It doesn’t take a genius to work out that the cash games at PokerStars are currently the most popular cash games on the internet. But what is the reason for this? Are players actually making more money on Stars? Or is it more of a case that people simply want to do what is popular?
We are going to run a simple experiment to help us compare the two networks. The details are as follows.
Challenge – To play 12,000 hands on each network and keep a log of our progress. When we have sufficient data we will run some analysis, taking in to account the following:
- Softness of the games and our winrate over the sample
- How much rake we paid and what the VIP system is like
- General tendencies of the population
- Any other factors that may influence the profitability of our play
Expectations and Projections
A perfect version of this test would involve a much larger sample of hands. 12k is enough to make observations about the population, analyse the rake levels, and comment on the quality of the games. It is not necessarily enough hands to see an accurate representation of our winrate. This would ideally involve samples well over 200k hands on each network.
There is still a reasonable likelihood that we will end up having a higher winrate on 888 than we do on Pokerstars due to a prior expectation that 888’s games are softer. It is completely possible to run hot on one network and cold on the other however. While most players have a strong tendency to look at graphed results, it should be kept in mind that looking at population tendencies will likely provide a stronger basis for making estimations regarding possible winrates in the future.
Log and Analysis
There will be two additional stages to this report. If you prefer to get straight to the point and see the final analysis/outcome of the challenge then head to the analysis section.
If you prefer to see the nitty-gritty of each day of the challenge and some selected hands that were played then there is a log for both the 888 and Pokerstars grind. Warning: some serious bad-beats and coolers coming up!
Some things might generally go without saying but we’ll say them anyway.
Firstly we will always have 100 big-blinds or more in our stack on both sites. If we fall below this threshold the auto-rebuy feature of both Pokerstars and 888 will automatically take care of this for us. This is a standard thing to do because the poker strategy we are employing works especially well with 100bb stacks, so it makes sense to spend as much of our time playing close to this effective stack as possible.
Having said that, we won’t be overly concerned about cashing out our stack if we get deep. And usually there are one or two winrate implications to this. Generally speaking, the deeper the effective stacks, the higher our potential winrate, because players typically make much bigger mistakes when they have big stacks.
The sessions will all be played on weekdays, Monday to Friday. We expect there to be some variations in winrate based on which time of day we play and whether it is a weekend or weekday. It will be impossible to draw big conclusions on this given the small sample of hands we will accrue over this challenge, but we’ll try and keep the test reasonably fair with an even distribution of play times.
Our total goal is 12,000 hands and we will dedicate 5 days (Monday-Friday) on two consecutive weeks. It doesn’t take complex maths to establish that we should be playing roughly 2,400 hands per day. We mostly stick to this during the challenge, but are not overly concerned if we are over or under by 100 hands or so each session. So long as the final total is around 12,000 we are happy. Each of the 2,400 hand days are broken into 3 sessions of roughly 800 hands, but there were variations in this based on external factors.
All hands will be played at the 10nl (buyin $10) tables, in the fast-format 6max games. (Zoom and Snap respectively). It’s usually possible to play around 800 hands in an hour with 4-tables open, so projected grind time for each day is just over 3-hours not including breaks. The report made the mistake of assuming it was possible to play 4-tables of snap at any one given time. It is in fact possible, but one of those 4-tables needs to be of a different limit to the other 3. This could effect the quality of the project significantly since we’d anticipate a different winrate at a different limit. As such we played 3-tables of Snap on 888 and accepted the fact that the grind would now take 4 hours instead of 3.
This target audience for this report is actually recreational players. We will provide general analysis of our observations, and basic breakdowns of some of the hands we play. We will not however do any highly advanced database analysis or detailed commentary on the posted hands.
Such analysis would be possible, but we’d typically set different parameters if we had a more advanced target audience in mind. A significantly higher sample size of hands would be necessary for a start. The goal here is to help recreational players make a logical decision regarding which site is best for them in the long run.