Improper shuffling - how to exploit?

[GuitarMarc]

Yes, I have covered this. I think you're getting hung up on the word "random". Call it "proper mixing" or something like that instead. We could sit here all day and argue that any kind of shuffle will produce a deterministic pattern.

A human would not be able to pick up any obvious patterns after 7 perfect rifle shifles from its original state. I also agree that 7 imperfect rifles shuffles is better than 7 perfect ones. My original post wasn't even about this, just mentioned the study in passing.

Can you add anything intelligent to my original question instead of hammering away at an irrelevant side issue. No card room that I know of does 7 shuffles of any kind.

[Jay36489]

You asked this:

[ QUOTE ]
The famous card shuffling study said you need 7 perfect rifle shuffles to properly randomize a deck. Most brick and mortars apparently do not do this.

[/ QUOTE ]

You must not mean perfect shuffles. You need 7 normal shuffles to get a deck close to random. Perfect shuffles are not random at all.

[ QUOTE ]

Should this produce any kind of recognizable pattern from the previous hand over the next few hands? Like the flop may be similar or the hole card combinations could be similar? If so, what to do about it?

[/ QUOTE ]

No you will not be able to recognise a pattern. They do 3 or so shuffles with a box because it is enough to make it so the cards can NOT be predicted. That is the reason the standard deal is 3 w/ box. The fact that they are not truly random is far from them being possible to predict.

[IsaacW]

[ QUOTE ]
He was "anti-intellectual" because he was questioning the merits of a well-known ground breaking study that experts in the field readily accept.

[/ QUOTE ]
Are you serious? I was not questioning the merits of the study. I provided a link to a page that indicates the study did not look at perfect shuffles, as your original post stated; rather, that page indicates that the study looked at imperfect shuffles. As soon as I get back to school I'll be looking up the paper and finding out for sure, but at the moment I have no reason to believe that I was wrong. I stand by my statement that no self-respecting mathematician would call the result of perfect riffling a deck of cards 7 times a random result.

[ QUOTE ]
He was a "know-it-all" because though he didn't really understand the study, he felt he could BS about it.

[/ QUOTE ]
See above.

[IsaacW]

[ QUOTE ]
Yes, I have covered this. I think you're getting hung up on the word "random." Call it "proper mixing" or something like that instead. We could sit here all day and argue that any kind of shuffle will produce a deterministic pattern.

[/ QUOTE ]
This is better. When dealing with mathematics, it is important to be careful about the words you use. Calling a deterministic process random is a pretty bad error.

Since you were talking about perfect riffles, I know that you were considering out-shuffles because you said that 8 repetitions will return the deck to its original state. Therefore, I can predict exactly what the top card will be after your "random 7" shuffle simply by watching which card begins the process at the top of the deck. If you shuffle in this manner and I see you put the A[img]/images/graemlins/spade.gif[/img] on the top of the deck when you start, I know that the player to your immediate left will receive the A[img]/images/graemlins/spade.gif[/img]. If I saw you put the A[img]/images/graemlins/spade.gif[/img] 6 spots in, I could calculate where it would appear in the deck after 7 perfect out-shuffles and I would know who had that card. To say that no player could discern the pattern of the cards after 7 perfect out-shuffles is plainly ridiculous.

[GuitarMarc]

[ QUOTE ]
If I saw you put the A 6 spots in, I could calculate where it would appear in the deck after 7 perfect out-shuffles and I would know who had that card. To say that no player could discern the pattern of the cards after 7 perfect out-shuffles is plainly ridiculous.

[/ QUOTE ]

Well doesn't the cut take care of that? But if you can personally determine something after 7 shuffles, can't you figure something after 3 or 4? That's what I've been asking all along. According to your answer, my original question should be looked into more seriously.

You should start a new post about the merits of 7 imperfect rifles vs 7 perfect rifles. That's never what this post was actually about. Pretend I just said 7 rifle shuffles and not 7 perfect shuffles. I know they're different. It's really irrelevant for what I was asking. It was background information for my main question. Most people didn't seem to be aware of the study so at least I raised awareness of it. The spirit of it was the important thing in prefacing my question.

Please tell me something interesting about the way casinos ACTUALLY do it and what the repercussions of that are. Other people have addressed that and their posts were helpful.

[JoshuaD]

[ QUOTE ]
Can you add anything intelligent to my original question instead of hammering away at an irrelevant side issue. No card room that I know of does 7 shuffles of any kind.

[/ QUOTE ]

No. With the wash, box and shuffle, it is going to be very hard to make predictions with out following a particular card.

[TomBrooks]

I think doing two riffles in a row tends to put the cards back where they were, so a hand mix in between riffles might be the best way to maximize randomness in the least amount of time.

[MtDon]

Improper shuffling can cause the cards to be far from random. But the shuffling needs to be much, much worse than your example.

An example from last night at a 2-6 spread limit dealer's choice game (Holdem and Taho/Split). The dealer was young and very bad shuffler. It was common for noticable "clumps," "packets" of cards an 1/8 of an inch thick to stay together in his shuffling. One Holdem hand came down to heads up. The winner had four 10's. For the next two hands two 10's came on the board. I then asked the dealer to make a few extra shuffles.

There were a few other problems with the dealers last night, such as dealing the turn card before the last player acted. The floorman was called at least two times in about 6 hours.

Dealers like that make you appreciate good dealers.

I guess my point is that even if you can get an advantage on a few hands with a dealer who shuffles bad, you gain much more by having a dealer who does a good job of dealing.


PS: Both the dealers who had problems are relatively new to dealing. And the card room is new and using the dealers it can get.