February 06, 2015

Online poker is rigged

We all hear the arguments in cardrooms, on forums, and especially after a suckout occurs. The standard rebuttal is that these sites are making plenty of money, why would they risk losing everything if they juice their flops? There are a few things wrong with this rebuttal.

First, these online poker sites were making plenty of money. But how much is plenty? With increasing competition and a move towards publicly traded companies (Party Poker IPO, Empire Poker IPO, Paradise Poker takeover by SportingBet, and the pending Pacific Poker IPO), the pie is being divided up into smaller pieces while to pressure to continue growth rises. When all you have to do is spend your cash, it is easy to stay fair. Why rock the boat? But when the industry is shattering into a million tiny poker sites and your board of directors is holding you accountable for a few billion of investors’ money, you might try to squeak out a few pennies.

Second, are they really risking everything if they are certain they will not get caught? If Party Poker and Poker Stars can find subtle ways to juice their flops, it would be damn near undetectable. Over a few billion hands the difference of a fraction of a cent of rake per hand would make such deception worthwhile. So the two big questions are why and how.

Why Would Online Poker Be Rigged?

Why is simple. Money! As mentioned earlier, external and internal pressures are increasingly weighing on poker sites. Boards of directors and shareholders need to see returns, and customers need to be retained. Think of things from the perspective on an internet poker site who is considering their cash flow. They have a few problems facing them.

First, their players are going broke. Bad players will supposedly lose over the long run. Once they are gone, the poker sites are left with fewer players and therefore less rake. This must be a huge problem for poker sites. If they can tip the scales in favor of the fish occasionally, they can avoid detection. Most “lucky runs” are well within the realm of statistical probability, so they can give a loser a break every now and then to rejuvenate his rake producing bankroll.

Second, if overall growth stalls, then the only way to grow revenue is to squeeze more money per hand out of the poor saps at the tables. This is already happening. Party Poker has been peaking at 70,000 players for over a year. Their astronomical growth took them from 40,000 at peak hours to 70,000 in about a year. It has not moved from that point. Party Poker’s response was to rewrite their rake requirements to rake up to $3 from tables starting with 6 players rather than 7. This has had a huge impact on the rake generated at the 6-max tables, which have proliferated in recent months. They obviously need to squeeze their players, why not cheat them to squeeze out an extra dime per hand?

Finally, there is an intangible benefit to creating action. Not only does it produce more rake, but it gives the gambling type a feeling of exhiliration. Players feel like they were involved in a significant pot and boast of their wins and whine about their losses after the hand. No one brags up a 2 big bet win. Everyone tells the story of the 30 big bet pot they scooped.

It is obvious, then, that there is plenty of motivation to cheat. However, rigging online poker is a dangerous game. If they are accused and it is proven that there is a discepancy in the randomness of the cards, that internet poker site is done. They will be ass-raped and crucified by the playing public. Therefore, they need to insure they are not caught.

How Would Online Poker be Rigged?

There are two basic types of rigging. One is designed to retain players who would otherwise bust and leave, thus maximizing long-term rake, and one is designed to maximize the rake on an individual hand, or pump up short term rake. Each goal would require different methods.

Fluctuations are a normal part of poker. Everyone knows that. But poker sites are in a unique position where they can control those fluctuations. If a player is about to go broke and another player has done well, software could easily detect that they are at the same table. One or two bad beats will ensue and big pots will be taken by the poor player. Setting this up would not be difficult for a clever designer. As long as it does not happen to any individual player too often, it would be undetectable.

Another trick that would keep players playing is the so-called cashout curse. This is when a player withdraws and then goes on a slide, losing a chunk of change. It does not even need to be real, as long as the fear is there then people will be less inclined to withdraw their money and will continue to play. Rigging online poker to punish people who cash out would be the same as any other form of rigging.

Pump Up the Rake!

What if Party Poker intentionally dealt AA and KK to players once more than would normally occur every 1000 hands? If these monster hands generate an extra $0.50 of rake, then Party Poker would have made an extra $1,350,000 over the 2.7 billion hands they have dealt so far. While this is not a lot of money given their huge worth, imagine if they could flop an extra flush draw here, an extra out-kicked trip aces there. It would add up and put additional millions on the bottom line.

All of these internet poker sites claim to be audited by third party companies who check the veracity of their randomness claims. There are a few problems with this. The third parties verify the data given to them by the online poker site. The poker site can control what data they get to see and thus control their assessments. If a poker site were intentionally cheating, would they turn over proof to a consultant? Never! In fact, it is possible to cheat under the consultants’ noses. If a consultant were to examine 1 million flops they may not see aces flop more often than expected. However, if the consultants cannot see all of the players’ hole cards, they cannot make a judegement on randomness. Sure the flops might be random, but if three aces are out preflop, would it be too difficult to flop the last ace more often than expected and flop fewer aces when fewer aces are dealt preflop? By using the hidden nature of the players’ hole cards, the online poker sites can effectively skew the deal while appearing random.

Finally, the huge amounts of data collected by poker sites could be mined to find profitable patterns. If a player chases flush draws every time he flops them, then the software could flop him a lot of flush draws while keeping the flops random. This player will miss and fold the river often enough that other players will never know what he may have had or how likely it was. By knowing which players will bet and raise certain hands, a good data miner can create specific subtle flopping rules to entice that playr to bet and raise, increasing the rake. They could also calculate a deposit threshold for players. If they know he redeposits at $400 they may let him get there just to get the money into their system. The possibilities of effective and undetectable rigging through data mining are endless and a little scary.

So Is Online Poker Rigged? How the hell should I know? I have never sat at a table with 9 friends whose cards I could factor into calculations for the 500,000+ hands that it would likely take to uncover any statistical anomolies with an acceptable margin of error. Is it rigged? Maybe. Is that scary? Sure. The point is that these poker sites have too much motivation to juice their flops and bouy up their shitty players. Rigging their site may seem to be an easy way to go.

No comments:

Post a Comment