When I play a card based game, I shuffle the cards by dealing them into three piles, which I then shuffle with a regular overhand shuffle, and finally stack back into one big pile.
This technique makes it impossible for cards that are adjacent at the end of one game, to also be adjacent next game. I feel like this is especially important in matching or set collecting games.
Recently, a player has objected to this technique, claiming that card games are designed with traditional shuffling in mind, where it's likely for cards to stay adjacent between games.
He wasn't mad or anything, he just pointed it out and we continued playing without a problem. But I've been wondering about his comment since.
What are your thoughts on this?
 It's All Fucked Shirt $22.14 |
 CRIME Shirt $21.68 |
It's All Fucked Shirt $22.14 |
>He wasn't mad or anything, he just pointed it out and we continued playing without a problem.
I have no opinion, nowadays there's no point taking a stance if it doesn't make someone mad.
He's talking bullshit. I've never heard of a game that suggested a specific number of shuffles or method of shuffling.
just because card games don't mention how to shuffle doesn't mean they don't assume a regular overhand or riffle shuffle is used. 99.99% of players don't use pharo shuffles or techniques like OPs that separate every card from each other, so why bring it up? Additionally, in some card games, like klaverjas, cards are dealt two or three at a time, which does seem like it's trying to take advantage of near cards staying together in a normal shuffle
>overhand or riffle shuffle
Those are entirely different and have hugely different effects on the resulting distribution. Not caring which of those is used means you don't care what the frick someone does.
Also analysing this sort of thing is really quite hard, and I don't for a moment believe that any remotely traditional games were designed with it in mind. They simply make the magical assumption that "shuffling" completely randomises the cards.
>traditional shuffling in mind, where it's likely for cards to stay adjacent between games.
This is not and has never been a thing and he convinced himself in the moment that it was so he could fuss over nothing.
Supposedly it's been mathematically proven that it takes 7 random riffle shuffles to properly randomize a deck of playing cards.
>This technique makes it impossible for cards that are adjacent at the end of one game, to also be adjacent next game.
So it's intentionally not fully random. And in such a way that if you happened to have a pile of cards at the end of the game that you wanted evenly distributed throughout your deck (as opposed to fully random) this would give you an advantage. Like lands in MTG.
In short it's mana weaving. A bit obscured, but the three decks tat don't get shuffled with each other also ensures that you won't accidentally shuffle properly and fully randomize away the weaving.
>if you happened to have a pile of cards at the end of the game that you wanted evenly distributed throughout your deck (as opposed to fully random) this would give you an advantage.
That's true. I didn't think about a more even distribution being a possible competitive advantage in some games.
>One practical problem you'd have to solve is that the line between proper play and cheating can get blurred.
Hm, but this is even a problem with a traditional overhand shuffle, right? Shuffle an uneven amount of times, and the bottom cards will end up near the top an vice versa. Shuffle an even amount of times and the top and bottom cards with stay roughly near the top and bottom.
I think it's clear enough in my technique that I'm not attempting a trick, nobody has implied I was doing it to cheat before.
>in some card games, like klaverjas, cards are dealt two or three at a time, which does seem like it's trying to take advantage of near cards staying together in a normal shuffle
I do this sometimes in the interest of saving time dealing cards, but I didn't know there were games where this was the standard way of dealing. It does seem like my technique would affect the feel of the game.
>Hm, but this is even a problem with a traditional overhand shuffle, right?
To some degree, i suppose. But it still ultimately gets fixed if you shuffle enough.
>I'm not attempting a trick
I'm thinking less about deliberate cheating and more situations where you can kind of cheat by accident. Like, if there are four of each card, but only three piles, one of them will have two of a certain card. So what do you do if you accidentally notice that the top four cards are aces because that's the match you put onto the stack last, meaning there will be more aces in pile 2? You can't really erease it from your memory.
Maybe you have a way to solve that, or maybe you're just better at not looking for that kind of thing, and things like that will obviously happen less if you aren't trying to do it deliberately, but i still prefer to just randomize and avoid that kind of problem all together.
>maybe you're just better at not looking for that kind of thing
Yeah, I am usually chatting with the rest of the table as I shuffle the cards.
>They simply make the magical assumption that "shuffling" completely randomises the cards.
See, I feel like that as well, which is why I came up with my more thorough technique, which might not be truly random, but at least results in a less similar card order.
>Weird bait, but here's your (you)
Thank (You), my friend.
>There's nothing in your weird-ass method that prevents adjacent cards from staying adjacent, and your overhand shuffle almost guarantees that some will
It sounds like you think I'm cutting the deck into thirds? But I'm dealing the cards one by one into three piles, so adjacent cards will always end up in different piles.
I suppose 2 sets of adjacent cards may meet again when I stack the shuffled piles together again, but that chance is tiny and will result at most in 2 adjacent card reunions, which is fewer than I'd expect with a few regular shuffles.
>it's been mathematically proven that it takes 7 random riffle shuffles to properly randomize a deck of playing cards.
7 is less than I expected. It may be faster than my technique, especially for games with larger draw piles. But I wonder what a "random" riffle shuffle means.
My riffle shuffles are a bit sloppy, I'd probably need more than 7 shuffles to achieve the same effect.
>But I wonder what a "random" riffle shuffle means.
It's based on the Gilbert-Shannon-Reeds model, which was developed by observing actual human shuffling. Basically you're more likely to split the deck close to evenly than not, and when riffling are more likely to take the next card from the larger part than the smaller one.
>Gilbert-Shannon-Reeds model,
This is pretty cool to read about, thanks.
>Get a math degree before you start obsessing
The tone of your post suggests your feelings on this are a lot stronger than mine, but OK.
>But I'm dealing the cards one by one into three piles
Somehow that's even worse. I'd slap the deck out of your hands before you got the ninth card down doing that. Christ on a bike, you're so worried about about two cards being next to each other again that you've devised this overcomplicated mess of bullshit to compensate for it. Here's the truth: randomizing the deck is the goal, and random does NOT mean "every card in a different place and order." Random means sometimes two cards will be in the same place they were before. And all riffle shuffles are sloppy, yours aren't significantly worse than anyone else's. That's how they're *random*. For a game with your friends, three or four riffles and a cut is plenty to make the deck unpredictable and that's all you need. Get a math degree before you start obsessing beyond that.
The people math degrees say 7 riffles to completely randomize a 52 card deck. You riffle 10 times and your 60 or 100 card deck should be random enough.
If you're against mana weaving, you should also be against overhand shuffling or insufficient riffle shuffling.
You see, out here in the real world where we do things like "get together to play cards with our friends," we're not concerned *completely* randomizing the deck because we're not obsessive autists. Perfect is the enemy of good, and good enough is good enough. My euchre and spades groups would not stop talking shit about the OP method. Go to a tournament or have money on the line? Sure, get the full count in, but no one would tolerate dealing out three stacks and shuffling them separately. Hell, if money is on the line that shit probably gets you thrown out.
Did I say "copy OP"? No, I said riffle shuffle at least 7 times if you care about a fair game, and if you care about fairness do not allow overhand shuffles, because they're as bad as mana weaving.
And I'm saying "fair" is not a binary state for people without a diagnosis.
>Not binary.
Sure. There's a point where it becomes 'good enough'.
I would be inclined to accept 4-5 riffle shuffles as "good enough"; and 700-800 overhand shuffles as similarly "good enough". If I'm being casual enough to tolerate overhand shuffles, I'm going to tolerate mana weaving too, and we're clearly not playing for money or prizes.
>If you shuffle enough.
I have never seen anyone overhand shuffle the requisite 11k times required to randomize 52 cards, let alone more for 60 or bigger sized decks.
https://hackaday.com/2023/05/28/math-reveals-how-many-shuffles-randomizes-a-deck/
The general presumption is that shuffling is supposed to randomized the deck, but i don't see any reason in principle you couldn't do it differently as long as it's symmetrical.
One practical problem you'd have to solve is that the line between proper play and cheating can get blurred. If you're shuffling randomly, you just put all the cards in a stack and shuffle. But if you're doing piles, it suddenly matters what order you put suff in.
Is it cheating to put my cards back in the initial pile in a certain order that's advantagious after shuffling? Can i play the game in such a way that cards will line up in a certain way next round? Memorize or write down the order? How do you pick which of the three stacks goes on top?
Weird bait, but here's your (you).
What you're doing is weird and wrong. There's nothing in your weird-ass method that prevents adjacent cards from staying adjacent, and your overhand shuffle almost guarantees that some will. Seven riffle shuffles has been show to have sufficient entropy, and some casino tables - where randomness is dollars to them - use a table wash. If we played a game and you did that bullshit that you describe, we would stop until you were fully convinced of your sins, and not allowed to shuffle again until you had changed your ways.