flop all-ins analysed for fairness – 6 million hands tested by Online Poker Watchdog
Online Poker Watchdog has just published the results of a new series of ‘bad beat’ tests. These tests focus on flop all-ins and will detect a bias in the deal towards the ‘underdog’ in all-in confrontations. They will also detect a bias towards the players that calls the all-in bet.
See the full details of the tests including explanation, method, results and conclusions at ‘Is Online Poker Rigged?’
Large Sample Size
Many tests that are run independently by poker players fall down because they only have a small sample of hands and so the results become meaningless. This isn’t the case with these tests as over 6 million hands have been analysed – more than 1 million hands for each site tested.
Wide Scope of Tests
Major independent sites have been tested: PokerStars, Party Poker and Bodog Poker.
Large networks have also been monitored: Ongame Network, Entraction Network and Merge Network were analysed. Between them, these three networks have over 150 member sites that share the network.
Altogether, these sites/networks comprise 85% of all online poker traffic.
Results
For all sites and networks the results show that there was no evidence of a bias in the deal – all the sites passed the test.
Conclusions
These tests are limited in that they only detect bias in the turn and river cards of flop all-ins. Although, similar tests have been performed for pre-flop all-ins, so a bigger picture is beginning to form.
Also, bear in mind that not every online poker site has been tested. Sites and networks that don’t have their hands ‘tracked’ by hand history providers cannot be monitored. These include sites that actively try to prevent data-mining, sites with ‘anonymous’ tables where a player’s screen name doesn’t show and small sites that hand history providers don’t track. In our opinion, players at these sites are at higher risk.
However, for the sites and networks that are ‘tracked’ it seems that the theory that the favorite hand sees the wrong end of more ‘bad beats’ than they should when the chips are all-in is untrue.
I find this pretty hard to believe in the case of the Merge network, suggest you re-run your numbers. I play at these tables every day and it is pretty ridiculous the bad beats that consistently take place.
I recommend that you run your own hand histories through the test – that way you will know if you get more bad beats than you should. I imagine that you should have a decent sample size if you play every day. I’d be keen to see the results if you want to post them here.
I used Holdem Manager instead of Poker Tracker. Can I still run the test?
Online Poker Watchdog have run this test using poker tracker and the download is only available for Poker Tracker. However, I’m sure that you could configure a similar custom report in Hold’em Manager and then use it to run the tests. If you go to the method page at Online Poker Watchdog they have detailed what the outputs are from the custom report.
My red line in HEM is 70 buy ins below EV in the (admittedly small) sample of 511 tournaments on Merge. That being said my perception is based on all hands that I have seen, not just the hands that I have been in. I think Merge is pretty suspect.
Were these results also tested on the ipoker network? Being a player here it would be good to know the results of this, but I cannot see a mention above…
Hi Phil – good question regarding the i-poker network. To date there have not been tests run on i-poker – there was originally some difficulty with obtaining hand histories that had hole cards shown. This was resolved but then there was an incompatibility issue when running the tests. Online Poker Watchdog are working on this. First they aim to complete a series of ‘turn all-in’ tests to see if good players get nailed by the river card too much – after that they intend to test i-poker thoroughly.
I have one question, how do we know Online Poker Watchdog has not been bought out by the big poker sites? These big dogs can afford to give “donations” to get a favorable review. There should be another independent group to corroborate their findings.
Thanks for the post Charles – this is a good point.
When Online Poker Watchdog designed these tests they realised that some players would be skeptical of their results (and indeed that it is right to be skeptical). That is why they have made their tests available for anyone to perform by having the Custom Poker Tracker Reports available for free download.
Therefore players can test their own hand histories, or even buy a million hand histories from the site/game that they play at and test them. In this way players can be sure that the results are not biased in any way.
Of course there rigged but bot to gain money but ensure the tournaments are finished in a particular time..pokerstars Sunday million would take a week or 2 to finish if it wasn’t the case and the Sunday storm with 25000 entrants would be a month long tourney..
I’m not quite certain what the poker site will gain on this. It is my belief they stand to gain nothing in the long run, and it will only give them a bad rep. They’ll lose more if they rig the games. These large poker sites are earning more than enough money anyway. Doesn’t make sense if they have to rig it just as a casino don’t have to cheat because HE makes sure they always earn nonetheless.
Thanks for posting Timothy. There are many players that agree with you on this – for already profitable sites there would be a lot to lose if they got caught rigging the deck. I think that your argument is good and makes sense but the problem with any argument is that it is open to debate and counter argument. As you will see in the forums the debate goes on and on and round in circles! There are a lot of players that are convinced that online poker is rigged.
That is why Online Poker Watchdog run these tests – to provide some hard evidence in an area where it is scarce. The results seem to be backing up your argument – no evidence of rigging has been found so far.