Quads - Expected Abberation?
 

So, I've been playing on an online site (fairly popular one)... I haven't played on this particular in a long time. The last time I played on it, I noticed a TON of quads. I didn't think much of it, and I chalked it up to variance.

Well this weekend, I started playing on it again. I've played about 700 hands... and here's what I've seen.

Total of 11 SHOWN quads. And, 2 'quite possible' quads where there was no showdown. Basically, flop of 555 and TTT, and a bit of action.

Of the 11 shown quads, 2 of them were pocket pairs. One flopped quads, and the other was a set that improved to quads.

The reason I started counting was that I saw three quads in a stretch of about 30 hands.

Anyway, this feels totally wrong. And, I'm not one to accuse any site of having a screwy RNG. What do you think? Is it more likely that I am seeing some aberration that is expected or more likely that the RNG is not quite right?

-RMJ

An Update.
 

Played another 100 hands since my original post... and well, someone had quads with A5 on a board: 55xx5

-RMJ

Re: An Update.
 

My guess is you see more quads online, because lots of people stay in with weaker holdings to see what happens. The percentage of quads in a game where only 2 people see the flop would be much less than a Nofoldem style game. It does seem like the quads are running high on the variance spectrum at your tables recently, but that is to be expected sometimes.

Re: An Update.
 

Well, I play quite a few hands on a handful of sites. It is only this one site that I've ever seen the excessive quads.

From what I've learned reading posts here, if someone says they observed something is that far out of line from normal, then there's a good chance it wasn't normal or the observation was fabricated.

Now, my question is... if you observed what I observed, and knowing nothing about where I observed it (online, live, whatever). Would you think something was fishy? Would you think I was lying about what I saw?

It's just odd that I've seen this on this particular site on more than one occasion. And, I don't think it has affected the games much, just found it really weird.

-RMJ

Re: An Update.
 

On any individual hand, a player has 13 * C(48,3)/C(52,7) odds of getting quads, I believe, or about .17%. So if everyone has this same opportunity, there's a roughly 1.7% chance that a 10 person table will have quads on any given hand. So you could expect this roughly every ~58-59 hands. I know these numbers are at least slightly off...but I think it falls well within the realm of possibility that quads will show up 11 times in a 700 hand stretch at a 10 person table. The table would have to be pretty loose to see all of them, but I don't think it's *that* big a stretch.

But online poker is rigged. Move to a non-pattern mapped table.

Edit: My math should work for one person being dealt 7000 hands, as opposed to 10 people being dealt 700 hands. I don't have time to compute that probability, but I don't think it's going to be *that* far off.

Re: An Update.
 

Thanks for your reply... but I am not so sure about your "edit" section.

I would have guessed that 1 person being dealt 7000 hold'em hands would be vastly different than 10 people being dealt 700 hands. This is due to the existence of the community cards that everyone shares. But, I might be wrong. As I said, it is my guess. Similarly, I think that it is for this reason that you are more likely to see stronger hands more often in a game such as 7-card stud.

Anyone care to expand on this?

-RMJ

Re: An Update.
 

I think that it's very far off.

~D

Re: An Update.
 

I think its very similar, if you have a board showing a pair and only 1 player playing, they will have quads about 1/1100 times. If they play ten hands, on this board, they will hit quads about ~1/110 times. If there are 10 players at the table, they will each have quads about 1/1100 times 1*10/1100 = 1/110. For every case the same thing can be done whether there are 3 of a kind on the board one pair or 2 pair.

Re: An Update.
 

[ QUOTE ]
I think its very similar, if you have a board showing a pair and only 1 player playing, they will have quads about 1/1100 times. If they play ten hands, on this board, they will hit quads about ~1/110 times. If there are 10 players at the table, they will each have quads about 1/1100 times 1*10/1100 = 1/110. For every case the same thing can be done whether there are 3 of a kind on the board one pair or 2 pair.

[/ QUOTE ]

I think this reasoning is flawed due to independence issues. Can someone else chime in?

-RMJ

Re: An Update.
 

You definitely need to apply the inclusion/exclusion (or similar) principle here to get an exact answer. I was just reaching for a ballpark estimate that might be within the first standard deviation when propogated over a few thousand hands.