PT Total Agression Estimates for O8b

[Luv2DriveTT]

Fellow 2+2'ers & Omaha fans:

I am in the process of TRYING to develop a rule set for Poker Tracker. I am basing the rule set on similar parameters that both the Bison & DRIVETT (developed separately with similar thought processes) rule sets have. So far I am thinking of developing a rating system based on three criteria:

1) Percentage of hands played pre-flop (tight, semi-loose, or loose)
2) Total Aggression (passive, average, aggressive, over-aggressive)
3) Went to Showdown % (tight, semi-loose, loose)

For those of you who have used Poker Tracker to play Hold'em and are familiar with my rule set will notice that I am not using pre-flop raise% for the Omaha 8/b ruleset. With my limited knowledge of Omaha I think it might be less important to rate a player's ability by how often they raise pre-flop, it seems that how often a player goes to showdown is far more important in evaluating Omaha opponents.

My biggest quandary however seems to be total aggression. My very small O8b database to date has less than 150 hands played but there seems to be one common theme.... total aggression levels are generally much lower than I have found within Hold'em. Therefore I am looking for good descriptions of various levels of total aggression, and how they correspond to everyday play. Can Omaha/8 2+2'ers please help me complete my quest to develop a auto-rate system by providing me solid examples of total aggression statistics??

TT /images/graemlins/club.gif

PS: since PT for Omaha is so new this will have to be a work in progress that gets updated as the best players develop databases consisting of thousands of hands. But for now, I'd be happy to find a good starting point, nothing more!

[Nick709]

I think it is a great thing you are doing, especially because I would not be able to do it myself.
But I definitely have to argue that PFR% belongs in autorating. As far as total aggression goes, I wouldn't know where to start.

[BradleyT]

Are you looking for us to post our individual stats or is there a summary page we can post some numbers from?

[Luv2DriveTT]

[ QUOTE ]
I think it is a great thing you are doing, especially because I would not be able to do it myself.
But I definitely have to argue that PFR% belongs in autorating. As far as total aggression goes, I wouldn't know where to start.

[/ QUOTE ]

And why is that? I mostly play low-limit O8b where total aggression seems to be a much better qualifier of a player's ability than PFR is. Hold'em is a different animal since the frequency of pre-flop raises is a great qualifier to determine if a player is aggressive or passive, where as in Omaha a PFR often does not follow with continued aggression on the other streets.

If I am wrong, please advise.

TT /images/graemlins/club.gif

[Luv2DriveTT]

[ QUOTE ]
Are you looking for us to post our individual stats or is there a summary page we can post some numbers from?

[/ QUOTE ]

I am looking for stats of good players, average players, and poor players. Tight and loose in all three categories. My Hold'em auto-rate settings were designed to provide a qualified projection of your opponents abilities within a 35-40 hand period. I would guess it will take 60-80 hands for the auto-rate to begin to be effective when Playing Omaha. Of course much of this is just a stab in the dark since we have never had such powerful rating tools available before.

TT /images/graemlins/club.gif

[gergery]

For the auto-rate discussion, I think it’s going to take quite awhile to get good #’s locked down. I think O8 plays much different than holdem, so I am not convinced the same 3 stats will be important (VPIP, pre/post aggro), and I definitely think the “correct numbers” will be different.

I also think we should set up different threads for PLO, PLO8, and limit O8.

Here are my initial thoughts on a good autorate system. Comments are for limit O8 unless otherwise noted.

VPIP will still be very important, probably the most important criteria, as in limit holdem

Preflop aggression will not be that important. Many good players don’t raise much preflop.

Aggression after the flop will still be very important, but getting a good read on this will be difficult. The % of time calling is the correct play is much higher in O8 than holdem. And the % of time you want an opponent calling is much higher in O8 vs. holdem. In holdem, you generally want your opponents to fold when you have marginal thru good hands and want callers when you have a monster. In O8, you generally want callers with many of your good or marginal hands, because you are getting only half the pot. Net, the objective of your raise in holdem is often “get others to fold”, whereas the objective in O8 is often “have others call when my equity is better”

Win % at showdown seems to be a very important criterion. Both winning something, and scooping (not sure if PTO can track that). O8 is a game of showdown-able nut hands, and you usually shouldn’t be going to showdown without strong chances one with with at least some chances the other way.

Went to showdown seems pretty helpful.

Blind defense should be higher, particularly since hand values run more closely together.

Shorthanded vs. full-ring will have a bigger impact than in holdem. Some hands like 3488 can be good head to head but terrible multiway. Hands like A268 can be good multiway, but poor head to head. Holdem hands vary too, but not the same extent at O8, I think. So the rating system will need to take into account # of players at table more.

--Greg

[BradleyT]

Well said.

Won at showdown as well as maybe scoop vs. split at showdown but I'm not sure if PT shows that.

[pipes]

Ugh, one of the reasons why I like Omaha8b was because it was pure from all of this stat tracking. Now we are gonna have all these hold'em pokertracker freaks coming over.

[sammy_g]

[ QUOTE ]
Ugh, one of the reasons why I like Omaha8b was because it was pure from all of this stat tracking. Now we are gonna have all these hold'em pokertracker freaks coming over.

[/ QUOTE ]
Time to take up stud/8? /images/graemlins/smile.gif

[pipes]

[ QUOTE ]
[ QUOTE ]
Ugh, one of the reasons why I like Omaha8b was because it was pure from all of this stat tracking. Now we are gonna have all these hold'em pokertracker freaks coming over.

[/ QUOTE ]
Time to take up stud/8? /images/graemlins/smile.gif

[/ QUOTE ]

lol Yep, I'm way ahead of ya

[JoshuaMayes]

I posted this in response to another thread on this topic and thought I repost it here. I mainly play $25 PLO8 at party, so thats where my stats are from. Many of the points have already been made, but for what its worth:

[ QUOTE ]

I just downloaded PTO yesterday, so I only have about 2000 $25 PLO8 hands from Party loaded in. My VP$IP is around 20.2. A 20 cutoff for "tight" might be right, as I believe I play a little loosely preflop. I think the cutoff between semi-loose and loose might be more indicative of true fishiness, however. Many of the players whose game I respect are rated semi-loose, with VP$IP in the low 20s. None have VP$IP > 25, however.

The default pre-flop agression stats seem off to me, at least as far as they are meant to correlate with good/weak players. Many good PLO8 players do alot of limping and not alot of preflop raising. All of the players whose game I respect are rated either tight/passive or semi-loose/passive preflop. No player in my database has a tight/aggresive preflop rating. In fact, only two players have an aggressive preflop rating, and both of them are players that my notes claim are "total maniacs," and both are big losers over the course of the hands that I have played with them. My guess is that preflop agression is just not a good indicator of skill at PLO8, at least at the $25 party tables.

The overall aggression stats don't factor in bet size, and thus are misleading with respect to how weak a player is postflop. If someone calls pot-sized bets with weak or one-way hands/draws, I want to play against him. In contrast, if a player calls min bets behind several others with longshot draws when the pot is large and he is likely to get paid off if he hits, I might not want to play against him quite so much. Again, several of the players whose game I respect, and who I would not classify as passive players (they always bet/raise the max, they aren't afraid to put all of their chips in with a good draw, they steal in LP or when opponents show weakness, etc.) are rated as passive overall, probably because they call small bets when the price is right and because raising and re-raising without the nuts or a huge draw in small buy-in PLO8 is not a good idea -- too many players call with garbage. Moreover, many times, calling is the correct play when you have a very strong 2-way hand or a nut hand with redraws, because you want to keep players in.

Anyway, these are my preliminary thoughts, and I look forward to hearing from some of the more experienced players.

P.S. I am earned a little over 6 BB/100 hands for these 2000 hands. Anyone know if this is a normal rate? Last time I tallied up my records, I had earned around $7/hour at these tables, mostly playing two tables at once, but I have no idea how many hands/hour I was playing.


[/ QUOTE ]

[Rudbaeck]

[ QUOTE ]
I also think we should set up different threads for PLO, PLO8, and limit O8.

[/ QUOTE ]

If just looking at aggression is interesting it will have to vary with stakes for PL. There are just so many many single bb bets into 20bb pots in low stakes PLO that any aggression factor of a good player will be artificially low.

I don't know why people even bet when they give you the odds to draw to a 2-outer... But they do. All the time.

[Luv2DriveTT]

[ QUOTE ]
[ QUOTE ]
I also think we should set up different threads for PLO, PLO8, and limit O8.

[/ QUOTE ]

If just looking at aggression is interesting it will have to vary with stakes for PL. There are just so many many single bb bets into 20bb pots in low stakes PLO that any aggression factor of a good player will be artificially low.

I don't know why people even bet when they give you the odds to draw to a 2-outer... But they do. All the time.

[/ QUOTE ]

For the initial discussion I don't think Pot Limit can even be considered. NL Hold'em auto-rating has yet to be a proven success for Poker Tracker; there is a possibility that Pot Limit Omaha may remain the same however I think it will be much easier to qualify than NL Hod'em. So for the porposes of this thread I only wish to discuss O8b limit games. Step two will be to compare Omaha limit games vs O8b to see if the auto-rate system is compatable, and then step three will be to try to develop a system for pot limit games. I've asked Ptrack Mike to get involved in this discussion as well, lets hope he can add some insight and in turn add some Omaha specific data such as Scoop at SD%.

TT /images/graemlins/club.gif

[Luv2DriveTT]

BUMP

[Luv2DriveTT]

Ptrack Mike PMed me today to help him think out the process for PTO additions that are necessary for Omaha. He is in the process of adding WinHi%, WinLow%, and Scoop%, as well as the equivelent autorate settings.

Since I am far from a master Omaha player (in truth I suck), I was wondering if a few solid players here would be willing to zip up 5,000 or so hands for me to use in the process of developing an autorate system that works for O8.

Thanks for your help!

TT /images/graemlins/club.gif

PS: wait till you guys all see what they have in store for you!