Question about data from Pokerroom

In the thread below, the consensus seems to be that the data is tainted because it includes the losing players with the winners. That's fine if you want to know how the EVs of KTs and JTs compare when expertly played. But what if we wanted to know if Joe Blow could play it profitably?


If the data shows that a particular hand is consistently profitable for everyone, could we then assume that said hand would be even more profitable in the hands of an expert? Can we assume that if a hand shows a profit when played by the masses, that it is then a profitable hand in that position?


For example, if QJo is shown to be profitable only within 3 of the button, should we then assume that the hand is playable once it gets to that position, and should be folded before that? After all, if all of these poor players are able to show an overall profit with it, shouldn't a competant winning 2+2er be able to do the same thing?


All feedback welcome.

you still have to take into consideration the texture of the game. which the stats dont mention. its just a general overview...


"should we then assume that the hand is playable once it gets to that position, and should be folded before that?"


you may be asking for too definitive of an answer here. this is why the data is skewed a little and can be misleading.


b

"you still have to take into consideration the texture of the game"


Bernie, on one level, no duh. Obviously anyone takes game texture into consideration when deciding what to play where.


But on another level, maybe not. Its not like Joe Blow is necessarily doing that. If over 7 million hands, any retard is able to play QJo profitably 3 from the button, surely that occurs over a range of situations, no? I think that it is an idea that seriously needs to be examined. We are dealing with average tables and average results. So we aren't talking about "very tough" games here.


"This is why the data is skewed and can be a little misleading"


The data is correct. If it says AA on that site has had an EV of 2.44BB over the millions of hands played, then that is what it is. The data is right. It is the meaning that is up for debate.


It is only misleading if you try and get it to be something that it isn't. That is why asking appropriate questions is key. Just because we don't necessarily know the correct questions to ask right now, doesn't mean that we shouldn't at least make an effort. To dismiss the data as being useless out of hand is silly, IMO.

i think it would be interesting to compare from one site to the next. see if there are any parallels. never said it was useless, just dependent.


b

Can we assume that if a hand shows a profit when played by the masses, that it is then a profitable hand in that position?


This certainly sounds logical on the surface. I'll have to think about it.


Another issue that just struck me:


The "Conspiracy Forum" always has posts about bots being used on poker sites. For some reason, certain players think that makes the game harder rather than easier.


So, do these stats include the alleged bots?

"I'll have to think about it"


Do that. I think its a very interesting question. I think there is quite a bit here that is worth discussion, if nothing else.


I agree with Bernie that I wish other sites would publish this info. I would love to have access to similar data from Paradise.

Is my question too difficult, or too easy? I can't imagine that only 2 people have an opinion on this one. So its either real easy and I'm missing the boat, or its difficult an no one wants to stick their neck out with an opinion.


Love some feedback here, folks.

Clarkmeister,


The problem with data and statistics is that you are able to infer (nearly)anything you like by using the particular elements you choose. Or by setting up your hypothesis with specific parameters as you wish to in your question above. I believe the stats are a bunch of cool numbers that you may use to justify playing a specific hand in a certain position if you like.


Using your QJo example: You may get the hand the 125 times it loses in late position before you see the 212 times it wins (not real numbers here). If that happened to you would you keep playing it after 125 consequetive losses no matter what the stats said? Methinks NOT!!!! You would probably analyze the particular game you were in and se why it did not work in "your specific circumstances".


Not that I am a competent 2+2'r but you asked for a response.


Jimbo

My opinion is that Joe Blow is a much better player than 2+2ers give him credit for, and if some batch of stats says that Joe makes money in online games with such and such a hand in such and such a position, that info could be useful in a real poker game and I want it.


Let me know when the stats come out. If they don't include position, save the stamp.


Tommy

thomas


i know you are a busy man, but take a quick click on this :


https://www.pokerroom.com/evstats/totalStatsPositions.php?players=10


it will show you the info that is available to you if you want it

I mistakenly thought I had to download a casino to get to this info. Thanks!


Tommy

Clarkmeister - I have not seen the statistics, have only read the posts. I believe that the statistics are flawed when they compute EV and, fotr that reason, cannot be used the way that you propose. Jim B says that the overall EV is positive because not all hands have an equal occurance. If true, then the stats need to be weighted os that 72o can be compared to AA or QJo to K10s with some degree of confidence. You can then make your assumptions and they will be valid.

For example, if QJo is shown to be profitable only within 3 of the button, should we then assume that the hand is playable once it gets to that position, and should be folded before that?


If you belive:

1. That most of the players are dopes,

2. That the data is correct and even they can make a profit with these hands from a certain position. Then...

3. These hands must be played by good players.

Ok, heres mine. My only concern with these statistics is how the experts did with these hands. I think we would agree that an expert player will always have a higher ev than a weak player, no matter what hand is played. The SM hand rankings are for optimal play when used in the right situations. The Poker room data supports average play by the Joe Blow's that sit at our tables. Since we don't strive to become suckout artists, this data should be taken with a grain of salt for the hands that we expected a

-Ev with, but showed a +Ev on these sheets.

I do however, like the info on the -Ev hands. It shows that some hands are trash no matter who plays them.

I did some number crunching of their data, and the overall EV is NOT positive. In fact, it ends up exactly where you would suspect it would, given that most of this data is from 1-2 through 5-10 games. Their EV calculations appear to be spot on.

Tommy,


That page has the macro view. There are other charts on that site with a micro view, by level, position, etc. The numbers I sent you were calculated using the large EV charts and their stated actual number of times each hand was dealt, depending upon position.


When I get home from St Louis, I'll have some more time to get inside the numbers. They seem to corobarate some things that I have long suspected.

I don't understand. You are saying that poor players are able to turn -EV hands to +EV hands by becoming "suckout artists"?

I agree that you can incorrectly infer many things from this data, depending upon your goal and the way you phrase the question.


That's why I am starting with a very simple question: If a given hand is +EV in a certain position over many trials with many players, can we then assume that an expert should play that hand in that spot and earn an even greater profit?


I think that is a critical question to analyze.


The reason I think so is this:


When a hand performs worse than an unexpected hand, people say "well, the (worse performing) hand was played suboptimally, and really is a higher EV hand when played correctly"


Or "well, they run into kicker problems so that minimizes the hands EV" or "they hang onto it too long and chase too much" or "they aren't being sensitive to position or game texture".


So if these players are playing so horribly, but STILL show a profit for a hand in a position, how in the world can it NOT be truly profitable for an expert in that spot, in the long run?


I want to know why not?

Thanks.


1. I don't necessarily belive this to be true. From what I hear, the avg online player is significantly more competant than the avg B&M player. However, since it is composite data from all players, we certainly must assume that no hand listed, except perhaps for AA , is played optimally. This is why I think that only looking at the +EV hands is in fact conservative, if anything.


2. I do believe their data to be correct. I analyzed the EV/seat in several scenarios, and the EV's were reasonable for each round. The data is also internally consistent.


3. If so, then there are some fascinating follow up questions that stem from this conclusion.

Clarky----heres my opinion. I don't like the data chart.


For example: 72o and 62o played utg have a negative ev of -.11 thus 1/10th of a big bet. If you wait to play these hands until you are the button, they are only -.02 Am I reading this right? Shouldn't they be WAY more negative than that?


But who is going to play them at all? Not me----and probably not you. But somebody did -- to get them in the sample for the chart.

I do not see how the data is of any value as new information. The data posted pretty much falls in line with previous accepted hand ratings.


I like to look at my hands as return on investment with a limited capitol base to invest from. If I didn't do this, I could sit at a $4-8 table, lose $150.00 an hour for eight hours, and think I had an okay night, because every hand I played did have a positive ev up until some point in the round. There comes a point where an investment is not the best investment. I think there is a fallacy that is being promoted from this data that some hands are better than they really are.


As for Joe blow, he does play a lot of junk profitably. If he didn't there wouldn't be a game to play in. He just doesn't do it consistantly. One of his brothers or sisters takes over for him quite often when he busts out. Good hands played well bring in good money. it's hard work, and like anything else in life, I don't think there is a magic shortcut.


I am not sure I replied with anything other than hot air.....

No, I'm saying that there are some winning hands in -ev plays by suckout atrists, that in the hands of a pro, would never have been played at all. I don't think that anyone on this board would cold-call Ax off for 2 bets and go runner-runner Aces up to beat AK. The better players would never be in that hand in the first place. That's where I think some of this data is flawed.