Card Basic Problem

[Kayle]

@Nopoke: That's probably much smarter, but it feels odd to me to shuffle too much after checking the deck - not that I would want to undershuffle, but more that I am concerned with overshuffling because:
- I used to get bad hands all the time
- I was told I overshuffle
- I noticed I shuffle as a habit just when I'm walking around league
- Stopped shuffling to keep my hands busy
- Started getting better hands
???
So I prefer to check the deck and then shuffle until my opponent is satisfied it's random again (and no more). =\

Randomization is a weird topic with me recently.

 

[Roarkiller]

If it helps, this is what I do before every game.

1) Shuffle. Or table-shuffle (separating into piles), whichever I take fancy.

2) Look through deck to check for clumps.

3) If satisfied, I shuffle one last time.

4) Break up clumps.

5) Repeat from Step 1.

Simple steps to ensure as few bad starts as possible. Still happen though, but meh. I got 4-2 at my states last weekend with Jumpluff, with my two losses because of my bad decision in removing my 3rd sunflora the day before.

 

[Raen]

Simple steps to ensure as few bad starts as possible. Still happen though, but meh. I got 4-2 at my states last weekend with Jumpluff, with my two losses because of my bad decision in removing my 3rd sunflora the day before.

Statements like the bold one there disturb me. I almost feeling too much like people here really are actually getting dangerously close to cheating. Anything you do to "ensure" good starts is simply not right. You shouldn't be able to ensure ANYTHING related to randomness. Random is just that; uncontrollable. Anything you do to control the result of the randomness feels dangerously close to cheating to me.

Not saying what you do specifically is cheating, but what I'm trying to say is you people need to be more careful.

I for one will now always be shuffling my opponents deck rather than cutting it from now on. Seems safer that way.

 

[mikeg542]

I've caught an eerily high percentage of my pokemon opponents cheating. This is just one of the many ways

 

[yoyofsho16]

Actual objects moving and colliding with each other will ALWAYS be more random than a computer if you shuffle enough (barring regulated conditions). You're saying that "redshark is random, so look at how unrandom our real hands are" when it's the total opposite.

 

[Roarkiller]

I almost feeling too much like people here really are actually getting dangerously close to cheating. Anything you do to "ensure" good starts is simply not right.

There's a big difference between "ensuring good starts" and "ensuring as few bad starts". If you look at my step 3, I clearly state that I shuffled one last time after breaking up clumps. If doing this i equivalent to cheating, then all forms of self-shuffling would be considered cheating, because you can argue that his "style" of shuffling is letting him "ensure" a good start.

I mean really, if you see a stack of energy cards together, who WOULDN'T break up the clump?

Point is, there's no rule that limits you to how many times you shuffle. Whether you break up one clump or ten clumps doesn't matter (unless you're stalling), because so long as you shuffle your deck in the usual manner afterward, it would still be perfectly legal.

I think the root of the problem here is whether the situation happens during setup (i.e. before the game), during the game, or after. There is no rule that prohibit a player from looking through his deck before shuffling it, as long as he does not attempt a stack-shuffle. Similarly, during a game, there is no rule that states that you cannot rearrange your cards during a search, so long as the final requirement of randomly shuffling your deck is done.

And obviously there's no rule whatsoever that prevents you from doing anything to your deck after a game.

So until a player actually breaks a rule (stalling, INTENTIONAL cheating, intimidating, rules lawyering, dubious game action, etc), anything else you do is perfectly legal.

IF at the end of this post you still regard such shuffling as cheating, remember that you, the opponent, has the RIGHT to shuffle or cut the deck one last time, and if both players are still unhappy, just get a judge to shuffle it again.

And that should be a sufficient enough action.

EDIT: Good reference from good compendium, thanks to the other thread.

Q. I was playing with this guy and he rearranged the order of the cards in his deck prior to shuffle it, would this be allowed in a tournament?
A. Yes but you would have the right to shuffle it yourself or cut it as well if you were uncomfortable with his shuffling method. (Sep 14, 2000 WotC Chat, Q125)

Q. I play an Oracle, and search for two cards. While I am searching, I find that I have a HUGE energy pocket coming up. So I break it up and add some trainers in between them. I then shuffle and put 2 cards on top. Is that legal?
A. As long as you shuffle sufficiently afterwards. (Mar 25, 2004 PUI Rules Team)

So there you go. Perfectly legal.

 

[Kayle]

Actual objects moving and colliding with each other will ALWAYS be more random than a computer if you shuffle enough (barring regulated conditions). You're saying that "redshark is random, so look at how unrandom our real hands are" when it's the total opposite.

I'm given the impression that you know what you're talking about enough to feel like you have a point, but not enough that the point is quite correct.

If you perform a 'perfect' riffle shuffle (it has another name but I don't know how to spell it), dependent upon the number of cards in your deck, a certain number of shuffles will predictably get the deck back into the same orientation as before.

If you perform an imperfect shuffle, you can guarantee that there is at least one group of at least two cards that will maintain their grouping in the new shuffle. And overall, you can guarantee that this will happen every time you shuffle. The less perfect of a shuffler you are, the more of these groupings there will be (but the less random your shuffle will be as well, since perfect riffle shuffles can be predicted). In practice you can assume that each shuffling has several of these groups of 2-3 cards; next time you riffle shuffle, look through the deck before you put the two piles together and see just how many groups you've got.

From this it's easy to conclude that it's entirely possible, even likely, that if you start with a certain deck configuration and riffle shuffle repeatedly, your finished configuration will have several 2-3 card clusters that are identical to the previous orientation, just moved around in the deck.

The best way to get around this is to six-pile shuffle but the more you six-pile shuffle (compared to riffle shuffling) the more predictable your final configuration will be.

The grouping is even worse if you do what I call "cut shuffling" though I bet it's got another name, too - where you take some portion of the deck and insert it in parts in the remainder of the deck; a common quick method of shuffling I see at tournaments.

 

[TechMachine]

The issue here is #3 where you state ONE last time.

3) If satisfied, I shuffle one last time.

It implies that you shuffle once.
If someone only shuffles only once after looking at their cards, I would certainly consider it cheating.

In order for a 60 card deck to be sufficiently randomized more than 7 GOOD (non-perfect, but not 3 or 4 blocks either) shuffles are needed.

This is based on Dave Bayer & Persi Diaconis' work where 7 shuffles are considered a good distance towards randomness for a 52 card deck and little additional randomness is added after 11 shuffles.
Seven shuffles for a 52 card deck assumes that each of the cards is unique. With Blackjack (where suits are ignored) and Pokemon (where up to 4 of a particular card may be present), the number of GOOD shuffles is lower.
In addition, if the deck starts out in a non-ordered state, the number of shuffles needed is even less.

Humans don't shuffle well a lot of the time, so the number is increased again, making 8 shuffles probably an acceptable number.

You are also confusing "Random" with "Evenly Distributed."
When you are checking your deck, if you find a block of energy cards you break them up to evenly distribute them, but since in a "Random" deck the probability of any one card being next to another card is the same, having all the energies together does NOT mean that the deck is not random.

In summary, you need to randomize your deck, with non-perfect rifle/faro shuffles (and even a pile shuffle or two) around 8 times EVERY TIME you look at your deck for it to be near random. Any less, and, especially for 3 or less shuffles, the ability to guess the upcoming cards is there.

I'm not saying that I shuffle 8 times every time, nor that I require it of my opponents, but by accepting 5 or 6 instead of 8 I'm accepting a deck that is still 8-10 seconds away from being "random."

With regards to:

There's a big difference between "ensuring good starts" and "ensuring as few bad starts". If you look at my step 3, I clearly state that I shuffled one last time after breaking up clumps. If doing this i equivalent to cheating, then all forms of self-shuffling would be considered cheating, because you can argue that his "style" of shuffling is letting him "ensure" a good start.

There is difference between "style" of shuffling and adequate shuffling.

In the references from the compendium, the "shuffling" being referred to is "adequate shuffling" not "one shuffle" and defined in the Pokemon Tournament Rules (Section 20) as:

20. Shuffling
Each player’s deck is expected to be fully randomized at the start of each game and during the game, as
card effects require. In order to achieve randomness, players are allowed to riffle, pile, or otherwise
shuffle their decks until they are satisfied that the deck is random. Randomization must be done in the
presence of the player’s opponent and must be done in a reasonable amount of time. Care should be
taken to assure that the cards in the deck are not harmed or revealed during the shuffle.
...
Players engaging in questionable shuffling methods may be subject to the Unsporting Conduct section of the Penalty Guidelines.

Therefore looking at the cards constitutes "revealing" them and the deck needs to be fully randomized again (which may create clumps of cards).

A single shuffle after looking at them is not adequate.


References:
http://projecteuclid.org/DPubS/Repo...w=body&id=pdf_1&handle=euclid.aoap/1177005705
http://en.wikipedia.org/wiki/Shuffling
http://statistics.stanford.edu/~ckirby/techreports/GEN/2002/2002-16.pdf
http://www.dartmouth.edu/~chance/teaching_aids/books_articles/Mann.pdf
http://projecteuclid.org/DPubS/Repo...ody&id=pdfview_1&handle=euclid.aop/1147179990
http://assets.pokemon.com/assets/cms/pdf/op/tournaments/2011/Pokemon_Tournament_Rules_EN-1-31-11.pdf

 

[Kayle]

You are also confusing "Random" with "Evenly Distributed."
When you are checking your deck, if you find a block of energy cards you break them up to evenly distribute them, but since in a "Random" deck the probability of any one card being next to another card is the same, having all the energies together does NOT mean that the deck is not random.

I don't think you are responding to me, but I want to address this anyway.

You're right; an even distribution isn't the same as random (and they are often in some measure of conflict). Going through and breaking up a clump of energy does not mean that you're further randomizing your deck.

But what about if you - to restate a point I made earlier - gather up your cards at the end of a game, like your fully-set-up Dialga G Lv. X with 3 Specials and a Belt, and you put them into a pile and then shuffle THAT? There are clumps of cards in your original deck state, let's say the top ten cards look like this:

[benched Pokemon] [setup Dialga]
[Energy Gain][Garchomp C][Uxie][Call Energy][Special Metal][Special Metal][Special Metal][Expert Belt][Dialga G][Dialga G Lv. X]

Pretty reasonable for having just gathered up your deck, right? You perform several riffle or cut shuffles, which will move things around in the deck but may not adequately break up existing clusters, as stated here:

If you perform an imperfect shuffle, you can guarantee that there is at least one group of at least two cards that will maintain their grouping in the new shuffle. And overall, you can guarantee that this will happen every time you shuffle. The less perfect of a shuffler you are, the more of these groupings there will be[.]

They may not be near the top of the deck, but I bet even if you do shuffle a great deal or very well, you will find that perhaps one out of three times, if you check your deck after what you consider to be "good shuffling", you'll find that there are probably one or two 2-3 card clusters that are sampled from the above. Say, all your Special Metals stay together. Or say you draw Garchomp C, Uxie, Call Energy in your next game. Or perhaps Special Metal, Expert Belt, Dialga G. Et cetera.

(It gets worse when you think about the often generous spread of very useful trainers and decent Pokemon in your Discard Pile, especially in non-tank decks.)

In this way, the clusters make your deck predictable, which is the opposite of random in this case.