randomness...online vs live

[Prowler]

it occurred to me late last night that online play is more random than live play to a certain degree (i was a bit tired and had a few drinks)

in a live game once the cards are shuffled and dealt, the flop, turn and river are already predetermined, there is no way of changing them without cheating in some manner

online the flop, turn and river are not predertermined until the last player has clicked the button and then a card is randomly generated

now i'm not sure how much difference this makes and someone who can work out stats or %'s might need to help me with this (dice? phil?)

the chance of hitting say a flush in both games has the same % but in a live game the deck is already determined, would it be more likely to get a heart say if you took a card from the top of the deck after 3 had already come out or if you picked one from anywhere in the deck at random? is it more likely that all the hearts are bunched together in a deck or are randomly split?

statisticly what are the chances of picking 5 random cards from the deck and making 4 parts of a str8 as opposed to making 4 parts using the flop turn river style?

also human shuffling, if someone is not a particularly good shuffler and does the basic shuffle then the cards are just moved around in blocks rather than say with a riffle shuffle they are all seperated 3 to 4 times, say after a big hand with three aces and 3 kings in and a bad shuffle after all those cards are clumped together and either will come out together or not be seen at all.....does this affect your odds or the randomness factor?

[Dice Man]

I have no idea, but if the cards were shuffled for an infinite amount of time then they would be as random as online. Finitely it depends on the style of shuffling being used. I don't think you could make some calculations and determine that more hearts are going to come out on average this time than any other time with a real deck, unless you watched extra carefully how the cards from the last hand were collected with the remaining deck, and then shuffled in, and that's very difficult.
Maybe someone should do an experiment where all the cards are chipped, do loads of different shuffles and you could determine the patterns in which shuffling occurs and how cards move around between each shuffle. Also, what is the chance of two adjoining cards still being adjoined after the shuffle than beforehand? Usually not every card gets shuffled around, so if you knew that pre-shuffle the Qh and 7c were next to each other, then after the shuffle you got dealt the 7c, you might be able to say that there's a 50% chance that the player on your left, who got dealt the adjoining card after the shuffle, has the Qh. It would be more difficult, though possible, to do that to pedict the community cards because of the burned cards in between.
You would have to be a Rain Man type genius to be able to make those kind of calculations though, and I'm sure it would be called cheating were it discovered, and banned just like card-counting.

Also, one could argue that the cards online are pre-determined. It's a bit controversial wherther random number generators are actually random. Some of them are determined to look random, but by doing this they are essentially non-random. I don't no what kind of generators poker websites use, but they have to be detetrmined somehow, even if it's on the sub-atomic or quantum level. Of course, if we could predict that sort of stuff it would be put to much better uses than cheating poker websites, but in principle it's predictable. It would be much easier to predict deck shuffles though, and that would be insanely difficult.

I've seen that in some casinos they've started using machines to shuffle lots of blackjack decks at once, and this is a continuous action while the dealer is not dealing. Those machines must be programmed to operate in a certain fashion. Maybe they have random number generators that determine at which points of the deck it should shuffle. But if they don't then maybe one could predict those as well, though with as many decks as they use, that too would be insanely difficult.

[barbarian]

If the deck is shuffled randomly it won't make any difference where the next card is taken from in the deck. Also even if the shuffle wasn't very good isn't doesn't seem that there would be any reason for cards to be grouped in the deck in any particular way (unless the game is draw poker where people will be grouping cards together through the discard process). Maybe if you remember the community cards from the last hand, and see the shuffle isn't good, then see one of the previous hands community cards dealt to you, you could give a higher probability to certain cards being in your opponents hand.

[Prowler]

so certain random factors could be an influence in a live game? so is online more random then?

personally i think hitting a flush online happens more than in live games and thats taking into consideration the amount of hands played online compared to live

the way the rng works means any card can come at any time provided it hasnt been used already but in a live deck the card is already decided so does that take away some of the randomness?

[barbarian]

I don't think the fact that the card is already being decided affects the randomness. Mathematically it certainly doesn't, i.e. if the deck is shuffled randomly then taking the top card (determined card) or randomly taking a card makes no difference. You could think about it like this: If you play two hands, one where the hole cards are dealt, the community cards are taken from the top of the deck (with the appropriate burns) and placed face down on the table before the hand starts, then turned over at the appropriate time. This is exactly the same as a normal live game, except the cards are on the table at the start (no effect on the randomness). Now think of another game where exactly the same is done except the community cards are taken randomly from the pack and placed on the table at the start of the hand. These cards are now determined before the hand starts, but this doesn't seem to affect the randomness, assuming the deck is shuffled randomly. I think the only difference between randomness online and live could be the shuffling. Unless there are people remembering the cards, and making inferences based on this, i don't think even this will affect the game at all.

[Dice Man]

Tosome extent both live and real deals are determined. Nothing can really be randomly determined. Even the rolling of a die is determined by how you dropped it, and other physical considerations.
In statistics, the probability you can assign to an event depends on the knowledge you have of the event. Online I have no knowledge of what cards will come next, because I don't know how the random number generator works. In a real game, if I chose to investiaget scientifically I might be able to discover knowledge of shuffling techniques that produce certain patterns of dealing. Of course I would need to know where some of the cards were in the deck before the shuffle. But I don't usually know or care about this, meanin that my knowledge of the real deck is identical to my knowledge of the online deck, and hence the probability should be the same.

[barbarian]

To some extent both live and real deals are determined. Nothing can really be randomly determined.

This is an interesting side issue. I think i'm correct in saying the random number generators will produce a number that is 'determined' on the quantum level, i.e. on a purely probabilistic level. This isn't really determined in the usual sense of the word. You couldn't trace a line of causality back and find the specific cause of the specific number that was produced. The 'cause' of the number is, essentially, random.

[Thor2007]

Originally Posted by Prowler

now i'm not sure how much difference this makes and someone who can work out stats or %'s might need to help me with this (dice? phil?)

What you are refering to is basically one branch of mathematics known as combinatorics. About chances of hitting certain cards with no repetition allowed (obviously as we can't have two Aces of spades etc).

There was a formula that allowed you to work this out, and for the life of me I cannot remember it. I will dig it out, I have it somewhere from my uni days...

I doubt there is any real difference in randomness online vs in a standard pack but somehow I always feel safer and more comfortable with the deck in front of you.

[Prowler]

this is a good article on how a rng works although does get a little complex,

http://www.av8n.com/turbid/paper/turbid.htm

the original question or thing that made me think about this is, say post flop you have a hand with 10 outs, in a live deck one of your outs is either there or it isnt, whereas in an online game the rng hasnt been determined yet so at that point you can still hit any one of your 10 outs, the only way a live deck could produce the same results is if it was continually shuffled until the turn card was required

[barbarian]

I guess in a way the probability of the turn and river being what they actually are is 100% and the probability of them being any other card is 0%. Like you say there's only one thing they can be unless you shuffle the deck. But from the point of view of the player, who has no idea what they are(hopefully!) the judges of probability you make will be the same for an rng as a real deck.