After spending a couple of days back at PKR, playing the low level cash games as a break from Full Tilt, and I have to voice a concern about the shuffle. I just re-loaded and then withdrew my funds again after a few hours.
Now I am sure the RNG is sound and fair, I don't suspect any rigging. But I did notice stuff that I noticed many months ago and voiced my opinion on.
I am talking about the amazing number of straights and flushes that seem to hit there. And if one doesn't hit, the board certainly makes sure there is a draw.
Now according to Poker Office, I have played over 7,500 hands on Full Tilt now, and in all that time I have not seen anything like what I see on PKR in just a few hundred hands. My concern is that there may be an error in the shuffle procedure.
While playing there, I had holes cards that a vast majority of the time were connected (10-J, J-Q, 7-8 etc). If I called with them, there seemed to be a high chance I got a straight draw. It is very common to see a board of 7-8-9-10. And also it was very common to get 3 or 4 of the same suit. Many times last night, there were open ended straight draws and flush draw possibilities. Other players at the table were noticing the same thing.
Checking the hand history from one session, there were 127 hands played. In those hands, there were 12 'known' straights. By that, I mean straights that were shown. I know I had 2 or 3 that I didn't show in the hand history but I never show my hand if it isn't called so there may be others that weren't seen.
Flushes were similar, with 8 'known' ones. Again I had 2 or 3 that I didn't show.
Checking my Poker Office, in those 7,500+ hands, I have had only 40 straights and 31 flushes. These were in Sit and Go's. So I hit a straight about 0.5% of the time. That is significantly lower than the percentage on PKR. (I think I had 5 or 6 straights in 127 hands, around 4%-5%)
I have played only 475 hands of cash games on FT. In all those I have had 1 straight and 5 flushes, again significantly less than PKR.
Now I have no idea why any of this is. Like I said, it is something I noticed a long time ago and it did contribute to me stopping playing there. And going back brought it all back.
I also understand that 127 hands is hardly a good amount of hands, but the memories came flooding back.
I am not alone either in my thoughts. A few players I spoke to agreed that flushes and straights came out a lot. A few of them called any connectors because of it, playing the system.
The thing that concerns me is this, extract, taken from the PKR website:
Quote:
|
We create an array with each of the 52 cards in order (2c, 2d, 2h, 2s, 3c... ... Ah, As). We then use the RNG to assign 32-bit random numbers (from 0 to 4,294,967,295) to each card. We then sort the array in increasing order using the number assigned to each card, generating a random permutation of the order of cards in the deck.
|
So why use such HUGE numbers? There are only 52 cards, so why not assign them a number from 1 to 52? PokerStars just use number 1 to 52 to shuffle and they have no problems.
By my thinking, using such big numbers, it is quite easy to get sequential numbers. For instance, the 8c is numbered 2,000,000. Now even if the 9s is numbered 100,000,000, a huge gap, it could still easily be placed next to the 8c because of the size of the numbers involved.
Now it may all be in my imagination, but one thing I do know is that I noticed this many months ago, moved to Full Tilt and didn’t see anything like it in 7,500 hands and moving back to PKR I noticed it all over again. It’s hard to think its just coincidence.
If there is something wrong on PKR, I am 100% convinced it is unintentional. I don’t think for one minute that they would rig anything and I am well past the “online poker is rigged” stage. I personally do not like the shuffle system they use and it could be flawed, creating these scenarios. I can’t see me playing there seriously and it’s probably why so many players have left and despite the fact they say 1 million people have signed up, 7,000 seems to be the peak number.