One more thing, AATT loses against AAA. Explain how you lost with a set?
Exactlyā¦and not the first time Iāve seen a glitch of that nature here, but 1st one that happen to me.
If your AAA lost to AATT, then you should report this hand to Replay immediately. I would actually be interested to see this hand myself, so it would be nice if you could post the hand number. I have never experienced this type of glitch myself.
Seen a lesser straight rake the chips against a higher straight twice and seen 4 deuces beat 4 fours once. All software glitches at times (no matter what a dev might say 
). Not something I see regularly, so just chalked it up to occasional glitches. And as you said before, online pokers does proceed faster than real in most cases, by the time you figure āI should have took a screenshot of that!ā and can react to take one, youāre already many, many hands after the fact. Well got to get back to seeing how many stupid ways I can loseā¦ 
You donāt actually need to take a screen shot. The Replayer saves about one week of hand histories, and you can go back and save these hands. I encourage you to do so for any hands that are amiss, and forward them to the devs and/or post them in the forums for inspection.
1 Like
yes thanks
i looked at the replay rng certificate
it is dated 2011 thats ten years ago
perhaps a current one would be reassuring?
Thanks @krista, I donāt know how it could be any more reassuring than the current one.
As I said previously, we know that Replay previously had an algorithm that got certified so all they have to do is submit that algorithm for re-certification. I really donāt see the point of the exercise!
If we believe that Replay are ārunning an honest gameā then the old certificate is quite sufficient. If we do not believe then even a new certificate does nothing to dispel those doubts.
The doubters really need to collect data for about 100k hands and submit them for analysis.
Hope this helps.
Regards,
TA
Youāre doing well @Boo2u2 ā¦ I have played just over 100k hands in one year and have yet to see anything at all out of the ordinary.
There are many others who have played many multiples of that number of hands without seeing anything at all out of the ordinary.
I also would like to see the hand number. I looked through your history, all 20 pages, and didnāt see any results where AATT won a hand that you lost.
Of course I recommend that you follow AKFolds advice by noting the hand number and reporting it. I have had several discussions with the support staff here and my experience is that they are extremely helpful and friendly.
Regards,
TA
3 Likes
Fairness isnāt a problem - itās the same for everyone, and thatās the very definition of āfair,ā so Iām not here to whine about why my pocket aces got cracked (it happens) or why someone sucked the one and only out on the river he had to beat my King high flush (again, it happens). But I am here to offer some (hopefully) constructive observations for the creators of the game - thing Iāve noticed that do seem a bit too weird and indicate that the algorithm used for this game could use some tweaking.
Some of the algorithm is probably the way it is to keep things interesting - and I get that - too many crappy hands all the time and it would get boring for a lot of players.
But anyway, I generally play the S&G turbo three player tournaments, and hereās what I see:
-
Beats on the river, bad beats on the river - they happen in real games. But this game hands out beats on the river like Halloween candy on October 31. Iāve been both beneficiary and victim more times than I can count, so Iām not complaining - just an observation that it happens much, much more often than would seem natural. And Iām talking āall he/she has is one outā kind of beats, and they happen all the time . Sure, beats on the river happen - but here they happen so often that it has become what I usually expect to see - Iām really never surprised anymore.
-
For whatever itās worth, in a 3 player tournament setting I wouldnāt expect to see, from hand to hand, as many straights, flushes, sets, etc. - basically, I wouldnāt expect to see as many strong hands as I see. They seem way too numerous to be realistic - and Iām not saying I see one in a hand - I see two players with them quite often - too often. Again - part of the algorithm might be to ākeep it interesting,ā I donāt know - itās just my unscientific observation.
But thatās it. Itās fun poker site, and maybe thatās enough and we should leave it at thatā¦ But those two things, to me, are the areas that the algorithm might need looking at if you want more realism. But then again, the sit is fun and free (if you want it to be free), so this is most certainly not a complaint. Itās just a suggestion of two areas where I see unrealistic play, and wouldnāt mind a bit more realism.
1 Like
What Elviod just said! He just said it better than me! Donāt feel site is rigged for me or against me, more āriggedā to make more āexciting and interestingā play. Side effect being, things feel āoffā and āartificialāā¦
1 Like
hi hi!!!
elvoid you stated the apparent situation brilliantly thank you
no one suggests the site is āriggedā for certain players to be more successful
it appears to me & others that the rng (random number generator) is predisposed to produce more āactionā & interesting handsā¦ surely we have seen many improbable hands; quads over quads, full houses over flushes, flushes over trips in the same tournament?
someone said the probability to have 2 players with quads in the same hand would be equivalent to being hit by lightning twice?
so what wrong with that? makes for more interesting play & fun
i disagreeā¦
many like myself play on replay, as itās inexpensive to try out various strategies & tacticsā¦ prior to migrating new concepts to $usd sites
if the play on replay does not mirror statistical & probability realitiesā¦ it negates that possibility
many on replay have financially invested by purchasing chips
with the explicit understanding that the play is realistic of actual poker
if the purpose is to have interesting, big hands then perhaps playing on replay events where small cards are removed might be more āfunā (royal games i believe they are called)
to suggest that the replay rng certified ten (10) years ago, would be applicable todayā¦ really? the software has had major changes, in migration away from flashā¦ same rng & algorithms applicable now?
very easy to alter a rngā¦ if these cards have fallenā¦ then make these type of cards more likely
granted my personal observations of statistically improbable events happening in replay are anecdotal, however i have discussed & see others here echoing similar concerns
very simple solution - replay poker - have your rng currently certified & i sure concerns would be alleviated
regards
krista
1 Like
Boo nicley summed up the feeling you get (āoffā and āartificialā) when playing, and kcorbee and I are certainly on the same page.
Playing virtually nothing but the 3 player sit&go, I am as certain as I can be that the cards are ājuiced.ā There is no way that many good hands keep showing up hand after hand. Donāt get me wrong - there are stinkers, too - of course not every hand is juiced - just an inordinate number of them.
It doesnāt hurt me so much - I play accordingly and do fine. Win some, lose some, all good. I just donāt feel like Iām playing anywhere near ārealā poker - Iām playing arcade poker.
Maybe the problem is my expectation - maybe that is too high a bar for a computer poker game? It is what it is - no one is forcing me to play.
Arcade poker is fine if everyone knows thatās what they are playing - but Iāve had good players comment to me in play ādo you even know what you are doingā or ātough combination to beat - lucky and bad,ā or things like that. I know EXACTLY what Iām doing. The problem is, they think theyāre playing something that resembles real poker - theyāre not - and they have not adjusted accordingly. And theyāll keep losing (mostly) until they realize that. (Again, I only speak for what I see in the 3x3 tourneys)
Should clarify something I said above - when I said āthey think theyāre playing something that resembles real poker - theyāre notā that was an overstatement, and i didnāt mean it to be. This game certainly āresembles real poker.ā I just mean that they are playing something closer to what Iād call āarcade pokerā than āreal poker,ā and there is a difference in how those two should be played.
Thanks for moving away from āReplay is riggedā - that particular dead horse has been whipped so much, thereās not even bones left now!
This suggestion that the algorithm is somehow programmed to make the game more in āinterestingā is interesting 
It seems that you, @Boo2u2 and @krista agree that thereās no particular bias, just more āgoodā hands than you would expect to see.
The problem is that the āalgorithmā doesnāt have control over who bets, when they bet and how much they bet. Equally, every player at the table is independently able to choose to fold their hand at any time.
This is a problem for an āexcitingā algorithm - the cards have been set up to show, say, a flush on the river beating AAA. Letās imagine that Iām holding AA and raise preflop to a fairly large amount. One player calls, holding 9c4c and everyone else folds. The flop comes AcTd2h. Itās āobviousā that the āalgorithmā is going to show clubs on the turn and river cards to complete the backdoor flush. However, the player who is āsupposedā to win doesnāt know this and folds when I bet 2x pot on the flop. The independent action of this player has completely āspoiledā this āexcitingā hand.
Moving on, if there is no bias in respect of a particular player or group of players then these ābad beats on the riverā need to be distributed fairly evenly amongst all of the participants. Curiously enough, if the PRNG was producing a random distribution, I would expect to see ābad beats on the riverā evenly distributed amongst all of the participants exactly as you describe!
The fact is that something like 72o is going to beat AA something like 15% of the time! You have AA, shove preflop and the clown with 72o calls for no apparent reason. 3 times out of 20, the clown is going to wipe you out! That is perfectly normal and expected.
This is where we get to the tendency of people, including me, to remember ābad beatsā far more than we remember āgood beatsā (is there such a thing? lol)! I truly canāt remember the last time that I won with AA ā¦ itās happened, as expected, a fair bit of the time and there is nothing at all memorable about yet another win. However, I clearly remember the exact scenario that I described above happening to me in a tournament about 4 months ago. I had aces and was middle stack on the table and villain had me covered. I shoved, villain called with 72o (!!!) and knocked me out. I was furious! It still annoys me!
Yes, I know and so do you now, that thereās no guarantee of winning with AA but getting knocked out of a tourney AA vs 72o really hurts. Iāve had aces cracked by other āunlikelyā hands as well and, without too much effort, I could remember quite a number of them. My memory tells me, now that Iām thinking about it, that I seem to losing with AA at least as often as Iām winning. That is only because winning with hands that I fully expect to win with is not at all surprising and there is no reason for me to pay any attention to that happening.
Please, try to not overthink this! Bad beats happen, as you correctly said, and we have a very high bias towards remembering those hands and emphasising those hands over time. That is just the way our minds work.
More to come but I have to go to work now!
Regards,
TA
2 Likes
Thanks for the info. I absolutely agree that we tend to remember the things that stand out much more than the things that are generally mundane, and this can often give a āit happens all the timeā feel when in fact, it doesnāt āhappen all the time.ā
The ājuiced cardsā scenario is not rampant to me - just noticeable. Letās put it in this wholly unscientific, unsupported by data, but observationally anecdotal way:
In this game, if Iāve got a great hand (not the nuts, but a great hand), and weāve gone all in, after the turn or the flop so now I can see what Iām up against and now the draw will determine, it seems the odds of getting beat in this game are much greater than in a real game - that it simply happens more often than I would expect it to naturally happen.
That doesnāt mean that in that particular situation I donāt win most of them - I do - anyone does. But a 50 percent change in those odds of losing (say from 20 percent up to 30 percent) is noticeable.
Put another way - when Iām on the bad side of that āall-inā and the only thing that can save me is, say, a 7 on the river; or I need two particular cards on both the turn and the river to win, though i know Iām going to clearly lose this situation the majority of the time - in this game, it seems like I still hit it just enough to where it feels a bit odd. And Iāve hit that kind of thing just enough to be much more hopeful that Iām going to suck out what I need than I would be in a real game.
Another analogy would be that in this game, I fold far less (about 51 percent according to my stats) than I would in a real game. And itās not because the money is fake and it inspires me to take more chances on marginal hands - itās that Iāve seen marginal hands win much more often than Iāve ever seen in real games, so Iām more inclined to give them a run at times - or at least take a look at the flop if Iām OK stack-wise. Iād get killed at a real table if I played 50 percent of my hands to conclusion.
In any event - the discussion about realism is fun, but in the end, the site is fun - the game is fun - I just play it a bit differently.
Elviod is reading my mind and saying the things I been trying to say. Especially the part about hands willing to try to see till the shootout. Feels like I have to play ābass-ackwardsā in order to win most times. Games feel more like how a slot machine works than a deck of cards.
And if everyone else is playing the same amount of hands, what do you think will happen? If youāre playing 50% of hands, you are going to be holding the wrong end of the lucky stick a massive amount of time.
Putting some maths to this: If thereās only 4 players, each of whom are playing the top 50% of cards, each individual has just 25% chance of winning if everyone plays through to the river. If thereās 6 players, each individual has 16% chance of winning if everyone plays through to the river! Those ājunkyā and āmarginalā hands are going to get through far more often than you, and most other people, expect!
You will notice far fewer junky and marginal hands taking the pot if you narrow your range and raise more aggressively preflop. Even at the lowest stakes tables, people start to fold some of their junk to a 10BB preflop raise. If the flop hits you hard, bet hard! You will still lose a noticeable amount of the time, thatās just poker and the PRNG doing random things, but you will start beating out those junky and marginal hands far more frequently.
Hope this helps,
Regards,
TA
I donāt understand why replay poker, a free website that doesnāt take money unless you buys chips would change its algorithm to be bad beats for everyone. Iām not an experienced coder, but Ive coded enough to know how hard it would be to make it so that there was just enough combinations for each hand so that it wasnāt that obvious that it was rigged but rigged enough to be rigged. Those calculations would be hundreds of times harder than if you just did a random number generator for each hand which is very simple. What is the replay team getting out of it? They dont want to lose their honor of being a fair site and that is a risk much worse than trying to generate action at a free poker table. I would ask someone to do a few thousand or so hands (which is probably not enough but if it was rigged it would start to show) but puggywug already did this on another forum and it showed this site is completely random! Here is the link right here:
This is because there is nothing to understand about Replay; the fault lies in other peopleās cognitive bias. The statistical probability of AA getting cracked, for example, does not change on Replay, but AA appears to get cracked more often because (1) people are playing a large number of hands per hour, which increases how often we will see AA get cracked over time. The higher frequency at which we observe AA get cracked is related to the number of people who see flops and subsequent streets, which is related to the loose and passive play that prevails on free poker sites. Both of these factors contribute to cognitive bias as people are making assumptions based on variance or the illusion that the game is somehow rigged or juiced for action.
Itās like you have aptly stated WATCHOUT8. It would take an enormous amount of work to construct an algorithm that singled out particular players or changed the deal based on their actions. This work costs time and time costs money. There is absolutely no benefit to Replay for doing this, since it is possible to play on this site without ever spending a dollar on chips.
The real problem is that people would rather blame the system for their losses, rather than their own play. In a previous post I talked about ātaking ownershipā of my losses, even the bad beats. This involves scrutinizing my play and figuring out what I did wrong and how that contributed to the loss. Of course, sometimes losing in poker is inevitable, no matter what you do, but analyzing your losses is always a good practice.
To put this in perspective, I aim to win about 60% of the hands I get involved with, which probably seems pathetically low to some readers. Still, this is enough to chip up over the long-run, although I still hold myself back with bad decisions. Letās not forgot that we should not conflate bad outcomes (losing) with a bad decision. Yesterday, I called a 3-bet shove preflop with AK suited, and lost to a player with JJ in the hole. I called the preflop shove because I knew (from previous observations) that AKs either dominated my opponentās shoving range outright (hands such as AQ, KQ, KJ) or was a coin flip (TT, JJ) enough times to make this a +EV call.
4 Likes
So Iāve added a very small, statistically insignificant, probably worthless but āhey thereās an ice storm outside, Iām not playing golf this weekend so here it is cause what else do I have to doā experiment.
And the results are contrary to my original feelings on this - I think it is clearly more of a case of what we remember vs. what we donāt - and while I always agreed that of course we remember the stronger hands, the bad beats, etc. more so than we remember anything else, it still seemed to me that the cards were ājuicedā for stronger hands in general for nothing more than keeping play interesting (an understandable goal for a web based game) - not any bearing on fairness - just keeping it interesting. Iāve never said anythin was āfixedā or any players were targeted to win or lose - only that the cards were generally more fun and ābettableā (ok, I know thatās not a word) than you would expect naturally - keeps the game interesting, but it sure does suck that time your ace high flush got beat by a full house that fell on the river (yes, I still remember that one, and I recognize it furthers the āwhat we remember vs. what we forgetā argument).
Well, I donāt think anything is ājuicedā anymore. Hereās why (again, pretty small sample - but still - it changed my mind):
I went back and counted what I would call the āstrongā hands from my last 118 hands at 3 player sit&go turbo. Of course, these were only the hands I got to see, not those that were mucked - but anyway, hereās the count:
118 hands (40 to showdown, so I really only had a 40 hands of data):
Three of a Kind: 4
Straight: 3
Flush: 2
Full House: 1
Frankly, I was surprised the numbers were so low - I expected more - I remembered (aha!) more. Now there were probably a few more of the sort that never made it to showdown, but I didnāt get to see those - but I doubt it was a drastic number - but I donāt think it matters, because of the following:
Broke out the real cards, started shuffling. Started dealing, and yes, re-shuffled after each hand. Three players, dealt to showdown, recorded what the hands were. I got to 35 hands and stopped. Why?
Because after 35 hands, I had already recorded:
Three of a Kind: 2
Straight: 4
Flush: 2
Full House: 0
I believe Iām seeing something closer to ānaturalā than ājuiced.ā
Small sample, I know - but I do have a life, even in an ice storm. āThe Outlaw Josey Walesā starts in 10 minutes. Hey - I didnāt say how exciting a life, just a life.
3 Likes