How many hands before I can...........

[Silverback]

How many hands do you think I would have to play before I can realistically get a good idea of what hands are profitable from each position, or played in certain ways from each position, e.g 65s on button after 2 callers,


Note I would be looking at hands as a group performance, eg A7-A2s might be 1 group, QJ,QT,JT, might be another, 98s,87s,65s,54s, or 66-22 and so on.

I would be looking seperately from each position of the table with 5-6 players present at table.

I would simply stop playing some hands based on there results.

Would 60k be enough?

Has anyone else done this?

Note, if anything Ive been playing slightly loose, so hands are there to be cut out,

[BreakfastBurrito]

I don't have any hard evidence, but I would guess that 60k wouldn't be even close to enough. Most relatively marginal hands are going to be in the +.1 to -.1 BB/hand range. For say, 65s, you're going to have seen that roughly 190 times after 60k hands, so you'll have about 33 hands at each position. I don't think there's anyway that your variance is small enough at that point to determine that your +.12 BB/Hand from the CO is reliably profitable. My guess is that you'll need something on the order of 2 million hands to get something close with most of the hands you're talking about.

[kapw7]

[ QUOTE ]
My guess is that you'll need something on the order of 2 million hands to get something close with most of the hands you're talking about.

[/ QUOTE ]

So the odds are that you die before you know the answer. Just tragic. [img]/images/graemlins/frown.gif[/img]

[donger]

Check the EV stats on pokerroom:
http://www.pokerroom.com/main/page/g...stats/expValue

here's 6-handed:
http://www.pokerroom.com/games/evsta....php?players=6

[Silverback]

You mention 65s, however I clearly stated not looking at 1 individual hand, but looking at a group performnace of hands.
AQ might be losing 1st in from CO, but id obviously still play it. 98s,87s,65s as a group might be losing 1st in so I fold them.

Combining actual results against what I might expect results to be and using some common sense in judging the results and coming to a conclusion do people still think 60k hands is not enough.

Since making the changes my vpip is around 23-24 and pfr 15%, which is in line with forum TAG play.

To say 2 million hands seems absurd, it would take many hands to realise raising 72 UTG is a losing play. Many other results are borderline, so a few I just cut out, as long as Im playing enough hands to make a "potential" 3-4bb/100 then why waste time on marginal hands which will just increase my variance, Im better off playing real tight with winning hands and possibly adding more tables.

[cartman]

I have done this, but I read a post (I think by StellarWind) which made me feel like a complete moron for doing so. He basically said you need a jillion hands to be able to tell. I did it in the way that you describe where I take a group of similar hands and look at them as a whole.

Cartman

[Silverback]

I did use these stats when evaluating my results,

Notice how 87s, 65s are +EV UTG, but if you look at 87s,76s,65s across a few positions you can see they are loosers. SO if my own results are similar or even very borderline winners, then these are future folds for me.

Actually some of these did profit for me, UTG, but I had played them few times so must have picked my spots well, or been plain lucky, my conclusion was to always fold these unless maybe once in a blue moon to mix my play up on the right kind of table, but generally i dont see point of that unless playing against people for long time, which is rare as its easy to move tables online.

[meow_meow]

[ QUOTE ]
Check the EV stats on pokerroom:
http://www.pokerroom.com/main/page/g...stats/expValue

here's 6-handed:
http://www.pokerroom.com/games/evsta....php?players=6

[/ QUOTE ]

Thanks for that link.
I wonder how many hands pokerroom is using for their stats. I just can't understand why 88 would be worth more utg, in mp and co than on the button...for example.

[Silverback]

I dont think u need a Jillion hands.

We know the vast majority of playable hands from each position and how they should be played, so that makes an easy 20vipip.

Looking at my on results, along with mass sample stats like those from PokerRoom and also posters in this forums opinons as well as what certain books tell us hands are profitable from given positions.

[Snow]

[ QUOTE ]
I just can't understand why 88 would be worth more utg, in mp and co than on the button...for example.

[/ QUOTE ]

I might be totally off but could it not be the simple fact that ppl respect a raise from early position more? (and i assume the data is collected from real hands showdowns)

Regards
Snow