V Low content (101K Tourney Hand stats for the online is rigged crowd)

Iago

LOW CONTENT

So I was going through Poker Tracker tonight and I realised I've played just over 101K Tournament hands, so I thought I'd post some starting hand stats at this stage. This is aimed at the online poker is rigged crowd, and yes I know 101K hands is close to nothing but I don't play as much as most of you!

Starting Hand Win % Win %WSF Won at SD Over/Under Index
AA 91% 86% 80% +2%
KK 90% 83% 77% -10%
QQ 83% 72% 68% +4%
AKs 81% 63% 62% -7%
JJ 78% 64% 63% -6%
Ako 78% 61% 63% +4%
AQs 72% 55% 59% +8%
TT 71% 52% 61% +9%
AQo 67% 48% 57% -%

Lowest Win %: T3o, 94o, T2o, T2s, 62o

Highest Indexing Hands: 53s, k4s, t7s, 64s, k3s
Lowest Indexing Hands: K8s, 54s, KK, 92s, 65s


So what does this tell you??

1. Over time the hands you expect to win will win, it all evens out

2. I get to showdown way to often with the 2nd best hand

3. As a player I'm generally "less lucky" in that I under index on the top hands.

/LOW CONTENT

edited to include correct figures

RoundTower

what is the index meant to mean, you got AA 22% less often than you would expect over those 101k hands? You expected to get AA-QQ 1400 times or so but you only got them 1100? That sounds statistically unlikely and probably is good evidence that the site is rigged.

Iago

RoundTower wrote: »

what is the index meant to mean, you got AA 22% less often than you would expect over those 101k hands? You expected to get AA-QQ 1400 times or so but you only got them 1100? That sounds statistically unlikely and probably is good evidence that the site is rigged.

Exactly, the index is based on the overall hands, so if on average I should get each hand 100 times, what I've actually seen is

AA 78 times
KK 69 times
QQ 80 times
AKo 159 times

and so on. It's based on tourneys played across Tribeca, FT & Ipoker. Probably a 50/20/30 split give or take a couple of percentage points. I suppose I could do an exercise in splitting the sites out and check the variance between them, but I'm not sure I've enough hands on any one site to be statistically significant.

mewso

Ah yeah but it's the cash games that are rigged.

RoundTower

OK, you got dealt 101,000 hands. You expect to get a big pair (JJ+) 4 times out of 221 which means you get an average of 1828 big pairs. You won't always get exactly this number, though, but 2/3 of the time you will get within 42 of this number. This is called one standard deviation. 99.7% of the time you will get within 127 of this number (three standard deviations).

You actually got 1371 big pairs, which is 457 away from the average, or 10.5 standard deviations. This means you are incredibly unlucky. If every man, woman and child in the world 12-tabled online with a fair deck for 40 hours a week for a year, there would be a one in a million chance someone was that unlucky.

So it could have happened -- you could be naturally the unluckiest poker player in the world. Or you could have made an error recording or reporting the data. The only other explanation is that at least one of your three sites is utterly rigged.

pok3rplaya

Iago wrote: »

I've played just over 101K Tournament hands,

Beat.

Iago

RoundTower wrote: »

OK, you got dealt 101,000 hands. You expect to get a big pair (JJ+) 4 times out of 221 which means you get an average of 1828 big pairs. You won't always get exactly this number, though, but 2/3 of the time you will get within 42 of this number. This is called one standard deviation. 99.7% of the time you will get within 127 of this number (three standard deviations).

You actually got 1371 big pairs, which is 457 away from the average, or 10.5 standard deviations. This means you are incredibly unlucky. If every man, woman and child in the world 12-tabled online with a fair deck for 40 hours a week for a year, there would be a one in a million chance someone was that unlucky.

So it could have happened -- you could be naturally the unluckiest poker player in the world. Or you could have made an error recording or reporting the data. The only other explanation is that at least one of your three sites is utterly rigged.

Well I think option 2 is far more likely than option 1 or 3. So I'll check the data when I get home, my guess is I have a formula problem within my data. I wasn't really concentrating when I did it up.

I've no doubt that when I review it I'll find something really simple is messing up the data.

pok3rplaya wrote: »

Beat.

lol, that depends on your perspective I guess. People can slate tournaments all they want but as long as I'm making a decent return from them I'll be happy enough. STTs can be mind numbing in isolation, but I find 4-5 tabling while watching TV or reading is fine. Just switch on auto-pilot and count the winnings!

TommyGunne

Are you sure the over/under index doesn't refer to how often it was expected to win compared to how often you won with each hand? This would be something akin to Sklansky bucks but not quite.

This would explain a much higher variance, as it is clear that one particualr hand can run below expectation for a long time. Each pair has a sample size of only 457 so this is in essence quite small.

I think I'm wrong though, as unpaired cards should then have a smaller over/under index, as there is a larger sample size, but this is not the case in the data set.

But teh more obvious answer is poker is rigged against you.

RoundTower

how often is any hand "expected" to win though? I don't think that means anything.

Iago

changed the data to reflect the correct information, i'm still under indexing on the top hands but only just...

with the exception of KK which at -10% is a pain