Testing if a BJ game is rigged
I have a quandary and seek assistance.
My favourite ONLINE game is an Atlantic City RNG Blackjack. the house edge is a mere 0.35% and I've had many successful sessions on that game. Probably ahead overall.
But I cannot get it out of my head that it is rigged to give a better experience at low stakes.
E.g. Typically if I buy in for 25 with an exit goal of 50, I SEEM to win more often that I lose. If I buy in at 50 with a win goal of doubling my buy in, again I tend to win more than lose. These are at low stakes, typically 1,2,5,10 maximum.
But on occasions like tonight, where I buy in for much more, in this case 3 x 250, and wager bigger, typically 25,50,75, I get my ass handed to me on a plate.
It sure feels that they adjust the house edge to encourage low stakes gamblers to ramp up.
I'm not the only one to suspect this and I've done half assed experiments to support the theory. Hence my assertion that I've done well a few times at stakes <=5
Would anyone care to offer a statistically sound test of the hypothesis that higher wagers encounter higher house edge? Preferably a test that doesn't involve losing my shirt?
The general consensus was that the app followed some kind of algorithm that made hands more "exciting" - that is, lots of higher flushes versus lower flushes, river cards completing flushes and full houses simultaneously, high pairs being completed to three or four of a kind then beaten out by higher completed pairs or straight flushes, etc.
Also, the consensus was that the app favored newly created accounts over seasoned ones, within that skewed algorithm.
However, I think all that was going on was that so many hands were played rapid fire that within that pool of so many hands, there was more likelihood that extraordinary events could surface. Something like where if you play enough Baccarat shoes you'll have a higher chance of seeing long runs.
As well, at a recent celebrity poker tournament, I played a pair of jacks where two other players were dealt kings and 10s, and the pair of 10s ended up winning due to a 10's landing. So there we had a hand, very early on, where 3 players went all in with pairs, which when that sort of thing would happen in the poker app I used to play, people would say that it was rigged, not to mention the donk idiocy of going all in before the flop anyway, which is not so smart especially early on in a tournament, but I was stuck at a table of donks who kept going all in every other hand. There are ace + not-suited-low-card donks everywhere, not just on the internet, and sometimes those guys win, doesn't mean that things are rigged.
Quote: MDawgThe general consensus was that the app followed some kind of algorithm that made hands more "exciting" - that is, lots of higher flushes versus lower flushes, river cards completing flushes and full houses simultaneously, high pairs being completed to three or four of a kind then beaten out by higher completed pairs or straight flushes, etc.
link to original post
There was a casino near me that tried offering an "electronic poker table" dealing "Texas Hold-Em".*
The players were quite surprised when they saw more fours of a kind in a few hours than might be seen during the entire WSOP main event, including highlight reels from prior events.
I don't recall the exact details, but the poker room closed, and there were some apologies. Something about the players were randomly ranked at the beginning of each round, and exciting hands were assigned. If the favored player folded, the next ranked player made a hand... etc.
I don't recall the manufacturer (maybe Pokertek? I honestly don't remember, and do not intend to cast aspersions). I think I recall that they were raking every pot, flop or no, to not have a dealer conduct the game.
Cards and chips were simulated on touch screens at each player position.
To complete the donut, this could be the type of exciting poker action the app is attempting to deliver.
*Quotes are used because I doubt the game actually being played was any form of poker; I think the underlying game may have been a variant of spin the bottle.
I once ran my bankroll up to 5k playing 5-25 bets
As soon as I upped to 100 I lost almost every single hand and the entire 5k within an hour
I can lie though running it up to 5k in less than a day was odd IMO playing the smaller stakes
Butt what the hell do I know I have no proof
Quote: MDawgThe general consensus was that the app followed some kind of algorithm that made hands more "exciting" - that is, lots of higher flushes versus lower flushes, river cards completing flushes and full houses simultaneously, high pairs being completed to three or four of a kind then beaten out by higher completed pairs or straight flushes, etc.
link to original post
Calling DRICH calling DRICH
What you described sounds about right for how the hands are constructed at least from what I know about them
wow this looks almost identical to the one Drich developed
https://www.jackpotdigital.com/pokerpro.html
I should also add i looked up the patents for "pokertek" and it is a company - which holds a patent for variable rake poker games and then i got the "jackpot digital" from them being half owner of some of the pokertek patents
Quote: OnceDearIt sure feels that they adjust the house edge to encourage low stakes gamblers to ramp up.
I'm not the only one to suspect this and I've done half assed experiments to support the theory. Hence my assertion that I've done well a few times at stakes <=5
Would anyone care to offer a statistically sound test of the hypothesis that higher wagers encounter higher house edge? Preferably a test that doesn't involve losing my shirt?
link to original post
Before delving into statistics...
Do you have video?
Quote: Dieter
I don't recall the exact details, but the poker room closed, and there were some apologies. Something about the players were randomly ranked at the beginning of each round, and exciting hands were assigned. If the favored player folded, the next ranked player made a hand... etc.
link to original post
I should clarify that much of this was rumor floating around the players.
We had a few players at the office; they were decidedly prone to gossip. They definitely said that the hands they saw were more exciting than usual. They were likely speculating as to why.
The only part I can confirm is that the poker room closed, and is now filled with slots. Any apologies made may have been "We are sorry to announce that the poker room will be closing in November."
To all, I apologise for my poor memory of events from 6+ years ago.
Quote: DieterQuote: Dieter
I don't recall the exact details, but the poker room closed, and there were some apologies. Something about the players were randomly ranked at the beginning of each round, and exciting hands were assigned. If the favored player folded, the next ranked player made a hand... etc.
link to original post
I should clarify that much of this was rumor floating around the players.
We had a few players at the office; they were decidedly prone to gossip. They definitely said that the hands they saw were more exciting than usual. They were likely speculating as to why.
The only part I can confirm is that the poker room closed, and is now filled with slots. Any apologies made may have been "We are sorry to announce that the poker room will be closing in November."
To all, I apologise for my poor memory of events from 6+ years ago.
link to original post
no no no the algorithm works EXACTLY as the player sees it happening... just because the players can describe it though doesnt make it unfair or not random
the players can say whatever they want after the outcome but they never knew the outcome to begin with ... which is IMO the deciding factor to what "fair" actually means
its random and therefore can not be predicted - but the pseudo random number generator can do whatever it wants as long as the outcome or particular outcome is "randomly" chosen with a physical random number generator
Quote: heatmapno no no the algorithm works EXACTLY as the player sees it happening... just because the players can describe it though doesnt make it unfair or not random
the players can say whatever they want after the outcome but they never knew the outcome to begin with
link to original post
Any objection I might have to a game such as I described is based on it purporting to be poker while being something else.
Quote: DieterQuote: heatmapno no no the algorithm works EXACTLY as the player sees it happening... just because the players can describe it though doesnt make it unfair or not random
the players can say whatever they want after the outcome but they never knew the outcome to begin with
link to original post
Any objection I might have to a game such as I described is based on it purporting to be poker while being something else.
link to original post
this is what i am referring to
i think your comment actually makes me think, why didnt they say the word poker in the name of the game - they simply called it "wasioux" yet you look at it and assume your playing poker, and you are, yet it used an algorithm to make the hands more exciting
Quote: heatmapi think your comment actually makes me think, why didnt they say the word poker in the name of the game - they simply called it "wasioux" yet you look at it and assume your playing poker, and you are, yet it used an algorithm to make the hands more exciting
link to original post
Anything that makes the hands more exciting would seem to make it not poker.
The serious players I know would notice the extra excitement, infer a gaff, and tell their casual player friends that something doesn't seem square.
At the end of each showdown, there's still one winning hand, or maybe two that chop. I guess it's still a random way to trade money across the table - it just isn't poker anymore.
OnceDear: Please accept my sincere apologies for hijacking. Inflict such penalties upon me as due.
Quote: DieterQuote: heatmapi think your comment actually makes me think, why didnt they say the word poker in the name of the game - they simply called it "wasioux" yet you look at it and assume your playing poker, and you are, yet it used an algorithm to make the hands more exciting
link to original post
Anything that makes the hands more exciting would seem to make it not poker.
The serious players I know would notice the extra excitement, infer a gaff, and tell their casual player friends that something doesn't seem square.
At the end of each showdown, there's still one winning hand, or maybe two that chop. I guess it's still a random way to trade money across the table - it just isn't poker anymore.
OnceDear: Please accept my sincere apologies for hijacking. Inflict such penalties upon me as due.
link to original post
I agree. But (from what i think i know) when you are generating random numbers, you need to generate them pseudo randomly. The algorithm is the pseudo-random generation method.
What makes it truly random is usually a physical random number generator - mixed with the psudo random number generator.
The physical RNG is truly "random" and therefor can not be predicted - thats the seed which usually fuels the "pseudo" RNG.
When you have a seed big enough, you can usually assume that the pseudo RNG will output the correct amount of outcomes that you need.
Im sure theres some kind of scaling going on in there as well but that is something i know little about because i suck at math or truly understanding this entire concept we all know as "Random number generation" as defined by the jurisdictions laws.
And i would say that considering the names of things only matter in court if you infringe someones IP, but the "way" that a game is played is usually the deciding factor as to if someone gets a patent or rights to use the "method". Games are stolen all the time and named different things. Its the people who play the game who decide in their minds if they are playing a specific variant of a game. Which is why i think i see so many people who dont IMMEDIATELY understand 21+3 bets. They sometimes joke like "am i playing poker or blackjack" kind of thing.
Quote: DRichIf the game is being dealt out of Atlantic City and is regulated by New Jersey I would say it is much less likely to be non-random distribution of cards. My game was not allowed other than in Native American casinos even though it was approved by GLI.
link to original post
Its very possible that its being offered in new jersey, but Atlantic City Blackjack is the name of the game i think its produced by IGT
While many people maxed out the coins at $10 a spin, we never had a Royal Flush at anything more than five credits. While the machine accepted bets from one coin to forty, anything more than five coins and you could forget about getting a
Royal. I'm sure it was just a coincidence.
Nooooo. It is an RNG game operated by MicrogamingQuote: DRichIf the game is being dealt out of Atlantic City and is regulated by New Jersey I would say it is much less likely to be non-random distribution of cards. My game was not allowed other than in Native American casinos even though it was approved by GLI.
link to original post
It's 8 decks. regulated either by Alderney Gaming commission, Or UK gaming commission. Unsure as yet, but neither dealt by or regulated by any US organisation.
To be clear, I've played thousands of hours of this and I DO NOT think it is cheating me at all. There is nothing conspicuously suspicious about this game except that I seem unduly lucky at low stakes. As though it is cheating itself and sneaking me a few good cards at those low stakes. Legally, I bet they could do that and get away with it, since no-one complains about winning too much.
I'm just seeking advice on how to test this hypothesis without making many massive wagers.
I'm in England, playing a UK regulated mainstream casino.Quote: heatmapQuote: DRichIf the game is being dealt out of Atlantic City and is regulated by New Jersey I would say it is much less likely to be non-random distribution of cards. My game was not allowed other than in Native American casinos even though it was approved by GLI.
link to original post
Its very possible that its being offered in new jersey, but Atlantic City Blackjack is the name of the game i think its produced by IGT
link to original post
This one is by Microgaming, Regulated operator in Europe.
I trust the products that are hawked that come into my spam email folder more than I do online casinos
.
Quote: OnceDearThere is nothing conspicuously suspicious about this game except that I seem unduly lucky at low stakes.
link to original post
One might suggest you play at low stakes indefinitely and continue to harvest the positive variance.
That was my thinking too.Quote: DieterQuote: OnceDearThere is nothing conspicuously suspicious about this game except that I seem unduly lucky at low stakes.
link to original post
One might suggest you play at low stakes indefinitely and continue to harvest the positive variance.
link to original post
I can chip away with a 1,2,4,5,5,5 progressive day in day out and either win or lose a few hundred.If I increase my initial Buy in tenfold (again) then I could try similar at double those stakes then triple. That would give some sort of indication of whether there is a generosity threshold. Conclusive proof would be harder, but who cares if I can win?
Quote: OnceDearThat was my thinking too.Quote: DieterQuote: OnceDearThere is nothing conspicuously suspicious about this game except that I seem unduly lucky at low stakes.
link to original post
One might suggest you play at low stakes indefinitely and continue to harvest the positive variance.
link to original post
I can chip away with a 1,2,4,5,5,5 progressive day in day out and either win or lose a few hundred.If I increase my initial Buy in tenfold (again) then I could try similar at double those stakes then triple. That would give some sort of indication of whether there is a generosity threshold. Conclusive proof would be harder, but who cares if I can win?
link to original post
Indeed. It may be fascinating to figure out how the golden goose works, but collecting the eggs may be more valuable.
Quote: teliotFind out if the software was written by a company based out of Costa Rica. I know exactly one fair company from CR, every other company is crooked. At least, that was the state of the art in 2013 when I sold my auditing company Certified Fair Gambling, the one Michael sold to me in 2006 that is now owned by Charles Mousseau.
link to original post
Most of the games in domestic regulated markets are made by IGT, Scientific Gaming, Netent and other generally respected company’s. They are all different than the ones offered at the offshore places such as Betsoft.
I have previously written about a game that I found in NJ. While I couldn’t figure out the exact house edge, I believed it was at worst case a break even game. I played tens of thousands of hands and millions of dollars through it. Not including bonuses, rough estimate is that I lost about 1% on it. Every game you play does seem to have patterns that you pick up on. Although I believe these to be likely just selective memory, or whatever the term is. For instance it’s easy to say it seems like every time I have a 20 and the dealer has a low card somehow they always end up with a 21. Not a 20, somehow always a 21. But in reality the dealer is going to get 21 every so often. You just remember it more cause you had an assumed winner.
With that game there was one thing I noticed. When I had a two card 12 v a two card 11, I seemed to get a 10 and bust a lot more than I made a 21. So one day I counted it. I counted 50-100 times of each happening and whether I got a 10. I was wrong in my assumption about the 11. It turns out I got a ten about 30% of the time, right at expectation. But sure enough when I had a 12 I got a 24/60 times. Sure it’s a small sample size. So the next day I played again. Sure enough when I had a 12 I was getting a 10 about 40% of the time.
I seem to notice patterns like this in all of the games. I don’t know if they’re random. The other day a baccarat game went 17 hands against me in a row. Sure it can happen. You can pass that off as a one in a hundred thousand or million or whatever but besides that one stretch of bad hands it also seems to go 5 against me, then 8, then 10, all in a period of 500 or 1000 hands. Sure I can write off the one bad streak but when 300, 500 and 1000 to 1 streaks are happening multiple times you got to wonder.
The bottom line is who knows if any specific game is random. So if I don’t like a particular game I just say to myself you can keep playing it or stop. Nobody is forcing you to keep playing. As much as I think it’s rigged sometimes I am still up significantly over the past few years.
The hypothesis I'm testing is a bit different in that I suspect that it's helping me to win at trivial stakes, knowing that as I get cocky I'll ramp up. As a marketing ploy that would be a no brainer. It doesn't even need to ever pay short of the stated rtp. The pittance it gives me as a low stakes winner could easily be clawed back by the legit house edge as I wager bigger.Quote: Sandybestdog
The bottom line is who knows if any specific game is random. So if I don’t like a particular game I just say to myself you can keep playing it or stop. Nobody is forcing you to keep playing. As much as I think it’s rigged sometimes I am still up significantly over the past few years.
link to original post
OK. I'm going to tentatively take this approach. It will be a slow test andwill evolve.Quote: Dieter
Indeed. It may be fascinating to figure out how the golden goose works, but collecting the eggs may be more valuable.
link to original post
I've decided to credit £100 to my account and ONLY play this one game (except I'll cash out and deal with any freeplay outside of this test)
I'll strictly play 1,2,3,4,4,4,4 neg progression. Never taking wager above 4. I'll handle splits and doubles normally, so occasionally I'll have slightly more in play. Any hand with a net loss, even a surrender, I'll take a step on the progressive. Any win, I'll reset to 1 and any net push, I'll replay same stake.
If I get my 100 to 200, I'll continue at the same pace. If I get to 500, I'll start recording the action.
So far, in session 1 after a few hours !!! I've advanced to 136 then fallen back to 111. Not been below 98 yet.
Onwards and upwards.
So I have had a couple instances where I felt something wasn't right with brick and mortar casinos and I know how to test to confirm or deny my suspicions.
One instance was a situation that I believed and still do that cards were being clumped by shuffle machines which resulted in nearly every shoe or times through the cards (it was a double deck game) reaching a true count of either plus 4 or more or -4 or less, depending when the clump came out. There just were no shoes or times thru the cards that the count stayed mostly neutral or , +/-1, +/-2. This was unusual in that most times through the shoe in normal play, the count does stay somewhat neutral, or moves only slightly, so to see nearly every shoe hitting + or - 4 was weird. So I tracked over 400 shoes or times through the cards and to my astonishment 97% of the time the count reached + or - 4. There was no possibility this was not some kind of clumping.
The second situation was more difficult. I didn't know exactly what was wrong, but it just felt like my results were abnormally bad. I first counted my blackjacks received to be sure I was receiving one every 21 hands or so (or close to it) and then counted down 100 shoes to be sure every shoe didn't end with a significantly plus count which would indicate some missing high value cards. All my checks, checked out. I couldn't find anything wrong.
But with an online casinos, I wouldn't begin to know how or what to check. I would be afraid there was something in the software that resulted in all 11 double downs receiving a 4 or 5 or something like that. And if not all, maybe just a higher number than normal. That is all it would take to increase the house edge significantly. And I wouldn't know how to begin to check for something like that other than just tedious counting them up.
I need to rework this experiment. Session #1 playing 1,2,3,4,4,4 neg progressive was a fail. This is what happened...Quote: OnceDearOK. I'm going to tentatively take this approach. It will be a slow test andwill evolve.Quote: Dieter
Indeed. It may be fascinating to figure out how the golden goose works, but collecting the eggs may be more valuable.
link to original post
I've decided to credit £100 to my account and ONLY play this one game (except I'll cash out and deal with any freeplay outside of this test)
I'll strictly play 1,2,3,4,4,4,4 neg progression. Never taking wager above 4. I'll handle splits and doubles normally, so occasionally I'll have slightly more in play. Any hand with a net loss, even a surrender, I'll take a step on the progressive. Any win, I'll reset to 1 and any net push, I'll replay same stake.
If I get my 100 to 200, I'll continue at the same pace. If I get to 500, I'll start recording the action.
So far, in session 1 after a few hours !!! I've advanced to 136 then fallen back to 111. Not been below 98 yet.
Onwards and upwards.
link to original post
From £100 to £136 was a slow but smooth ascent over 30 minutes or so. Then a slow descent through 111 to 100. Then about four more hours of play !!! I bobbed about eventually reaching £48. This game was doing me no favours and I was BORED.
So I staked £24 and lost. staked £24 and won a few times. Then a few wagers in the region £10 to £30, I clambered to £101 and cashed out.
Playing stakes of 1,2,3,4 had done alright for 30 minutes or so, but turned into a slow grind down pretty much as one would expect min betting into a -EV game. That I clawed my loss back with a few risky big bets was incidental.
Apart from the early 36% gain, there was no anomaly. No golden goose. Just a slow grind that took me down over many hours.
I might employ a bot if I revisit this experiment.
Quote: OnceDearI need to rework this experiment. Session #1 playing 1,2,3,4,4,4 neg progressive was a fail. This is what happened...
link to original post
Lie to me. Call me names. Tell me you got it on video.
The session exceeded 4 hours. To much video to observe and transcribe for analysis.Quote: DieterQuote: OnceDearI need to rework this experiment. Session #1 playing 1,2,3,4,4,4 neg progressive was a fail. This is what happened...
link to original post
Lie to me. Call me names. Tell me you got it on video.
link to original post
Apart from feeling that I got a few extra blackjacks and a few early lucky breaks, it played out as a -ev game should play out: Mostly random walk trending downwards. I hoped and expected, and failed to slowly double my 100.
And if anyone says I should have quit while ahead or hit and run, they will be taken out and shot !
Funny that after abandoning my experiment, I doubled what meagre bankroll I had left. And later, doubled it again.
Some in this thread seem to expect me to reveal conspicuous or repeatable anomalies. I don't expect that at all, just MAYBE a favourable game at low stakes.
Quote: OnceDearI need to rework this experiment. Session #1 playing 1,2,3,4,4,4 neg progressive was a fail. This is what happened...
link to original post
The session exceeded 4 hours. To much video to observe and transcribe for analysis.
Apart from feeling that I got a few extra blackjacks and a few early lucky breaks, it played out as a -ev game should play out: Mostly random walk trending downwards. I hoped and expected, and failed to slowly double my 100.
I don't expect that at all, just MAYBE a favourable game at low stakes.
link to original post
To be clear, I still harbour the suspicion that this game has what I'll call 'Generous Mode' with reduced or even negative house edge at low stakes. I've only anecdotal evidence and gut feel after many hundreds of hours playing it and not particularly losing.
Proving it is too much to ask. But if I can USE it to my advantage while experimenting, then I see no harm in trying.
For now, I'm going on the hypothesis that 'Generous mode' is active for stakes below £5
Session #1, I attempted to take £100 to £200 using 1,2,3,4,4,4 neg progressive. I abandoned the playing pattern of this session at £48 and managed to recover my bankroll with a few big, lucky bets.
Session #1 was just hours too long to bear. The amount of action exposed and re-exposed to the house edge made failure likely.
Session #2, I flat bet £4. Session length was more acceptable and comprised about 200 hands. £200 target achieved!
I'll try at least one more session flat betting £4. If I can grow my initial £100 to £500, I'll consider that to be weak evidence unless I do so on 5 consecutive sessions, in which case I'll be emboldened to maybe flat bet more.
Might take this topic to my blog.
Quote: OnceDear
To be clear, I still harbour the suspicion that this game has what I'll call 'Generous Mode' with reduced or even negative house edge at low stakes. I've only anecdotal evidence and gut feel after many hundreds of hours playing it and not particularly losing.
...
Session #1, I attempted to take £100 to £200 using 1,2,3,4,4,4 neg progressive. I abandoned the playing pattern of this session at £48 and managed to recover my bankroll with a few big, lucky bets.
Session #1 was just hours too long to bear. The amount of action exposed and re-exposed to the house edge made failure likely.
Session #2, I flat bet £4. Session length was more acceptable and comprised about 200 hands. £200 target achieved!
I'll try at least one more session flat betting £4.
link to original post
Session #3 was a bust. Flat betting £4. Steady downtrend till I'd lost back the hundred won in Session #2
Oh hum. a -EV game playing and paying just like a -EV game!
I don't believe you would indefinitely get away with flat betting or a continuous negative progression without it eventually taking that sessions bankroll..
And therein lies the answer to a test - simply go in with 100 - and continuously play the way you do.
Your bankroll will eventually deplete.
You observe this happening at higher stakes, because the number of stakes in your bankroll is a multiple smaller than when you are down at lower stakes.
Or so the fair dinkum gaming world would believe... ;)
Quote: TwelveOr21What you are describing OnceDear is the slot machine effect.
I don't believe you would indefinitely get away with flat betting or a continuous negative progression without it eventually taking that sessions bankroll..
And therein lies the answer to a test - simply go in with 100 - and continuously play the way you do.
Your bankroll will eventually deplete.
You observe this happening at higher stakes, because the number of stakes in your bankroll is a multiple smaller than when you are down at lower stakes.
Or so the fair dinkum gaming world would believe... ;)
link to original post
You may be right.
Anyway, two scenarios encountered.....
1) Buy in small (say 50) double it slightly more often than I lose it when playing small progressive. Cash out and re buy in each time. I've done this a few times and seen some lucky bundles of sessions.
2) Buy in 100 once, double it with small progressive fairly quickly.
3) Buy in 100 flat bet £4 for many hours. lose steadily as one might expect.
Repeat #2 but at the end, don't cash out, but continue. Session became a losing grind.
So. Anecdotally, I only seem to encounter 'generous mode' when I cash out and re-buy in.
But this is SO subject to my own acknowledged cognitive bias that I'm deferring further experiment. Besides, it has cost me money and not provided supporting evidence of my hypothesis.
Quote: OnceDear
So. Anecdotally, I only seem to encounter 'generous mode' when I cash out and re-buy in.
But this is SO subject to my own acknowledged cognitive bias that I'm deferring further experiment. Besides, it has cost me money and not provided supporting evidence of my hypothesis.
link to original post
sadly this is a quote that i equate to what is possibly happening here. i dont think ALL of you are out to spread misinforation or even steer anyone here in the wrong directions on purpose, but the work to find out if something is what you think it is is just too much. whatever way they decide to randomize something is up to them, and if you see patterns possibly based on how they are generated, it doesnt matter whatsoever about seeing patterns. As long as you dont know the algorithm and cant predict the next iteration then its "fair".
As long as it is "random" over time, the variance that comes with the randomization is considered fair enough to justify whatever is actually happening, "fair" (to you in your own definition) or not
Quote: TwelveOr21What you are describing OnceDear is the slot machine effect.
...
Or so the fair dinkum gaming world would believe... ;)
link to original post
Quote: OnceDearYou may be right.
...
So. Anecdotally, I only seem to encounter 'generous mode' when I cash out and re-buy in.
But this is SO subject to my own acknowledged cognitive bias that I'm deferring further experiment. Besides, it has cost me money and not provided supporting evidence of my hypothesis.
link to original post
Quote: Heatmap...i dont think ALL of you are out to spread misinforation or even steer anyone here in the wrong directions on purpose, but the work to find out if something is what you think it is is just too much.
I'm almost insulted by that. I don't know who 'You' refers to, but I can assure you that 'I' am not here to spread misinformation, nor to steer anyone wrong! Nor am I American, corrupt or otherwise.
I strongly disagree with you there. As I understand it, a table game that is replicated in an RNG game should behave faithfully to a real random shuffled shoe, or real spun wheel. I do not believe US legislation would permit a game to apply a deliberately non-random distribution of outcomes, just because the total of a lot of outcomes would be right. E.g. if it was simulating roulette, it would not be legal to have the 'random results' that constantly toggled red/black and threw in a green exactly every 37th spin. Nor would it be legal to introduce some sort of catch up mode such that over x thousand spins there are exactly equal numbers of red/black or odd/even. The RNG job is to generate random numbers as faithfully as possible with all the expected trends / hot cold times / variance in line with truly random distributions.Quote:As long as it is "random" over time, the variance that comes with the randomization is considered fair enough to justify whatever is actually happening, "fair" (to you in your own definition) or not
link to original post
There have been, and maybe still are, some RNGs which are rubbish and might fold some absurd patterns into their output. But modern RNG's in popular commercial games have a much better behaviour. The RNG job is to faithfully make a random shuffle. The game play algorithm from there is to pull cards from that shuffled deck to the same algorithm as the real live game. Not selected who wins when.
The suspicion that I aspired to test was that my game MIGHT be 'doing me favours' on a few early, low value spins, but then to switch to a honest, faithful -EV game after a while, such as to lull me into thinking I was onto a winner. Such behaviour would NOT be faithful to a really randomised shuffle, so I expect it would be illegal, even though it would be cheating in my favour! It would NOT be considered fair enough or justified at all.
For now, there is no reason to test if higher wagers encounter higher house edge. First, you must test and prove that lower wagers are rigged to give a better experience when playing low stakes. There is a formula for that using standard deviations and confidence levels. You need to quit being lazy buckle down and start keeping track. No more of this... " I SEEM, Probably, half-assed experiment, suspect.Quote: OnceDearHi All,
I have a quandary and seek assistance.
My favourite ONLINE game is an Atlantic City RNG Blackjack. the house edge is a mere 0.35% and I've had many successful sessions on that game. Probably ahead overall.
But I cannot get it out of my head that it is rigged to give a better experience at low stakes.
E.g. Typically if I buy in for 25 with an exit goal of 50, I SEEM to win more often that I lose. If I buy in at 50 with a win goal of doubling my buy in, again I tend to win more than lose. These are at low stakes, typically 1,2,5,10 maximum.
But on occasions like tonight, where I buy in for much more, in this case 3 x 250, and wager bigger, typically 25,50,75, I get my ass handed to me on a plate.
It sure feels that they adjust the house edge to encourage low stakes gamblers to ramp up.
I'm not the only one to suspect this and I've done half assed experiments to support the theory. Hence my assertion that I've done well a few times at stakes <=5
Would anyone care to offer a statistically sound test of the hypothesis that higher wagers encounter higher house edge? Preferably a test that doesn't involve losing my shirt?
link to original post
Are you suspecting it was rigged in your favor to the tune of 5k? That sounds like a significant amount of money not to take a second look.Quote: heatmapI think it is
I once ran my bankroll up to 5k playing 5-25 bets
As soon as I upped to 100 I lost almost every single hand and the entire 5k within an hour
I can lie though running it up to 5k in less than a day was odd IMO playing the smaller stakes
Butt what the hell do I know I have no proof
link to original post
I myself have experienced the same thing as OD, unfortunately, it was only on $1 and $2 bet levels and there was a winning cap of a few hundred bucks. Not worth the time and effort one would have to do in order to make any serious money.
I'm thinking that to test an online RNG game for fairness you would have maybe two options. I know you mentioned BJ ,.but in this explanation I would like to use Baccarat for example.
First and possibly the most difficult proof would be to obtain and examine the program through whatever means are available if the casino is reputable and regulated and has nothing to hide.
Have it certified by an independent agency but using your own "winning playway" for both high bet sessions and low bet sessions. Compare the outcomes.
In this case , suspecting foul play and possible cover up, you would first have to be concerned that the program you obtained and are testing is the one that is actually being used online and not being switched.
So this program could in fact be randomly shuffled using a fisher yates and mersenne twister ( as Bovada uses) but have one or more sub programs to either 1.) capture larger bets or 2.) generously reward smaller bets or both, independently.
The second and easiest way I think to prove / test would be to use your proven consistent "winning playway" (that many " losing players" agree doesn't exist ) that works for you long term, and then compare the results to the RNG game in question.
I would it test by playing your " winning playway" through 3 SD's as follows:
Four Senarios
1.) Self deal and record results ( bankroll + or -)
2.) Play Live dealer online and record results BR + or -
3.) Follow the Wizard's suggestion and test it against his 250,000 Baccarat shoes. record results BR+or-
4.) Play RNG game in question record results BR+or-
The results will be No. 1, 2 , and 3, will show a positive WE ( winning expectation) and no. 4 will be negative WE.
There's your proof.
As an AP the goal is to make money gambling, so why care if it's an honest game if you can make money outsmarting their system?Quote: kewljThis is an interesting topic, Oncedear and the EXACT reason I don't play online casinos. I have just never been comfortable that I can be assured I am getting an honest game.
Here is an example: Let's say the online casino has a blackjack game that should have a house advantage of 1%, but they have rigged it to hold 10%, yet they are offering a bonus/promo/deal that gives you a 25% advantage. That seems like a no-brainer to me.
Quite right. So far, my 'sperimenting has been weak and badly documented. I need to screen capture the entire sessions and log the games to a database or spreadsheet, then analyse the hell out of it.Quote: AxelWolfFor now, there is no reason to test if higher wagers encounter higher house edge. First, you must test and prove that lower wagers are rigged to give a better experience when playing low stakes. There is a formula for that using standard deviations and confidence levels. You need to quit being lazy buckle down and start keeping track. No more of this... " I SEEM, Probably, half-assed experiment, suspect.Quote: OnceDear
But I cannot get it out of my head that it is rigged to give a better experience at low stakes.
link to original post
link to original post
I think my time would be better spent getting a bot to record it for me.
Quote: OnceDearQuote: TwelveOr21What you are describing OnceDear is the slot machine effect.
...
Or so the fair dinkum gaming world would believe... ;)
link to original postQuote: OnceDearYou may be right.
...
So. Anecdotally, I only seem to encounter 'generous mode' when I cash out and re-buy in.
But this is SO subject to my own acknowledged cognitive bias that I'm deferring further experiment. Besides, it has cost me money and not provided supporting evidence of my hypothesis.
link to original postQuote: Heatmap...i dont think ALL of you are out to spread misinforation or even steer anyone here in the wrong directions on purpose, but the work to find out if something is what you think it is is just too much.
I'm almost insulted by that. I don't know who 'You' refers to, but I can assure you that 'I' am not here to spread misinformation, nor to steer anyone wrong! Nor am I American, corrupt or otherwise.I strongly disagree with you there. As I understand it, a table game that is replicated in an RNG game should behave faithfully to a real random shuffled shoe, or real spun wheel. I do not believe US legislation would permit a game to apply a deliberately non-random distribution of outcomes, just because the total of a lot of outcomes would be right. E.g. if it was simulating roulette, it would not be legal to have the 'random results' that constantly toggled red/black and threw in a green exactly every 37th spin. Nor would it be legal to introduce some sort of catch up mode such that over x thousand spins there are exactly equal numbers of red/black or odd/even. The RNG job is to generate random numbers as faithfully as possible with all the expected trends / hot cold times / variance in line with truly random distributions.Quote:As long as it is "random" over time, the variance that comes with the randomization is considered fair enough to justify whatever is actually happening, "fair" (to you in your own definition) or not
link to original post
There have been, and maybe still are, some RNGs which are rubbish and might fold some absurd patterns into their output. But modern RNG's in popular commercial games have a much better behaviour. The RNG job is to faithfully make a random shuffle. The game play algorithm from there is to pull cards from that shuffled deck to the same algorithm as the real live game. Not selected who wins when.
The suspicion that I aspired to test was that my game MIGHT be 'doing me favours' on a few early, low value spins, but then to switch to a honest, faithful -EV game after a while, such as to lull me into thinking I was onto a winner. Such behaviour would NOT be faithful to a really randomised shuffle, so I expect it would be illegal, even though it would be cheating in my favour! It would NOT be considered fair enough or justified at all.
link to original post
I do not think you or anyone here is trying to spread misinformation, but what i was saying is technically what you are saying, that you have a huge roadblock ahead of you and its going to require alot of work and that it most likely will not be able to show you anything significant.
there has only been a few times where i can say someone has tried to prove that certain places like this were "rigged" and they were met with doubt because they didnt have the entire set of data. You need to be the casino, and be in their database, in order to see every single hand dealt in order for it to be tested as random.
when a computer shuffles cards they do not follow the same physical procedure as if it were a dealer. this is why i am always harping on this stuff. Most people actually believe that the cards are shuffled the same way as if they were to sit down at a table game and watch a dealer shuffle the cards when they are playing these digital blackjack or poker or whatever game you are playing. With computers anything is possible and if the laws of the jurisdictions dont specifically tell the manufacturers "how" to do shuffle the cards, then that is up to them.
also what i am speaking about when it comes to the algorithm is not programmed to do things at specific times, it happens EVERY time. The players who win are picked at random, and the cards are still picked at random. Its just that the entire hand is already made before.
I really wish drich would speak up about the algorithm because im kind of talking out of my ass here i have skimmed over the patent and he can explain how exactly they justify doing what they do.
Quote: AxelWolfAre you suspecting it was rigged in your favor to the tune of 5k? That sounds like a significant amount of money not to take a second look.Quote: heatmapI think it is
I once ran my bankroll up to 5k playing 5-25 bets
As soon as I upped to 100 I lost almost every single hand and the entire 5k within an hour
I can lie though running it up to 5k in less than a day was odd IMO playing the smaller stakes
Butt what the hell do I know I have no proof
link to original post
I myself have experienced the same thing as OD, unfortunately, it was only on $1 and $2 bet levels and there was a winning cap of a few hundred bucks. Not worth the time and effort one would have to do in order to make any serious money.
link to original post
i do think that if its rigged to make you lose its rigged to make you win
im going to see if they still have the history of the gameplay (msot likely not) , but since im doubling down here with the rigged thing, if you could have seen the type of stuff that i was doing when i did run it up, you would face palm yourself and walk away. I was doing some very very bad stuff, whereas BS was rarely used.
Ran it up from what? $50 to $5k would be more impressive than $4k to $5kQuote: heatmapI think it is
I once ran my bankroll up to 5k playing 5-25 bets
As soon as I upped to 100 I lost almost every single hand and the entire 5k within an hour
I can lie though running it up to 5k in less than a day was odd IMO playing the smaller stakes
Butt what the hell do I know I have no proof
link to original post
What game, where? What stakes?
Quote: OnceDearRan it up from what? $50 to $5k would be more impressive than $4k to $5kQuote: heatmapI think it is
I once ran my bankroll up to 5k playing 5-25 bets
As soon as I upped to 100 I lost almost every single hand and the entire 5k within an hour
I can lie though running it up to 5k in less than a day was odd IMO playing the smaller stakes
Butt what the hell do I know I have no proof
link to original post
What game, where? What stakes?
link to original post
It was the game they are speaking about - at least I thought it was. Pa pokerstars casino, Atlantic City blackjack.
Something is coming back to me about this particular session. I’m not sure if this matters in online blackjack where they (hopefully) are using a shuffle every round in order to make it CSM like, but I had a way to play “heads up” with a 4 deck game.
And I am nearly certain that I ran it up from less than 50 because at some points in the session I was all in and kept coming back it was a really bumpy road
I should also mention the session lasted 8 or more hours over multiple (I think 3 days and may have been longer than 8 hours)
Quote: heatmapwhen a computer shuffles cards they do not follow the same physical procedure as if it were a dealer. this is why i am always harping on this stuff. Most people actually believe that the cards are shuffled the same way as if they were to sit down at a table game and watch a dealer shuffle the cards when they are playing these digital blackjack or poker or whatever game you are playing. With computers anything is possible and if the laws of the jurisdictions dont specifically tell the manufacturers "how" to do shuffle the cards, then that is up to them.
link to original post
A common computerized shuffle is random.
Beyond the input deck being reordered to become an output deck, I think it's safe to say there is little similarity to most manners of hand shuffling.
Most hand shuffles are based on a series of cuts, riffles, and running strip cuts.
Computerized shuffles rarely do that.
https://www.pokerstrategy.com/forum/thread.php?threadid=510544
Quote:
To convert random bit stream to random numbers within a required range without bias, we use a simple and reliable algorithm. For example, if we need a random number in the range 0-25:
- we take 5 random bits and convert them to a random number 0-31
- if this number is greater than 25 we just discard all 5 bits and repeat the process
Finally, we use that method to do the actual shuffle:
To perform an actual shuffle, we use another simple and reliable algorithm:
- first we draw a random card from the original deck (1 of 52) and place it in a new deck - now original deck contains 51 cards and the new deck contains 1 card
- then we draw another random card from the original deck (1 of 51) and place it on top of the new deck - now original deck contains 50 cards and the new deck contains 2 cards
- we repeat the process until all cards have moved from the original deck to the new deck
So, how does it work? First, we need a number from 0 to 51 to get one of 52 available cards. To get such a number, we need 6 bits. We take the first six bits of our much larger stream of random bits, and never use them again:
010101111001011001110110100010001010111101010101011010101010101011...
010101
(use these)
----------111001011001110110100010001010111101010101011010101010101011...
(these are what's left)
If that number is from 52 to 63, we discard it as too large. If it is between 0 and 51, we use it to choose the card. In this case, 010101 is our six bit number, and it is "21", so we choose card 21 as the first card.
We continue down the bitstream as needed. We now need 0 to 50 (51 cards left), and the next six bits are 111001, which is 57:
------111001011001110110100010001010111101010101011010101010101011...
------111001
(use these)
----------------011001110110100010001010111101010101011010101010101011...
(these are what's left)
We discard that as too large and continue with the next six bits, 011001, or 25, and so on.
Each time the number of cards is reduced, the number of bits we need can drop, too. Here's a table showing how many bits of data we need to choose from N remaining cards:
52 = 6 bits needed 35 = 6 bits needed 18 = 5 bits needed
51 = 6 bits needed 34 = 6 bits needed 17 = 5 bits needed
50 = 6 bits needed 33 = 6 bits needed 16 = 4 bits needed
49 = 6 bits needed 32 = 5 bits needed 15 = 4 bits needed
48 = 6 bits needed 31 = 5 bits needed 14 = 4 bits needed
47 = 6 bits needed 30 = 5 bits needed 13 = 4 bits needed
46 = 6 bits needed 29 = 5 bits needed 12 = 4 bits needed
45 = 6 bits needed 28 = 5 bits needed 11 = 4 bits needed
44 = 6 bits needed 27 = 5 bits needed 10 = 4 bits needed
43 = 6 bits needed 26 = 5 bits needed 9 = 4 bits needed
42 = 6 bits needed 25 = 5 bits needed 8 = 3 bits needed
41 = 6 bits needed 24 = 5 bits needed 7 = 3 bits needed
40 = 6 bits needed 23 = 5 bits needed 6 = 3 bits needed
39 = 6 bits needed 22 = 5 bits needed 5 = 3 bits needed
38 = 6 bits needed 21 = 5 bits needed 4 = 2 bits needed
37 = 6 bits needed 20 = 5 bits needed 3 = 2 bits needed
36 = 6 bits needed 19 = 5 bits needed 2 = 1 bit needed
1 = 0 bits needed
If you add up all the bits you get 249, which is the number of bits we take from each of our truly random entropy sources.
Since we start with double the number of truly random bits needed (249 each from quantum entropy and user inputs), this is enough to ensure that even if we have to discard every other group of bits as "bigger than the maximum number we need", we have enough truly random bits to complete the shuffle.
link to original postPerfectly reasonable shuffle process. They seem to take a tiny risk of discarding too many bits and running out. Not 'enough to ensure'. Probably have a retry if that happens.Quote: heatmapGo to this page and apparently someone was open enough to tell someone the "algorithm" for pokerstars
So long as the shuffle is as honest as this AND they always deal from the top without a cheat algorithm, then there's no problem.
So, after getting a bit burned, I deferred the experiment. But in a few recent sessions just playing how I normally would, I've had good luck. I've been doing more of a 5,10,15, 20, 20, 20 progressive, so outside the low stakes I suspected of 'generous mode' Typical recent session buy-in 100, lose down to 50. win up to 800 and cash out.Quote: OnceDearBut this is SO subject to my own acknowledged cognitive bias that I'm deferring further experiment. Besides, it has cost me money and not provided supporting evidence of my hypothesis.
link to original post
Have put buy in limits on my account to slow me down.
I will revisit this experiment. First, I need to sort out efficient game capture.
