Only logged-in members can vote!
→ Click here to create a Mob account which gives you access to our forum and all our free bankrolls (no deposit bonuses) → Log in to existing account!
Joined: Feb '08
Location: United Kingdom
Age: 35 (M)
Posts: 1886
I have read here and other places that there is no such thing as random. There are always a series of inputs, which determine the outcome, even where the outcome cannot be predicted or explained. When a dealer shuffles a pack of cards the finishing order will be partly determined by the starting order so I ask the question:
Which is the fairest shuffle?
I realize that there may be no answer to this. After all, how do you determine what is more fair when both methods meet all the criteria? But maybe there is room for debate.
Joined: Apr '09
Location: Australia
Age: 40 (M)
Posts: 6483
Definitely a digital RNG. Followed by a card-shuffling machine, followed by hand shuffle. I wouldn't even call that an opinion, pretty happy to state that as a fact.
------------ Curious to see if any non-rigtards want to debate on this, for now I'll assume this is obvious. But happy to elaborate if anyone disagrees.
------------
Posted by awood88: I have read here and other places that there is no such thing as random.
This is some what debatable and really depends on just how strict your definition of random is.. But certainly digital RNG's seeded by external non-controlled environmental variables are pretty bloody random.
See here for an example - but there's plenty more RNG machines in the world created in a similar way.
Joined: Jan '09
Location: Italy
Age: 53 (M)
Posts: 3871
My choice is for hand shuffle: teoretically a machine can't output something really random ....pratically the inputs of RNG are unpredictable as well as their outputs.
Joined: Apr '09
Location: Australia
Age: 40 (M)
Posts: 6483
Posted by magatt966: My choice is for hand shuffle: teoretically a machine can't output something really random ....pratically the inputs of RNG are unpredictable as well as their outputs.
Seriously? Hand-shuffle as the most "FAIR" ? wow..
Joined: May '11
Location: United Kingdom
Age: 58 (M)
Posts: 1372
Posted by jessthehuman:
Posted by magatt966: My choice is for hand shuffle: teoretically a machine can't output something really random ....pratically the inputs of RNG are unpredictable as well as their outputs.
Seriously? Hand-shuffle as the most "FAIR" ? wow..
It's just natural human bias to trust a person over a machine. It's irrational, but humans are irrational animals.
Joined: May '11
Location: United Kingdom
Age: 58 (M)
Posts: 1372
Posted by sadamman: It depends whos shuffling the cards. An rng doesn't shuffle, it just generates numbers technically. So I'd say a hand shuffle also.
Shuffling is just a synonym for randomizing. How is generating random numbers worse than shuffling?
Edited by zeroster (17 September 2012 @ 15:04 GMT)
Joined: Feb '08
Location: United Kingdom
Age: 35 (M)
Posts: 1886
Posted by jessthehuman:
Posted by magatt966: My choice is for hand shuffle: teoretically a machine can't output something really random ....pratically the inputs of RNG are unpredictable as well as their outputs.
Seriously? Hand-shuffle as the most "FAIR" ? wow..
How are you defining fairness in this equation? For me they are all fair since no player has an advantage. But I can see how you might think fair would be an exactly equal chance of every card being dealt.
So if you have Ad Qd, your opponent has 8d 9d and the flop is Jd 9h 6d, it would be FAIR if you had exactly 47.58% chance of winning, no more no less. An RNG would certainly give this percentage very accurately assuming it was programmed to be as good as random.
It's hard to say whether hand shuffling gives this accuracy as the cards are already in a predetermined order once shuffling ends. The way the cards are collected and moved will determine that final order so there is a greater chance (possibly) of two cards remaining next to each other. But seeing as none of the players know the starting order or would be able to count cards moving in the dealers hand, it's safe to say no one has an advantage regardless of whether you're drawing or in front. And cos cards will not stay together over a series of hands, it can be claimed that the percentage described above is also true of dealer dealt hands.
Another factor I suppose is that with digital data there is always the possibility of corruption. There is the possibility that someone will have inside knowledge. That's not an argument that online poker is more likely than live poker to be rigged but it does raise sort of make it fairer if everyone at the table can see the hand being shuffled. I don't necessarily agree that means by hand is the right answer but that I don't think RNG is a concrete fact that doesn't need explanation.
Joined: May '08
Location: Lithuania
Age: 39 (M)
Posts: 10090
all of the shuffle methods have their own strengths & weaknesses. and it depends what do you mean by saying fairest? if you want to ask which method is generates most random shuffle, then putting aside the manipulation factor (and every method can be manipulated) hand shuffle is most random- since human factor is unpredictable one, rng uses formulas to generate "randomness", shuffle machines do shuffle in a manner manufacturer has set them to do (again simulated randomness)- human meanwhile cannot repeat identical shuffle if he wants to- again i must say that this is only when we are talking about non-rigged human shuffles- we all know there are pro shufflers who don't know what random means
Joined: Jan '09
Location: Italy
Age: 53 (M)
Posts: 3871
Posted by jessthehuman:
Posted by magatt966: My choice is for hand shuffle: teoretically a machine can't output something really random ....pratically the inputs of RNG are unpredictable as well as their outputs.
Seriously? Hand-shuffle as the most "FAIR" ? wow..
assuming fair=random
Yes it is imho.
If my life would depend from the colour of the first card to pick up from a deck: well my favourite option would be to blindfold a man (a 7-yo child would be good as well) and make him suffle the deck for me instead of putting my life into an algorithm simulating randomness (written by a man).
Edited by magatt966 (18 September 2012 @ 06:54 GMT)
Joined: Apr '09
Location: Australia
Age: 40 (M)
Posts: 6483
Posted by magatt966:
Posted by jessthehuman:
Posted by magatt966: My choice is for hand shuffle: teoretically a machine can't output something really random ....pratically the inputs of RNG are unpredictable as well as their outputs.
Seriously? Hand-shuffle as the most "FAIR" ? wow..
assuming fair=random
Yes it is imho.
If my life would depend from the colour of the first card to pick up from a deck: well my favourite option would be to blindfold a man (a 7-yo child would be good as well) and make him suffle the deck for me instead of putting my life into an algorithm simulating randomness (written by a man).
No offence or anything, but I do wonder if that thinking comes from a product of your age perhaps?
I find it very strange that you think an RNG would give you a less-random card (number) than a human shuffling.. Even if we take for granted that there's no deliberate action by the card-shuffler to make it unfair, surely we cannot accept that a 'shuffled' deck is == a 'random card'. When a human shuffles they basically just slot the cards in and out from each other, certain cards are bound to stick together and albeit a very complex pattern, the resulting deck is anything but random. Particularly if it's an ordered deck before the hand-shuffle.
To me as long as no one knows wich cards are about to be dealt it is random. The human factor is more important than the mathematics, because in the end poker is all about the decisions you make.
Of course a well done RNG is theoretically more random than physicly shuffling a deck, but it shouldn't influence a player's decision in any particular spot... So pretty useless question IMO.
Joined: Nov '11
Location: United Kingdom
Age: 40 (M)
Posts: 747
if all method are carried in an honest manner then they are all fair methods of shuffling. none would be fairer than the other, i can shuffle a deck of cards very well indeed the outcome would be just as random if a computer had shuffled the cards. dont have much expeience with card shuffling machines but i coould imagine they are designed to do just as good a job as a propper hand shuffle but saving time and effort so they would be no fairer just an end result of a random deck of cards. the main problem with this question is that 2 methods invovle taking an exsitsing order of cards and rearanging the order to a random specification and the RNG is a different concept completely where it does not rearange the order but completly generates a new random order from scratch and is used in a completely different scenario where the phisical "cards" dont really exist.
Edited by takingdrugs (19 September 2012 @ 00:10 GMT)
Joined: Oct '11
Location: Sweden
Age: 32 (M)
Posts: 1410
I would have to say by hand. Sure, it's not as much random as an RNG but I still would prefer that a human had done the shuffling.
There isn't any risks when it comes to errors which, lets face it, programs or machines eventually will encounter. Sure, a human might not be shuffle the deck to perfection but if you see the dealer doing it, at least you know that the deck is being shuffled. And according to me, it really doesn't matter if it's a human, a mechanic card shuffler or an RNG that does the shuffling. The odds of getting one of your outs is still the same. It's all math, you have a number of cards left in the deck and your opponents cards that you can't see and then you have your outs. The odds of getting one of your outs is still the same no matter who or what is shuffling the deck.
Bottom line is that I'm more willing to trust a person then I am to trust a machine or a program.
Joined: Jan '11
Location: United States
Age: 60 (M)
Posts: 3361
All I can say atwood is it's funny how one of your most basic poles has turned out to be one of the best as far as argumentive / opinionated. It's definitely created a bit of friction for the most part. Personally I trust the RNG over the human, but I won't even pick which one is my preferred. great topic for discussion!
Joined: Oct '11
Location: Sweden
Age: 32 (M)
Posts: 1410
Posted by jessthehuman:
Posted by erru9107:
... I still would prefer that a human had done the shuffling.
There isn't any risks when it comes to errors which, lets face it, programs or machines eventually will encounter
IMHO human error is far more likely than machine error.
Yeah sure, human errors are known to happen every ones in a wile, but if it happens I think that you've got a better chance on catching it before the cards are dealt and also, the dealer may notice something where a program wouldn't.
I think there's some good arguments for both really but as I said in my previous post, I would prefer a human dealer to a program.
Joined: Apr '09
Location: Australia
Age: 40 (M)
Posts: 6483
To illustrate this, I would say to anyone who knows even the most basic of computer programming, write a small console based program, to randomly pick a number from 1-52, then iterate it literally billions of times.. It will in all likelyhood complete those billions of iterations WITHOUT error.. I would love to see a human repeat the feat.. A human couldn't even complete the same amount of shuffles in their lifetime, let alone complete them without ever a single error in randomness. A computer can do it in less than a second.
Computer bit-error rates are extremely low, the amount of times you would have to play out the the iterated sequence (of billions of simulated shuffles) before you even hit a single 1bit error is monstrous. And considering all modern code run-times include error-correct for generally single-bit right through to 3 or bit errors..
------------
Posted by erru9107:
Posted by jessthehuman:
Posted by erru9107:
... I still would prefer that a human had done the shuffling.
There isn't any risks when it comes to errors which, lets face it, programs or machines eventually will encounter
IMHO human error is far more likely than machine error.
Yeah sure, human errors are known to happen every ones in a wile, but if it happens I think that you've got a better chance on catching it before the cards are dealt and also, the dealer may notice something where a program wouldn't.
I think there's some good arguments for both really but as I said in my previous post, I would prefer a human dealer to a program.
This came while I was typing my post, anyway - I strongly disagree. As I mentioned, computer code run-times include error-detection and correction. And have done so for decades now, but modern error correction and detection is extremely good.
Once again, I must say - without question a computer can perform a insanely larger number of shuffles without the likelyhood of error compared to a human.. If a computer simulated some 50 billion shuffles (in almost no time too) it would have a chance of error so low as to be negligible to any results (basically almost guaranteed no error over such a 'small' sample). If a human was to do 50 billion shuffles, you could almost guarantee the opposite - at least one or more of the shuffles wouldn't randomise the cards well enough, or the deck would be short (ever played a game then realised part way through you're playing solitaire to dawn with a deck of 51?).
I find this whole argument to be ridiculous to be quite honest. Maybe it's just because I come from a software engineering background and actually understand what is involved.
Edited by jessthehuman (19 September 2012 @ 00:45 GMT)
