We analysed Merge Network (see list of all merge network poker sites) heads-up cash games to see if they were rigged for winning players to receive more 'bad beats' than they should...
We started by buying a sample of 1 million hands* and selecting the pre-flop all-ins from these.
For the player that called the all-in we compared the expected number of hands won (in a perfect world) with the actual number of hands that won.
If there was no rigging at Merge Network the expected and actual numbers should be very close:
| Total No. Hands | Expected No. Hands Won | Actual No. Hands Won | Deviation |
|---|---|---|---|
| 3822 | 2035 | 2014 | -21 |
So, from just under 4,000 hands the actual number of hands won differs from expected by 21 hands... is this "very close"? To answer this question we calculated the standard deviation for the sample - if the actual deviation is within 2 standard deviations these numbers can be considered "very close".
Please note that this explanation of standard deviations is simplified. For a better explanation of how to interpret standard deviations see these short videos on how to interpret test results.
| Standard Deviation | Deviation |
|---|---|
| 28 | -21 |
The results show that the actual deviation is well within 2 standard deviations and therefore the sample can be considered unbiased.
Next, we separated hands that were ahead pre-flop and hands that were behind pre-flop and repeated the comparison.
If the Merge Network was rigged against good players then the 'ahead' hands should win less than expected and 'behind' hands should win more than expected. However, if there was no bias the expected and actual results should be very close.
| Group of Hands | Total No. Hands | Expected No. Hands Won | Actual No. Hands Won | Deviation | Standard Deviation |
|---|---|---|---|---|---|
| Ahead | 2165 | 1472 | 1468 | -4 | 21 |
| Behind | 1652 | 561 | 543.5 | -17.5 | 18 |
The results show that in both cases the actual deviation is well within 2 standard deviations and therefore the sample can be considered unbiased.
We can conclude that the Merge Network was fair with respect to 'bad beats' at the time of testing - there is no evidence of Merge being rigged.
This test was performed on a specific game type (25c/50c 2-player tables) during a specific period (May-July 2011) at a specific poker network (Merge) and the results can be considered true for these conditions only. Although these results are relevant to online poker in general other circumstances were not tested. Other poker sites may use different methods for the distribution of cards and other game types (e.g. multi-table tournament hold'em) or levels (e.g. $2/$4) at Merge could also use different programs for the deal. It is also true that the method of dealing at a given site could change in the future as the site updates.
Please note that the explanation of the analysis described on this page is simplified so that it can be viewed easily and at a glance. If you want more detail and a full explanation of the methods you should view the full report.
Online Poker Watchdog intends to continue to perform this test on other poker sites for a variety of games and levels.