| Author |
Message |
|
Posted: Fri, 23 Jun 2006, 12:33am Post subject: Analyzing A3s vs A2s... (interesting results)
|
 |
|
Two Pair
Joined: 17 May 2006
Posts: 38
WPP: 288
|
|
I've always had a suspicion that the poker site I play on is "rigged".
Of course, this could just be me being paranoid, but now I've finally found some interesting results when analyzing my starting hand stats via PokerOffice.
It seems like A2suited would be better than A3suited, since it is connecting, right? They're not really good hands in the first place, we know that, but that's irrelevant.
When I analyzed, this is what I came up with.
I was dealt A-3suited a total of 57 times, with a winrate % of 52%.
The total winnings with this hand is about $63.
vs.
I was deal A-2suited a total of 71 times, with a winrate % of 39%.
The total losses with this is hand is (negative) -$42.
Is that a mere coincidence? I thought so. Then I analyzed a few more.
AJoffsuit vs. AJsuited. Clearly the latter is better, correct?
Ace-Jack/off, dealt 169 times, 63% winrate, winnings = $83.
vs.
Ace-Jack/suited, dealt 71 times, 45% winrate, losses = -$70.
It COULD be just a result of the way I play the hands, maybe I get attached to the suited AJ more easily, and it's harder for me to let it go? Perhaps, but what if it's possible that certain hands are dealt to hit the flop better than others? To generate action, to generate rake? Anyone with AJsuited is going to throw some chips away to see the turn even if there's just ONE of that suit on the flop and no AJ. What if AJ offsuit always spikes a J or A on a rainbow flop? I can't prove this, but it's an interesting theory.
Another interesting analyzation would be 99 vs 88.
Finally, the software is logical.
I've won more with 99 than 88.
but since the former is slightly better than the latter, then it also raises the question... will I hit sets more often with 9's than I will with 8's? In real life brick-and-morter poker, this would never be logical. However, in a world where we play online and we're trusting software algorithms to deal us supposedly random hands... it does raise some questions.
Give it some thought, offer some feedback, and thanks for reading. =) |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 12:43am Post subject:
|
|
|
3-of-a-Kind
Joined: 09 Jun 2006
Posts: 96
WPP: 91
|
|
A3s is a better hand then A2s there both connecting cards since they can both only make 1 straight.
Although A3s vs A2s would only be better if both pairs hit otherwise its a split. I treat these hands as identical. |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 12:46am Post subject:
|
|
|
3-of-a-Kind
Joined: 09 Jun 2006
Posts: 96
WPP: 91
|
|
| Actually theres more I just realized this. A3s is a better hand because a straight comes as 245 while A2s comes as 345. Theres two hands that beat the A2s flop while there only one that beats the A3s flop. Also theres a higher chance someone will be playing 67 then 36. |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 2:36am Post subject:
|
|
|
Full House
Joined: 19 May 2006
Posts: 722
WPP: 43
Location: Philly
|
|
| You would need a massively larger sample size to see the actual difference between very similar hands. |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 5:08am Post subject:
|
|
|
4-of-a-Kind
Joined: 22 Jun 2005
Posts: 2195
WPP: 52
Location: NZ
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 7:52am Post subject:
|
|
|
4-of-a-Kind
Joined: 28 Feb 2006
Posts: 2958
WPP: 109
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 10:00am Post subject:
|
|
|
Flush
Joined: 24 Jun 2005
Posts: 286
WPP: 118
|
|
| Based on his numbers, this represents about 20k hands played. Not huge, but fairly significant...my experience has been that PT data tends to normalize around 15-20k hands. So, while some of it is sample size, the larger issue is how the hands are played, esp AJ vs. AJ sooted. AJ sooted can be a huge loser if you are always chasing a flush at any cost. A2 vs A3 sooted is a coin toss and therefore one hand could be more profitable than the other at any given point in time. Not sure it has anything to do with "rigged"...just depends on how you play the hands in my opinion. |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 10:04am Post subject:
|
|
|
HABITUAL LINE-STEPPER
Joined: 13 Jan 2006
Posts: 5226
WPP: 74
Location: Atlanta, GA, USA
|
|
the general data does normalise around 20k hand
however, you need a 500k sample to have accurate specific hand information.
I have 120k and i can still see variance in specific hands.
The longer I play the more I realize how little 5k, 10, 20k, or even 50k hands actually means. |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 12:01pm Post subject:
|
|
|
Straight Flush
Joined: 21 Jun 2005
Posts: 7707
WPP: 71
Location: Petra Marklund FTW ^^^^
|
|
| Renton wrote: | the general data does normalise around 20k hand
however, you need a 500k sample to have accurate specific hand information.
I have 120k and i can still see variance in specific hands.
The longer I play the more I realize how little 5k, 10, 20k, or even 50k hands actually means. | QFT |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 1:15pm Post subject:
|
|
|
Straight
Joined: 19 Jul 2005
Posts: 122
WPP: 80
Location: NYC
|
|
My "analyzation" is that yours posts invariably make me want to cry.
Don't be paranoid, just play good cards. |
|
|
|
|
|
|
|
Posted: Fri, 23 Jun 2006, 4:39pm Post subject: Re: Analyzing A3s vs A2s... (interesting results)
|
|
|
4-of-a-Kind
Joined: 15 Mar 2006
Posts: 2664
WPP: 84
|
|
| Atlplayer85 wrote: | | It COULD be just a result of the way I play the hands, maybe I get attached to the suited AJ more easily, and it's harder for me to let it go? |
Indeed this is it.
If you want to prove a theory about the pokersites rigging flops etc, then you would need to do the following:
- determine how they rig them. You say 'to generate more action', but in practical terms, what would they do.
- if they rig it, then there would have to be a discrepency over a very large sample with what would be normally statistically expected.
More to the point, if you hold AJs, the odds* that the flop has
- none of the suit: 46.6%
- one of the suit: 41.6%
- two of the suit: 10.9%
- three of the suit: 0.84%
So.. go ahead and round up a large enough sample and see if the flops are rigged. If they differ significantly from what the odds imply, then we'll talk.
* the math:
for one of the suit, 11 suited cards left out of 50 unknown cards. So 39 of them are offsuit to your hand.
none on the flop= 39*38*37/50*49*48
one suit on the flop= (39*38*11/50*49*48)*3
=> three possible configurations to fit the suit into the offsuits
two suits on the flop= (39*11*10/50*49*48)*3
=> again three possible configurations
three suits on the flop= 11*10*9/50*49*48 |
|
|
|
|
|
|
|
Posted: Sun, 25 Jun 2006, 10:58pm Post subject: Re: Analyzing A3s vs A2s... (interesting results)
|
|
|
3-of-a-Kind
Joined: 21 Mar 2006
Posts: 61
WPP: 106
Location: Japan
|
|
| Atlplayer85 wrote: |
Ace-Jack/off, dealt 169 times, 63% winrate, winnings = $83.
vs.
Ace-Jack/suited, dealt 71 times, 45% winrate, losses = -$70.
Anyone with AJsuited is going to throw some chips away to see the turn even if there's just ONE of that suit on the flop and no AJ. |
I think I figured out why you're losing money with AJs, but not AJo. Don't chase runner-runners. |
|
|
|
|
|
|
|
Posted: Mon, 26 Jun 2006, 1:31am Post subject:
|
|
|
4-of-a-Kind
Joined: 15 Jul 2005
Posts: 3248
WPP: 165
Location: Ohio
|
|
|
|
|
|
|
Posted: Mon, 26 Jun 2006, 8:00am Post subject:
|
|
|
4-of-a-Kind
Joined: 22 Apr 2005
Posts: 3173
WPP: 88
Location: Putney, UK; Full Tilt,Mansion; $50 NL and PL; $13 and $16 SNGs at Stars
|
|
This thread is teh funny 
Let's imagine for a second that online poker IS rigged. The only logical reason for this is for the house to take more rake, right? So it would have to both provide hole cards and flops that bring action.
However, the irony in this is that it DOESN'T FAVOUR ANY SINGLE PLAYER ABOVE ANY OTHER. The players are irrelevant - the house doesn't give a toss who wins and who loses! So you can say with certainty that even in a rigged game, you will win and lose, long term, just as often as you would if it were completely fair. |
|
|
|
|
|
|
|
Posted: Wed, 28 Jun 2006, 6:32am Post subject:
|
|
|
Full House
Joined: 29 May 2006
Posts: 1301
WPP: 111
Location: surfing in a room
|
|
|
|
|
|
FTR 5 Year $300 Freeroll on Full Tilt Dec 4th + FTR Gear!
