It's true you make less bb/100 by not being dealt in to each position, however there is no reason why seats you didn't get to play would go to a fish more often than a regular (unless your pool has more fish than regs...). In my pool it would be more likely to go to another regular because there are more regs (and bad regs) than fish.

Fish miss hands just like anyone else when they sit out. It's probably more likely they just keep playing til they get stacked enough times to hit their pain threshold or lose their roll - which in that case would not effect the amount of rake the site collects at all compared to a normal table.

All that happens when someone sits out and misses some positions is positive bb/100 are decreased and negative bb/100 are even more negative. There really is no benefit for the site when you sit out because less people playing = less dealt hands = less rake. If you have 102 players and 1 sits out, that is potentially a whole table of rake the site is losing (101 / 6 = 16.83333333333333) unless they allow less than 6 to sit. In the mornings on WPN sometimes the games pause for a moment when you fold because there aren't enough players.

The most likely beneficiary would be people who play longer sessions. Reason being if the software places you in to 6 instances for example, the most seats you could lose would be 30 (SB through EP ... 5 seats times 6 sessions), but you're not likely to lose that many. If you played 200 hands in a session and missed 15 hands that would be a pretty big deal compared to someone who played 1000 or 2000 hands in their session.

We don't exactly know how the software works either. I am sure there are sessions or instances as I described because if you watch - you will notice it starts with a single session and eventually you will go from MP in one hand to the BB in the next, then you will be dealt in to EP (the first session you were in) ... or another BB to open another session, so on and so forth.

For all we know a participant sitting out could completely kill a session or multiple sessions the moment a participant sits out, at which point NOBODY would benefit, except the players closest to the BB (EP,MP, player who was next to be dealt in to the BB) had the opportunity to play more hands they paid the blinds for. It would be like the dealer saying "this table is now closed" at the end of a hand and all players involved in the session(s) that player was in have to get dealt in to new sessions (and would miss seats just like the guy who sat out).

It's tough to say without actually seeing the code, and every site probably does it a bit different.

It could all be prevented by just removing the requirement for players to start in the BB for their first hand and instead just deal every hand that gets generated with every element of it being random, instead of fixing one element as not random, (new entrants to the pool start in the BB), and apparently every other element of the hand being left to be truly random, (although again I am sceptical of this since sites haven't tried to address this issue before and they likely were aware of this before players have noticed, and it is in their interests for increasing rake by fiddling with the other elements too to decrease win rates of the best players in a fairly covert way, such as keeping winners away from the fish in the pool etc.).

Currently if a player plays 60 sessions of 1 hand, he plays 60 BB's and doesn't play in the other seats. Removing that requirement means someone playing 60 sessions of 1 hand means he will only play 10 BB's on average.

That would be bad because then players could play then sit out without paying the BB. If they literally just did it randomly we would end up with the same problem we currently have actually - being that whenever you sit out you miss the seats in other sessions. Maybe it would be slightly more evenly distributed than it currently is now but there is a much easier to implement and more exact solution (described in my original post and below).

I am pretty sure what is happening now (and it is what should happen) is you pay for a BB and then within each session cycle through the seats as you normally would. All they need to do is add a sit out option that allows you to finish playing through your seats in each session before actually sitting you out. Right now you are just being sat out prematurely at your next BB seat.

I don't think players could sit out strategically as they wouldn't know when the next hand will be their turn to be in the BB as it would always be random. Perhaps they sit out thinking they just avoided a BB as it was surely their turn to be in the BB next, but actually unbeknownst to them their next two hands would have both been on the BTN had they stayed in the pool, so they would shoot themselves in the foot just as often as gaining from the sit out so that overall they would be indifferent to when they choose 'sit out next hand'. Essentially every player's next hand would always be a 1 in 6 chance of being in the BB, no matter what their recent run of hands were, even if they hadn't sat in the BB for the last 10 hands straight, they still have a 1 in 6 chance for the next hand.

Please can you explain your point to show why it would be bad as I am not convinced without an explanation.

I guess it could be "better" than the current situation if things end up being more evenly distributed - however implementing something like this is likely to be an absolute programming nightmare and would not yield as good of results as simply just adding a "finish all sessions then sit out" option. With the "finish all sessions then sit out" option - you would never have a discrepancy in the number of positions you play barring any short handed situations or if you just close the table without using the "finish sessions" option.

Adding the finish sessions option should be very easy in theory, like something that could be programmed in anywhere from a single day to a month, although there is always the chance the code required for something like this is much more complex than I imagine.

Why would it be a programming nightmare? It just looks at the pool and tries to create a hand from the available pool and then randomly puts 6 of those players together and sits them randomly.

If you are in a pool with 50 other players, then your next hand always has a 1 in 6 chance of you being in the BB and a 10% chance of any of those individual opponents being part of your 5 opponent roster for the next hand.

Obviously when you have multiple entries in the pool and so do other players, then the random opponent selection will need to not match up entries from the same accounts in the same hand, but that is the only slight allowance that need be made and the current programming can already handle that.

To do it randomly, they'd have to scrap probably most if not all the code they already wrote and write new code. Right now the code probably just looks for people who are next in line for a position (ex: your next position is cutoff, so now it needs to find 5 players who are ready for another hand and are supposed to be in the remaining positions) - so that code would have to be scrapped and instead say "these players are ready for a hand" and randomly put them in to a seat. This will result in fairly even (but not exact) distributions in the long run, but it doesn't make a lot of sense because inevitably there will be some poor soul(s) who pay the BB/SB significantly more than they should in the short to medium run.

There might be some ways to counteract that effect but it's still an inferior solution that would require more time / resources.

what is a samplesize?

also: what is stdDev?

Samplesize is ammount x of something you use to analyze something. In this cas hands played. stdDev is standard deviation meaning a random variable, sample, statistical population, data set, or probability distribution is the square root of its variance. In this cas how good / bad you can run with your winrate in set period of time.

Hope this helped.

Can anyone confirm whether this "bug" still exists or whether it has been fixed? I'm getting significantly less EP and HJ hands than I should be, with the extra hands seemingly going to SB and BB. The sample is too small to be conclusive though.

It's been too long and I can't edit the post above, but would just like to clear it up for anyone who is having the same issue and finds this thread.

I thought there was something wrong with the ZOOM dealing algorithm because my PT4 results were showing me getting the most hands from the blinds and significantly less from EP, MP, and CO. Turns out that if you fast fold then close out the table before the hand finishes, it doesn't get written to the HH file and consequently doesn't get imported into PT4. Before getting dealt the BB, you're most likely to get dealt a hand in EP or MP. So, for example say you click 'Sit Out Next BB', get dealt 72o in the MP, fold, then get sat out next hand and leave the table. The 72o hand won't be imported into PT4 and it'll seem as if you didn't get dealt it at all.

TL;DR The ZOOM dealing algorithm is fine, trackers just won't be able to import a hand if you close out the ZOOM session before the hand is finished.