and complex that even the best human players could not find a way to exploit it. The AI is simply better at randomizing its bluffs, value bets, and check/raises with perfect frequencies, making it impossible to read or get an edge against.
Conclusion: No human can beat these top-tier AIs in heads-up play over a large sample size. The AI is provably superior.
2. The "Not Solved, But Superhuman" Part: Multi-Player Poker (6-Max and Full Ring)
No, these formats are not mathematically solved, but AI has achieved superhuman performance.
The Complexity Problem: As you add more players, the game tree (the map of all possible actions and outcomes) explodes in size. A 6-player game is astronomically more complex than a 2-player game. Finding a true GTO solution is currently computationally infeasible.
The Breakthrough: An AI named Pluribus (from Facebook AI and Carnegie Mellon) was the first to defeat elite professionals in six-player No-Limit Hold'em in 2019.
How it Works: Pluribus doesn't play a perfect GTO strategy. Instead, it uses a highly advanced approximation of GTO. It learned by playing trillions of hands against itself, figuring out what works without being given any human strategic input. It's so good that it's demonstrably better than the best humans.
Conclusion: While not "solved" in the pure mathematical sense, AI plays at a level that humans cannot consistently beat. The same applies to other complex formats like Tournaments (MTTs) and Pot-Limit Omaha (PLO), where AI is extremely strong but the games are even further from being truly solved.
Summary for Part 1: From a human perspective, the argument "poker is more or less solved" is fair. The best AI strategies are so far beyond human capability that the top level of the game is now defined by how closely a human can replicate the GTO strategies pioneered by AI.
Part 2: How Do Poker Sites Prevent AI Cheating?
This is a massive, ongoing arms race. Online poker sites invest millions in their game integrity and security teams because AI "bots" and Real-Time Assistance (RTA)—software that advises a human player during a hand—threaten their entire business model.
Here are their primary methods of detection:
1. Behavioral and Data Analysis (The Biggest Weapon)
Sites collect a staggering amount of data on every player. They use machine learning to find patterns that are not human.
Session Patterns: Humans get tired. A bot can play 24 hours a day, 7 days a week, without any decline in performance. Inhumanly long or consistent session times are a huge red flag.
Decision Timings: A human will take longer for a tough river decision than for a standard pre-flop fold. A bot might take a suspiciously consistent amount of time for every action (e.g., 1.2 seconds), regardless of complexity.
Mouse Movements: This is a clever one. A human's mouse cursor might hover over the "call" button, then move to the "fold" button, then back again. A bot's cursor might snap instantly to the intended button with no wasted movement, or it might not move at all if it's injecting commands directly. Some sites even require you to click a button that moves around to thwart simple bots.
Inhuman Perfection: A bot playing a GTO strategy will make mathematically perfect plays with perfect randomization. While top pros are very good, nobody is perfect. An account that never makes a questionable play over tens of thousands of hands is highly suspicious.
2. Technical Detection
The poker client software itself is a tool for security.
Scanning for Prohibited Software: The client can scan the user's active processes for known cheating programs, screen-sharing software, or virtual machines (which are often used to hide bots). This is stated in their terms of service.
Analyzing Input Methods: They can detect if gameplay actions are coming from a standard mouse/keyboard or from a software-based input.
3. Statistical and Hand History Analysis
After the fact, security teams can run deep analysis on a player's hand history.
Solver-Like Play: They can compare a suspect's decisions in key spots against the known GTO outputs from popular poker solvers. If a player's actions match the solver's "perfect" play too frequently, especially in complex situations, it's strong evidence of RTA.
Impossible Win-Rates: Every poker format has a statistical ceiling on a player's win rate. An account that crushes games at a rate that is statistically impossible over a large sample size will be immediately flagged for review.
4. The Human Element
Player Reports: The poker community is very good at policing itself. When skilled players suspect they are playing against a bot or RTA user, they report the account. A flood of reports from credible players will trigger a high-priority investigation.
"Turing Tests": In some cases, a security team member may try to chat with a suspected bot to see if they get a human response. They may also use CAPTCHAs or other tests during gameplay.
This is a constant battle. As detection methods improve, so do the bots and RTA tools. The poker sites' goal is to make cheating so difficult and the risk of being caught (and having your funds confiscated) so high that it isn't worth it.
[09:14, 05/11/2025] Urs-Li: problem is that legally it never had consequences like jail time for the cheaters/bot rings because poker/gambling is prohibited in itself
Feel free to reach out to us with any additional questions, or visit our FAQ page here: https://coinpoker.com/#faq
Regards,
Matheus
CoinPoker