I was thinking about your algorithm (I've done a psuedoRNG simulator for shuffling a little differently in Perl). What size number would you need to generate to avoid getting bias from the mod operations at the endpoints? I'd think you want to do the largest number of significant digits your processor can do, which still may not be enough with this algorithm. So if you generate a rand between 0 and 1 to 32 decimal points, then multiply it to move the decimal to the far right, then use the mod operation, I think the endpoints values in your ranks and suits are going to be under-represented because they only have half a range to round from. Amirite?

There are ways to overcome that.

Well, the suit is not a problem because, although I've written the 'sum' as x % 4, you actually just mask off two bits which will evenly cover all requirements.

For the rank, yes, you would want to generate as many bits as you could.

Clearly if you only generated 5 (32 numbers) you would over time find that the ranks 0 to 5 were significantly overrepresented.

For normal simulations 32 bits is quite adequate as the error is well under 1 in a hundred million.

I would imagine a poker server would use a 128 bit RNG so that the error would be infinitesimal and you'd need to run it for longer than the age of the universe to detect it.

The other main possibity is to just reject the numbers that are causing the overrepresentation (e.g. 26-31 in the 5 bit example).

Thanks qpw.

People oftentimes they post here that "the sites are rigged" after they lose a big hand in hard-luck fashion like runner-runner. Typically their arguments make no sense at all - WHY would the site "rig" the hand in favor of your opponent? What would they really gain by pleasing that player but at the same time pissing you off royally? "Nothing" is the obvious answer, and the claim of "its rigged" is rejected.

My theory finds a middle ground where no, of course the sites don't favor one player over another. But they don't have to... They can simply alter the card distribution on the board ever so slightly so that it is "more clumpy" - cards of similar ranks and suits will hit the board slightly more often together. This produces slightly more straights, more flushes, more sets, etc. Nobody is favored by this because it's still a "random" draw but the overall variance of the game goes up and the rake stays high as more money "churns" thru the poker community. Think of a kettle of water. The "true random deal" kettle is calm, the water is still. The "clumpy random deal" kettle is bubbling, frothing, churning. The water is moving around quite a bit about even tho it's basically going nowhere, contained in the kettle.

Obv none of this is profound or unique, and there's probably been similar posts on this topic here but I'm too lazy to go look for 'em. I don't read a lot on this subject, but this IS what I believe. Contrary to some "its rigged" schemes, this "clumping" behavior would be absolutely trivial for the programmers to implement - this is another area where some of the "its rigged" arguments fall flat. But the main reason I believe this theory is my own personal observations in my 100K tournament hands. "Cards of a feather flock together". I haven't analyzed my HEM database to see if I'm a nutcase or not yet tho. haha Someday I'll probably try.

The obvious next question is, if I truly believe this, how do we take advantage of it? And the common sense answer is, if you're on a draw then you've gotta realize your equity is slightly higher than a calculator like PokerStove would tell you... and thus occasionally this means you will keep drawing in situations where conventional implied odds calcs make it a close call but on the "fold" side. I often wonder if THIS isn't the true reason we see some of the HSMTT pros playing a very aggressive, loose, borderline maniac game... they know the secret.

Ok, go ahead and rip me and my theory to shreds, I'm goin' to play teh pokerz.

OK, the problem with your theory is that this sort of tampering would be extremely obvious to anyone doing a fairly rudimentary analysis of hands.

It's extremely easy to take a few hundred thousand hands and determine if the table cards are truely random combinations.

Just enjoy playing teh poker.

no you are wrong

thanks for driving through

dude i have't hit a draw since 1972

Cliff notes:

If only there was a giant thread dedicated to this type of thing...

It could easily go gold.

think of how sweet it'll be when you do at the expense of a clearly superior player because the site doesn't want you to go broke too fast.

A post in another thread today handily disproves this by showing that the distribution of flops for about 1 billion hands spanning multiple sites, looks exactly like it was expected to look.

Next?

################################################## ##############################################
Stage #1616385812 Tourney ID 1950607 Holdem Single Tournament No Limit 60 - 2009-04-25 17:57:19.015 (ET) [ 2009-04-25 17:57:19 ]
Table: 28522693 (Real Money) Seat #4 is the dealer
Seat 5 - (1040 in chips)
Seat 6 - HOLLYWOOD7 (4253 in chips)
Seat 8 - T CASH THUG (4796 in chips)
Seat 3 - 1G0TASSRAKED (1350 in chips)
Seat 4 - GRIFROG (2061 in chips)
- Posts small blind 30
HOLLYWOOD7 - Posts big blind 60
*** POCKET CARDS ***
Dealt to [Jd Ks]
T CASH THUG - Folds
1G0TASSRAKED - Folds
GRIFROG - Raises 180 to 180
- Calls 150
HOLLYWOOD7 - Folds
*** FLOP *** [7h 8c Kh]
- Checks
GRIFROG - Bets 420
- All-In(Raise) 860 to 860
GRIFROG - Calls 440
*** TURN *** [7h 8c Kh] [Qd]
*** RIVER *** [7h 8c Kh Qd] [5s]
*** SHOW DOWN ***
GRIFROG - Shows [5h 5d] (Three of a kind, fives)
- Shows [Jd Ks] (One pair, kings)
GRIFROG Collects 2140 from main pot
*** SUMMARY ***
Total Pot(2140)
Board [7h 8c Kh Qd 5s]
Seat 3: 1G0TASSRAKED Folded on the POCKET CARDS
Seat 4: GRIFROG (dealer) won Total (2140) HI2140) with Three of a kind, fives [5h 5d - B:5s,P:5h,P:5d,B:Kh,B:Qd]
Seat 5: (small blind) HI:lost with One pair, kings [Jd Ks - P:Ks,B:Kh,B:Qd,P:Jd,B:8c]
Seat 6: HOLLYWOOD7 (big blind) Folded on the POCKET CARDS
Seat 8: T CASH THUG Folded on the POCKET CARDS

Stage #1616436398 Tourney ID 4113198 Holdem Single Tournament No Limit 60 - 2009-04-25 18:27:14.013 (ET) [ 2009-04-25 18:27:14 ]
Table: 28523236 (Real Money) Seat #2 is the dealer
Seat 4 - NJVCOOL (4043 in chips)
Seat 7 - (3052 in chips)
Seat 8 - CHAEGOECARDS (3160 in chips)
Seat 9 - OMG_DZACHO (2415 in chips)
Seat 2 - PYRO013 (830 in chips)
NJVCOOL - Posts small blind 30
- Posts big blind 60
*** POCKET CARDS ***
Dealt to [10c 10d]
CHAEGOECARDS - Calls 60
OMG_DZACHO - Folds
PYRO013 - Folds
NJVCOOL - Calls 30
- Raises 180 to 240
CHAEGOECARDS - Calls 180
NJVCOOL - Calls 180
*** FLOP *** [Qc 3d 4d]
NJVCOOL - Checks
- Checks
CHAEGOECARDS - Bets 100
NJVCOOL - Folds
- Raises 200 to 200
CHAEGOECARDS - Calls 100
*** TURN *** [Qc 3d 4d] [10h]
- Bets 860
CHAEGOECARDS - Calls 860
*** RIVER *** [Qc 3d 4d 10h] [2d]
- Checks
CHAEGOECARDS - Bets 500
- Calls 500
*** SHOW DOWN ***
CHAEGOECARDS - Shows [6d Jd] (Flush, jack high)
- Mucks
CHAEGOECARDS Collects 3840 from main pot
*** SUMMARY ***
Total Pot(3840)
Board [Qc 3d 4d 10h 2d]
Seat 2: PYRO013 (dealer) Folded on the POCKET CARDS
Seat 4: NJVCOOL (small blind) Folded on the FLOP
Seat 7: (big blind) HI: [Mucked] [10c 10d]
Seat 8: CHAEGOECARDS won Total (3840) HI3840) with Flush, jack high [6d Jd - P:Jd,P:6d,B:4d,B:3d,B:2d]
Seat 9: OMG_DZACHO Folded on the POCKET CARDS
################################################## ##############################################
Stage #1616438285 Tourney ID 4113198 Holdem Single Tournament No Limit 60 - 2009-04-25 18:28:21.012 (ET) [ 2009-04-25 18:28:21 ]
Table: 28523236 (Real Money) Seat #4 is the dealer
Seat 7 - (1252 in chips)
Seat 8 - CHAEGOECARDS (5200 in chips)
Seat 9 - OMG_DZACHO (2415 in chips)
Seat 2 - PYRO013 (830 in chips)
Seat 4 - NJVCOOL (3803 in chips)
- Posts small blind 30
CHAEGOECARDS - Posts big blind 60
*** POCKET CARDS ***
Dealt to [As Kh]
OMG_DZACHO - Folds
PYRO013 - Calls 60
NJVCOOL - Folds
- Raises 210 to 240
CHAEGOECARDS - Calls 180
PYRO013 - All-In(Raise) 770 to 830
- All-In(Raise) 1012 to 1252
CHAEGOECARDS - Calls 1012
*** FLOP *** [4d 3h 7s]
*** TURN *** [4d 3h 7s] [Kd]
*** RIVER *** [4d 3h 7s Kd] [3d]
*** SHOW DOWN ***
PYRO013 - Shows [Qc Qh] (Two Pair, queens and threes)
- Shows [As Kh] (Two Pair, kings and threes)
CHAEGOECARDS - Shows [9d 10d] (Flush, king high)
CHAEGOECARDS Collects 844 from side pot-1
CHAEGOECARDS Collects 2490 from main pot
*** SUMMARY ***
Total Pot(3334:2490,844)
Board [4d 3h 7s Kd 3d]
Seat 2: PYRO013 HI:lost with Two Pair, queens and threes [Qc Qh - P:Qh,P:Qc,B:3h,B:3d,B:Kd]
Seat 4: NJVCOOL (dealer) Folded on the POCKET CARDS
Seat 7: (small blind) HI:lost with Two Pair, kings and threes [As Kh - P:Kh,B:Kd,B:3h,B:3d,P:As]
Seat 8: CHAEGOECARDS (big blind) won Total (3334) HI3334) with Flush, king high [9d 10d - B:Kd,P:10d,P:9d,B:4d,B:3d]
Seat 9: OMG_DZACHO Folded on the POCKET CARDS

Dude calls a 3 bet all in for over 20x with 9 10 suited 3 handed and magicly connects a flush....?

What are you trying to get at?

He had 21% equity when the money went in. Is there a point to your post other than you got donked? And he had 41% equity against just your AK if the other guy wasn't in. Nothing magical about losing with your AK no-pair hand that you put all your chips in with.

About callin a 3 way all in for that price? Your wrong. So this is how you play? You justify this? I could light this thread up over and over with hand histories....I'll save the trouble, but I encourage others to do so...

I didn't say he made a smart play, but he made one that will win 21% of the time and you use it as evidence of something other than a fair game. He also had 4x the chips you had, and he could afford to take a chance with a suited connector 1 seat from the bubble with a giant stack. Smart, maybe not, but not amazingly terrible either. His hand has almost the highest all-in equity against AA of any possible opposing hand (only bettered by 98s - 76s).

Dude has played 237 SNGs on AP to make a whopping $7. Guess it's not quite rigged enough for him.

Do you think maybe, just maybe, he might be a bad player?

I just found out anyone can now buy over a billion cash hand histories at PTR. You can get a million Full Tilt and a million Stars for $106, at any stakes you want. Time for the ...ahem, skeptics to find the proof you so desparately seek.

The evidence awaits you....
Glory could be yours!!

What buy in should I play so i dont run into this type of play?

Who said anything about the buyin?

And why would you want to avoid people who call all ins preflop with T9s?

Dude, you were the underdog to a QQ and you are complaining because the QQ didn't even win. The third player is almost irrelevant to the hand from your standpoint, you got all in with the worst hand and you lost.

Move up to $500/$1K and play with folks who would never call two all-ins with T9s one seat from the bubble. You won't win, but you won't have that happen either. But you should welcome players like that, not avoid them.

I'm sorry I just don't have any sympathy for your non-bad-beat story.

Instead of worrying about this..perhaps you should stop hopping around different limits so much. hopping up and down from 50centers to $10 sngs probably doesn't really offer up much for getting better.