What is Random and How to Shuffle, or Why You Shouldn't Declump

[Bolt]

there's a lot of math in this thread. I look at it like this...

My opponent, mid-game, starts looking through their deck and rearranges cards. I don't know if they're "de-clumping" or stacking. I'd be fine with it if they showed me exactly what they are doing. I want to see what cards they are moving where.

Otherwise, I will just assume they are cheating.

 

[pokedad1]

Pokemon players, this is the solution to the problem ...

http://www.youtube.com/watch?v=17hVv28Yzzo

 

[SD PokeMom]

those don't work with sleeved decks...

 

[TheRolesWePlay]

Before this, there's been a lot of discussion about declumping. And this seems like some sort of natural progression past that. Or declumping where it worked perfectly. The reason why there wasn't that much of a deal with declumping before is because people just shuffled afterwards. And then people were like, "Well it's not actually cheating, because nothing is happening, because people are randomizing their deck." But I want to make something clear, that when you declump you have the intention of cheating. And for me that's the same exact thing, and you shouldn't do it. I know some people don't realize, and after they declump they shuffle correctly, but you shouldn't do it. And if you declump anyways, even if you shuffle correctly afterwards, you had the intention of manipulating your deck, in a way, to your advantage. And that's cheating, or the intention to cheat. And nobody should do either of those things.

I think declumping opened up the can of worms. "I can declump? What if I just shuffle a couple times afterwards?" They're trying to take it as far as they can. Pokemon's going to have to step in, smack some hands.

I have a bit of a bone to pick with this, specifically with the bolded statement. Declumping doesn't equal the intent to declump. I know people who go after clumps both via 'declumping,' and giving a brief moment to focus their shuffle on that part of the deck they know where a clump exists and then resuming a normal shuffle. Part of the issue that people are conveniently ignoring is that when some people see a clump it occurs to them that the deck isn't random. While yes, statistically clumps are random, that isn't how the person's mind processes it, and they declump as a result. Declumping is more of an action to set the player's mind at ease than anything else in that sense.

That's why intent is so crucial here, and also why this issue is also a far trickier issue to handle, since those who declump for the arbitrary reasons above aren't doing anything inherently wrong, but there are those who try to use similar means to their advantage. A good resolution to this isn't going to be easy to find, and resorting to calling those who declump cheaters, won't resolve the issue.

 

[bullados]

Declumping doesn't equal the intent to declump.

This is funny. I don't know why this is funny, but this is funny.

 

[signofzeta]

The thing is, there is no surefire way to check if something is random at any individual point in time. The only data you can gather is from previous shufflings, but that data doesn't matter, because the only data that matters is the shuffle you make right now, not the one after, not the one before.

People must realize is with random, what you get is what you get. You can't force something to be more random when it is already random, such as, you don't see any clumps, so you shuffle again until you get clumps. That isn't how it works. The fact that you didn't get clumps is the fact that it is random, and because the randomness picked a order of cards that doesn't have any clumps. That's what random is.

There is no such thing as not having certain combinations because it is random. It may happen less, it doesn't mean that it does not happen that you have to reshuffle your deck because it isn't random, because the fact is that, the deck is random.

The mistake I see people make here is that they use data from other shufflings when that data does not count. You only count the shuffle you make now, and not any shuffle. The only data to make is probably if you constantly get the same order of cards, then yeah, there is a problem, but just because you don't get clumps in one shuffle out of many doesn't make the deck not shuffled randomly.

The other thing is deck makeup. If a person's deck is built in a way that by randomness, you don't get clumps, ahem, singleton decks, but who makes those anyway, then they shouldn't be labeled a cheater because they built a deck a certain way.

The thing with random is you want to lose control over where your cards end up, hence saying, when you shuffle, don't think. By trying to make the deck more random, you are inadvertently controlling where the cards to go through your preconception of "random", and therefore, you aren't make your deck random. By doing that, you are thinking. Don't think. Just shuffle, and if you don't think, and don't look at your cards, even if the deck wasn't super random, your lack of thinking and looking would gain you no advantage towards knowing what the next cards are in the deck, because you didn't keep track of them while shuffling. Of course you probably have to shuffle it so many times anyway.

The other thing as to why people shuffle like crap is because they don't want to shuffle. They want to play. I've had fits of rage when I have to shuffle my deck and I keep on getting the same crap over and over, and it is annoying. Not random at all, until I discovered mash shuffling with sleeves. Before I didn't use sleeves and did the hand over hand shuffle technique, which sucked, especially if you have bent cards, so to fix that, I alter the starting condition of the shuffle, so that the end result would be different than before. I would move cards around without looking at the cards, and do the pile shuffle thing. Not as random, but you can't expect people, especially kids, which this game is aimed at, to be master shufflers who could do it in a short amount of time. Oh, that was when I was playing MTG all day every day, before I decided to split my time and cash on 3 TCG's. It doesn't matter now, sleeves + mash shuffling = win. Much quicker, and I don't have to deal with pile shuffling anymore, and it is more random, well not totally random, because 100% random is impossible. So long as it is 70% random, it is good enough for me. I am not going to shuffle till I am blue in the face just to achieve the 100% randomness, but then again, I shouldn't be thinking about getting that hard to get card order through random shuffling, because when shuffling, you don't think.

I do have one problem though, when shuffling MTG intro packs, yugioh structure decks, or Pokemon theme decks, I tend to have the urge to reshuffle when the foil/holo/ultra rare is the top or bottom card, and that is when I shuffle the deck and not play, because I felt like shuffling out of boredom, even though I know what I get is what I get. It maybe has to do with the way I cut the deck because foils/holos usually bend, and they mostly end up as the bottom or top card, which is kind of a bummer.

 

[NoPoke]

... The only data you can gather is from previous shufflings, but that data doesn't matter, because the only data that matters is the shuffle you make right now, not the one after, not the one before.... BIG SNIP....

Major, enormous, total, complete error highlighted in RED. The rest is bad too but the most egregious error is highlighted

Since shuffling cards is proving to be hard try applying your argument to the design of a paint mixing machine. The only way to know if the mixing machine is doing what you want is to look at the final result. The only way to know if the outcome from the mixing process is repeatable or a fluke is to carry out trials where the parameters of the process are varied and tests are repeated.

 

[khit]

Part of the issue that people are conveniently ignoring is that when some people see a clump it occurs to them that the deck isn't random. While yes, statistically clumps are random, that isn't how the person's mind processes it, and they declump as a result. Declumping is more of an action to set the player's mind at ease than anything else in that sense.

That's why intent is so crucial here, and also why this issue is also a far trickier issue to handle, since those who declump for the arbitrary reasons above aren't doing anything inherently wrong, but there are those who try to use similar means to their advantage. A good resolution to this isn't going to be easy to find, and resorting to calling those who declump cheaters, won't resolve the issue.

The bolded is exactly why intention means nothing. You can't tell why someone is moving cards around in their deck, which can be used to manipulate the odds in their favor, therefore declumping should not be allowed in my opinion. Just because some people incorrectly think clumps are a sign of non randomness, doesn't mean they are. If i say 1+1= over 9000 that doesn't make me right. Not to mention if you shuffle properly after the declump didn't matter, if you don't shuffle properly, you are cheating, whether you think 1+1= 9000 or 2, it's still 2

Maybe there shouldn't be any cards allowed to be moved during a search. I know a lot of players move cards to the front when they're deciding but honestly its not that hard to decide in your head, any competitive player knows what every card in his/her deck does, and if for some reason they don't, they can just look at it in his/her deck without moving it to the front.

 

[Otaku]

...since those who declump for the arbitrary reasons above aren't doing anything inherently wrong, but there are those who try to use similar means to their advantage. A good resolution to this isn't going to be easy to find, and resorting to calling those who declump cheaters, won't resolve the issue.

Everything is permissible"--but not everything is beneficial. "Everything is permissible"--but not everything is constructive.

I realize some will be instantly turned off by the source of that quote (1 Corinthians 10:23, NIV), and no I am not trying to turn this into a religious matter. The quote establishes my point and demonstrates it isn't a new idea to the world that there are some things a person can do that but shouldn't do.

It is not illegal to declump as long as you sufficiently shuffle afterwards. Such shuffling reduces declumping to a "good luck ritual" but still provides a smokescreen for various forms of cheating and can still result in someone gaining undue advantage against the rules (e.g. cheating) if they fail to properly shuffle... and we know not everyone can or will properly shuffle, due to inability or haste.

The only reason to declump is because "it makes you feel better". If you're heading to Nationals this weekend and you can't break yourself of it before, I get that you're going to declump and I won't hold it against you. Still, for the sake of a feeling you're putting others at risk and providing cover for those intentionally cheating; isn't that a good reason to make this the last season you declump?

 

[NoPoke]

The case against declumping is that it might be used to cheat.
The case against declumping is that you are going to shuffle properly anyway so why waste time moving cards around.

But what if you are not very good at shuffling? Now a sufficient randomisation that can break up clumps of three or four cards is going to take a lot of time and fall foul of the requirement for a brief shuffle after a mid game search. Avoiding a disadvantage you are not supposed to have is not the same as seeking an advantage.

So the case against declumping presumes that many players may cheat if given the opportunity and that they can shuffle properly anyway. Unfortunately neither assumption is really true. If you must declump keep it brief, the card says "search" after all. How a player searches is not prescribed, it would probably play into the hands of the rules lawyers if it was. If your deck is obviously not well randomised when you search tell your opponent what you are doing.

Declumping remains the smokescreen it always was. The fire that we should be concerned about is stacking. Judges and players should be looking for insufficient shuffling and not worrying about or legislating the detail of how a player searches mid game.

 

[M_Liesik]

I'll just leave this here...

 

[Otaku]

I'll just leave this here...

Well that certainly is an article. I want to agree with it, unfortunately it added some references in the middle that make me mistrust it.

 

[signofzeta]

Major, enormous, total, complete error highlighted in RED. The rest is bad too but the most egregious error is highlighted

Since shuffling cards is proving to be hard try applying your argument to the design of a paint mixing machine. The only way to know if the mixing machine is doing what you want is to look at the final result. The only way to know if the outcome from the mixing process is repeatable or a fluke is to carry out trials where the parameters of the process are varied and tests are repeated.

Ok, maybe I was wrong that the data doesn't matter, but when you are analyzing an isolated incident, you don't take what comes before, but what comes after as well, that is, it is obvious that guy shuffles like crap if that guy keeps on drawing the same cards all the time.

Again, with random, there is no such thing as should be this, should be that. When something should be something, it means the alternative doesn't happen at all. So you are basically saying that there is 100% chance that every shuffle will result in pairs of 2, but that is ludicrous.

When something can happen, it means that the alternative can also happen as well.

You can't say that for each individual instance of shuffling that there should be pairs in the deck, but you can say that in that person's overall shuffling "career", that you can say that there should be pairs in the deck more often than there aren't.

Labeling someone a cheater just because that person didn't get a pair in the deck in only one case is just foolish, and you are assuming everybody uses the same deck, which they don't.

Again, when you shuffle, don't think. The more you think, the more you try to look for patterns, the more you look for patterns, the more you try to shuffle to the opposite of that pattern, which means you are controlling that shuffle. When something is random, there is no such thing as pattern and control, but unfortunately it is hard to tell who has a clear mind when they shuffle.

Again, when you shuffle, for each individual time you shuffle, you don't look at your deck. You say that you should look at your deck to check how random it is, but that defeats the purpose of shuffling, so you don't know what cards are going to come up in your deck. Just shuffle and play, it's that simple. People aren't here to gather data on the order of their decks. They want to play.

Here is a cool fact. If people don't know what cards are coming up in their deck, then they can't think ahead and formulate strategies, no matter how not random the deck may seem. You don't force a deck to become random, you let the deck be random. In Soviet Russia,You don't control random. Random controls you.

 

[NoPoke]

When I say that a player should "expect to see pairs" after a shuffle I am making a prediction. A correct prediction. When you read this as "always see pairs" you are making an error. In addition why do you insist on declaring that I am saying players are cheating if they don't see pairs?

 

[signofzeta]

When I say that a player should "expect to see pairs" after a shuffle I am making a prediction. A correct prediction. When you read this as "always see pairs" you are making an error. In addition why do you insist on declaring that I am saying players are cheating if they don't see pairs?

Well you did say that people should look at their decks after they shuffle to check if their deck has any pairs, as if they were cheating if they didn't. The worst thing about expecting something to happen is that when it doesn't happen, you take drastic measures to make that event happen, like I don't know, manipulate your deck until you see pairs. When something is random, in any given individual scenario, you don't expect anything. So you didn't get any pairs. That's part of randomization.

When you absolutely have no data on the order of cards that some person shuffled into for the past 20 or 30 times, then you can't say that you expect that his deck will have pairs after the next shuffle, because it isn't guaranteed. Even if you gathered the data, and you plotted some sort of curve, there is no guarantee that for sure that the deck is expected to have pairs in them after the next shuffle. You can't expect anything when it is random.

When disqualifying people, predictions don't cut it. Cold hard facts, and the history of that player do. You can DQ someone for the fact that they kept on drawing the same cards, or that for every game, there seems to be some sort of pattern with the order of cards for that person. You can't DQ someone based on what you expect that person's deck order to be based on probabability, unless the probabability of getting pairs in the deck is 100%.

I say this again. There should never be a "should" when describing random scenarios. You expect event A to happen, or event A should happen. If there is only one alternative, that means that you expect the alternative NOT to happen, and that it should not happen, but that isn't how randomization works. Both events can happen, just one event happens more often than the other, it doesn't mean that the other doesn't happen.

I just hope you don't use those high expectations that the deck should have pairs to judge if people are cheating or not, because that is a poor way to judge someone. You can, although, expect someone to have pairs, if that person hasn't shuffled into some pairs at least once or twice in a row, or else, you can't expect anything. If on the other hand, that person has pairs after 20 or 30 or even 100 shuffles, then there could be a chance that that a pairless deck order may soon come up. Note that I never said anything about "should", because a pairless deck order isn't guaranteed.

Again, with randomization, nothing "should" happen. It "can" or "may " happen.

 

[Tapin]

you can't say that you expect that his deck will have pairs after the next shuffle, because it isn't guaranteed.

zeta, you've shown a lack of basic knowledge of probability and statistics over and over in this thread. Please stop.

Expectation is a basic tenet of probability. Expectation is not a guarantee. Expectation typically comes with a percentage attached, and I believe that Ian has already mentioned "90%" on more than one occasion.

There should never be a "should" when describing random scenarios.

As has been mentioned before, this is flat-out false. Everything about this game we all play hinges on probability of occurrences based on known initial conditions and two specific random elements -- each players' deck. To claim that any of the following statements is false shows a fundamental misunderstanding of the majority of the terms you're trying to use:

- If you have four copies of a card in your deck, you SHOULD draw one of them off your first seven cards 40% (well, 39.94996%) of the time
- If you have three copies of a card in your deck, you SHOULD draw one of them off your first seven cards 31.5% of the time

- If you have one copy of a card in your deck, you SHOULD expect to pirze it 10% of the time
- If you have two copies of a card in your deck, you SHOULD expect to prize both <1% of the time

When Ian claims "you should expect to see pairs" and "if you never see pairs you're not shuffling enough", he's saying that IF YOU WERE to check your deck -- which you may do at home, or while sitting around between rounds, or while fooling around with friends outside a tournament atmosphere, or (wild example) as a result of running a large Monte Carlo simulation because this topic is important enough to you to do research on it -- you SHOULD regularly expect to see pairs. Not "everyone needs to always look at their decks after shuffling"; that's just ridiculous, and takes a huge leap to somehow read that into his words.

I'd really like to know whether the straw men you keep attacking are intentional or unintentional constructions; but either way, you're not furthering the discussion.

 

[signofzeta]

When Ian claims "you should expect to see pairs" and "if you never see pairs you're not shuffling enough",

I guess I had the definition of "expectation" wrong then, and I misinterpreted his post as, you should always see pairs, and if you don't, shuffle until you do. Oops. But still, it is expected that people don't do shuffling research before they play high end tournament games, and check if they are shuffling properly. I mean, who does that?

But how about judging if someone is shuffling properly or not? Without the vast data that only that person knows about his deck order in the past 20 or 30 shuffles or more, you don't expect anything.

 

[chrataxe]

So, we're back here...again....really?

While OP tried to mold the idea of randomness to support his cause, I think he actually did the opposite. There is an idea in mathematics that deals with how random IS random? For example (to use Mike's link), how random is an MP3 player? Yes, it will play songs at random, but it will ONLY play the songs that are in your library. Um, well duh, right? But, in order for the selection process to be truly random, how can you say "play songs at random, but only use these songs?" Yes, there are elements of randomness, but it is not truly random.

I get it, its a bit too philosophical. The point I'm making is, OP's definition or random is very selective. Is rolling a di random? The outcome, yes. The choice to roll? No. What if I made a deck that played 59 energy and one basic. What would the variance of my opening hand be? Zero. The opening hand i will start with every game is 1 basic and 6 energies. Stupid deck? Yes, of course. But, the purpose here is to eliminate "randomness." Is it possible to have a random variance with this deck? No. But, oddly enough, according to OP's definition of randomness, part of randomness is to "have no definite aim or purpose." By eliminating variance, I've used both aim and purpose. Now, that deck is a bit ridiculous, so lets use "real" decks.

Lets use a card like Twins. It's a great card, but really should only be 4-ofs in decks like Lost-gar or Durant. Barring decks like that, two may be a bit much, 1 is very good, especially late game when you deck is thin and when you NEED it. So, lets say I've run one in my deck instead of two. Now, the idea is that I don't want to draw it early, I wan't it to be my closer in the ninth. That is is giving the deck itself "aim and purpose." So, how well I can I REALLY randomize a deck that I've purposely stacked to give me the best randomization? I can't.

The problem is, there is no central definition of "randomization" for anyone to base an argument. Lets use Magic as an example. Is there a need to clump or declump in magic? There are some draws in Magic where clumping is VERY good, say 2 lands and 3-4 burning tree emissary. That paired with a Champion of the Parish is hard to overcome. Would it be OK in magic to dig through my deck and put 3 Burning Tree's together, then shuffle? Not really. Then, why do I advocate that it is OK to declump in Pokemon? For the same reason that I don't necessarily have a problem with someone CLUMPING in MtG. If you just put all 4 together, I'm just going to six pile your deck, the riffle it a handful of times. One interesting thing in MtG is the mulligan...its optional. In other words, if you aren't comfortable your decks randomization, just shuffle it back in and let it be randomized again....if you don't like that, do it again...

If the purpose of the definition of random in Pokemon was to be implied as OP has suggest, there are two HUGE flaws in the game: first, the game wouldn't be able to exist because we have to have rules that govern deck building, which takes away from randomness in the first place. Second, mulligans couldn't happen. Yes, you HAVE to have a basic, but if the deck is randomized and spits out crap and you shuffle it back in, you've defeated the purpose of randomization.

OP and other players need to understand the TRUE purpose of shuffling. While randomization is part of it, the POINT is to ensure that players aren't stacking the deck and have no way of knowing what they are going to draw. If I declump and shuffle, I've met this burden.

Lets reverse the scenario: If I were doing a deck search for a basic Pokemon and I wanted to grab Mewtwo Ex, but I also want to see how many are in my deck for the purpose of knowing how many are prized. Its perfectly legal for me to move all four to the bottom of the deck. I then take one, leave the rest clumped, then shuffle. Is that a problem, especially when you consider that is is potentially more harmful than helpful? Lets assume its even less drastic: lets say I'm not sure if I want to choose Mewtwo or Mew. I dig, find a Mew and move it to the bottom, then grab a Mewtwo and move it to the bottom. I set my deck down and say "I haven't decided, I'm going to look at my hand." I take a quick glance at my hand, make my decision, then take one...but, by displacing that one card, I've changed the random order. Lets even go more innocent. Lets assume I play Skyla and start rummaging through my deck for a dark patch and pull it to the bottom...but, then another play dawns on me: I check my discard for a dark patch, see one, then decide I want my Dowsing Machine instead since I can get it back with Sableye AND discard the energy...so, I take the Dowsing Machine...does this not defy OP's definition of randomness?

TLDR; we get it OP, you don't like declumping, but that doesn't make it cheating. We can we get over this?

 

[Otaku]

What if I made a deck that played 59 energy and one basic. What would the variance of my opening hand be? Zero.

Actually, there is variance because... you have to have 59 different Energy cards. They may all be the same Type, but there would be (for example) :fighting: Energy #1, :fighting: Energy #2, :fighting: Energy #3, etc.

Otherwise, it seems like you are intentionally distorting almost everything you can from Raen's post. Yes, it is a legitimate debate tactic when someone leaves something completely open ended to take it to the extreme. Raen didn't do that; you aren't waxing philosophical. You're either

a) Suffering an epic reading comprehension fail
b) Intentionally obfuscating the issue by playing with semantics
c) Something I didn't think of in which case, let me know :thumb:

We are discussing the concept of a "random deck" in the Pokémon TCG. If you read the thread, you'll notice I even had a disagreement with the definition used.

The argument you make that low level stacking (a.k.a. "de-clumping") should be legal because as long as you follow procedure, all you're doing is wasting time and risking unintentional cheating. That said, the practice is permissible because of enforcing rules against it is nigh impossible and technically it is legal. I will not deny someone their in game "right" (using that word makes this sound far loftier than it is) to "de-clump", but I will let people know it is a poor practice, even if some great players still do it.

TL;DR: We get it, de-clumping is legal so long as you follow careful steps that most players either ignore, forget about, or have difficulty executing, and following them doesn't stop the behavior from providing an "out" for would-be (or successful) cheaters. It's a bad practice; continue it if you want but just like you can champion it, others can warn against it.

 

[vaporeon]

@chrataxe, you said this perfectly. Also your last point. The OP does not like declumping, as well as a hand full of others but just because you dont like it, it does not make it evil or cheating.