except the winners get bumped up and start losing (unless you're counting on the unlucky ppl at the 0-5th, 25-30th, 50-55th, and 75-80th percentiles of skill to keep losing at x bb/100 indefinitely to fund the winners). And before this all came out, they wouldn't even have any idea why they were suddenly losing. initial take is that a rough 3-4 tiers of players would be even more, or at least just as, awkward and distorting and disorienting for those involved than the "rank each player and get as close to perfectly even skill as possible".

think of the player segregations like you would think of a a 1/2 game at the following times: party poker 2003, party poker 2005, pokerstars 2007, pokerstars 2009, pokerstars right before black friday. If you split skill levels down those lines you get widely different skill levels, but can still have a wide array of winners as each time period has marginal and big winners. Once you put the marginal party poker 2003 winners in the pokerstars pre-black friday game they get crushed though. thats what you want to avoid as a site. the worst of the worst will still lose quickly. the best of the best will still win big. however, the marginal winners or slight losers wont be squeezed out.

And yeah if the best they can do is break the player pool into quartiles then I agree there will still be some winners (there would still be winners even if the skill matching were more granular than that since there's no way it could be perfect). I don't see how it's arguable that there will be MORE winners though.

The biggest winners will win less. The biggest losers will lose less. That money isn't being redistributed amongst the players though, it's going to the house.

do you seriously no understand?

CDL my "river bettor vs. Caller" example was just one way it gets "shady". If they intentionally seat 3 frequent PFRs with three loose callers early the problem is that same. I'm fine with Clark's quartile idea from a PR standpoint but if you're matching on anything beyond EV adjusted BB/100 I think its a much harder sell.

The thing for everyone to remember here is that casinos already offer a place where EVERYONE is -EV and yet they still do fine. People will play for variance if nothing else.

This is a good point. Run hot, get bumped up, lose it back.

I get what you guys are saying about how the matching could be more sophisticated than "sort by BB/100" (and if it is just sort by bb/100, hopefully the software is called IGGY).

However, if you think that at most they're going to split the player base into quartiles then how fancy can the software really be.

there will be long term winners bc there is not a player pool with infinite depth and they still need to make sure games run continuously at all levels.

if they get bumped up and start losing then you can move them back down. its not like an escalator that only goes up. it works both ways. also, winning and losing is not the ideal consideration for the sites. expected wins and losses should be what they use.

yeah except you're teleporting people between the groups as soon as they win or lose too much. (you have to, otherwise manipulation of what group you're in completely undermines the whole system)

yeah so long-term ev is 0 minus rake.

I guess a good result of this change could be getting solid winning bots away from the worst players

how far do you think you could split a player pool and still have it function? maybe into deciles at nl10, but literally no one makes a living there. at nl1000 FR games didnt even have 10 tables going at many times when I was still playing in 2008. You simply cant split that player pool too much if at all. NL400 could go into 2-3 ways max without a deeper player pool than I remember.

you dont understand how one could win big and get promoted and repeat a few times and then stall out as a .25bb/100 winner or move up in stakes at a lower skill level and still be a winner lifetime?

Well clearly PP split it up enough for people to notice without being informed.

long-term winrate EV, I should have been more precise. you can obviously steal a few bucks from fish then leave if that's what you're after.

biggest thieves in the world are the people who start structured settlement companies/money lenders.

world would likely be a better place if those dudes (or women, but thats highly unlikely) were assassinated

1. how could you even make 2 non-bot winrates EXACTLY equal vs. a larger player pool over the long run?

2. how can you set this to 0?

3. how can you find a large enough player pool so that the previous 2 conditions are satisfied?

4. why would setting their winrates at 0 be better than having one winner and one loser who redeposits or is replaced by another loser who redeposits?

the other downside is i have to see ****ing montel williams on my teevee every night of NBAing

CDL do you think this is good for poker long term?

Here's why you can't have marginal winners (or, to be more precise: you can have marginal winners, but it'll be arbitrary and skill-independent unless they manipulate the system):

The losers in quartile 1 will be better pokerers than the winners in quartile 2. So, if being a marginal winner in q2 is a possible equilibrium state, there'll be huge incentives to play exactly the level of poker to stay a marginal winner in q2 and nothing better (which'd get you bumped up). that's dog**** not poker.

You have to clearly define what you mean.

The goal isn't to match players as closely as possible, though. They'll realize that perfect matching, even if possible, creates few winners, maybe none, which will cause problems on their end.

I dont think it's really possible, though. Narrowing the pool finely enough that everyone is, say -2 to 0 bets/100, would limit the pool enough that getting games would be a problem. This would force the site to expand the group, opening up the win rates. It's impossible to say for sure, but intuitively I don't think you could keep games going smoothly without some winners since waiting for a game due to lack of opponents in one's class would be so bad for the site's rake.

So don't tell them what group they're in. If somebody thinks they're good playing 60/15 they won't be able to tell that their opponents are terrible. Would make for epic posts here though - "PP's algorithm must be ****ed, everyone I play against is a fish!"

well, I know that I would play midstakes again in a setup like this and I probably would stick to playing something like nl50 if it was legalized and the bumhunting continued at the pace it was at 3-4 years ago. so they would capture some of the semi-serious players like myself who aren't looking to make a living but wouldnt mind trying to make a few extra bucks.

I think its better for sites profits and better for the majority of players including practically all of the losing players.

I certainly dont think its extremely bad for anyone other than the people who play midstakes professionally.

.

But you're 50/50 to win millions! You either do or your don't. Tell me where else you can win 40 million dollars by spending 1 dollar.

Just saw the Western Sky post. lol.

Was watching a YT video the other day of one of their commercials and the comment section pointed out that either their 5k or 10k loan was at 116% APR.

You better be fearing for your life if you take out a loan from Western Sky.

see you outed yourself. you think this setup will allow you to make money. it wont. if you start making money, you get moved to the tougher bracket. have fun playing with that in the back of your mind.

Even if you don't bumhunt surely you'd like the chance for a terrible player to sir down at your table right