Head Judge Banning Coins

[psychup2034]

I've read the report. published here in full http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2789164/ (its a bit brief!)
The experiment was carried out some years ago.
It struck me as very odd that I can't find any repeats of the experiment to confirm some of the rather strongclaims made. 68% bias for example. easy to learn technique...

Being a skeptic does not mean I dismiss it as bad science. It means I seek corroboration. But I can't find any beyond anecdote.

One of the reasons why this report probably wasn't widely replicated is that

1. It's a relatively boring topic to do a study on
2. It's widely known in the field of statistics (and among statisticians) that the coin flip is one of the easiest randomization processes to manipulate
3. You don't get any grant money for doing a simple study like this, and it takes a lot of time to do it right
4. Thus, it's a waste of time to continue doing a study that any high school 10th grader with an elementary knowledge of statistics can do at home

 

[NoPoke]

I did not get the same results when I tried at home. I read the report and was very disappointed that it does not mention this easily learned technique.

Back to 11th grade for me

 

[psychup2034]

It's all about muscle memory. Flip the same Pokemon coin for half an hour, starting with the same side face up. Try to use the same force each time. Notice how much force you're using when you flip heads, then notice what you're doing differently when you flip tails. I'll post something about how to manipulate a Pokemon coin to Nick this weekend/next week, and have him show you.

That's why I always use dice. I don't want to potentially tarnish my reputation while using coins in Pokemon.

 

[NoPoke]

I've tried... but the coin always flips over many many times. The only way I can conceive that anyone can flip heads reliably is if the coin does not travel far and does not spin fast. Short flight time/path flips are against the floor rules so anyone who has mastered coin flips needs to have also mastered long flight flips: total flight path approaching 1/2 meter and 1/2sec. Personally the only way I can manage to get a coin to rotate slowly is to throw it into the air rather than flip it, but I don't class those as valid flips either.

So I can't do it. But maybe others can. A reason I'm skeptical is that I've tried just dropping a POP coin from around 30cm and even when I try to avoid any rotation at all the coin still bounces and yields many results that are not same side.

 

[Ditto]

I was glad the conversation had not yet steered in the direction of "I can flip a coin to 80-20% heads-tails." If the coin is not damaged, worn, altered, it really does not matter. Since cards that use tails as a favorable outcome are popular in format, it sounds like something that player may want to unlearn.


If the player had a history of using a damaged or invalid randomizers (in this case, a coin), then I could see where the judge ruled the player had to change the randomizer. That would have been a ruling for using an invalid randomizer as directed by the rules, not a penalty for using a coin. That would have been information that should not have been made known to you by the judge. You would have only seen the side of the conversation of "your opponent's randomizer is invalid."

From what you said, it sounds like that may not have been the case since it would have been beyond the judge's power then add "only the dice we provided" and "no coins allowed."

Right, I gave the whole story since people seemed to have wanted it. The important part was when the head judge came over and directly said that NO coins were allowed. Anything else is really just for completeness.

 

[PokePop]

Note: Unless I'm posting in Ask the Rules Team, I am not giving an "official ruling" unless I specifically state:
"This is an official ruling".

Now, Biggie did follow up and confirm what I posted. That's the important thing! He gives official rulings on things like randomizers.

As for the study that someone got 68% heads:
Well, duh.
That's how statistics work. You will get a bell curve of results, with some people getting 68% heads, others getting 34% heads, and most people clustered around the midpoint of that.
There will always be outliers.
That's why it's called "Random".

 

[psychup2034]

As for the study that someone got 68% heads:
Well, duh.
That's how statistics work. You will get a bell curve of results, with some people getting 68% heads, others getting 34% heads, and most people clustered around the midpoint of that.
There will always be outliers.
That's why it's called "Random".

Pokepop, unfortunately, that's not how statistics works. You're assuming a normal distribution without looking at what the data is.

Provided the sample size is sufficiently large, the 68% result could be statistically significant. From the study:

The highest proportion of heads achieved was 0.68 (95% confidence interval 0.62–0.73, p < 0.001).

In other words, there was a sufficiently high sample size such that one can be more than 99.99% confident (p < 0.001) that the 68% result not the result of a truly random flip.

If you'd like a more mathy explanation, I'm happy to give you one via PM.

 

[Pidgeotto Trainer]

The results from the linked study are stunning!

13 people were given minutes to practice a flipping technique and one was able to achieve a success rate (68%) that had a statistical likelihood of less than 1/1000.

That's INSANE.

Think of what is possible for someone (like a Pokemon player) if they practice for far longer than that. Coins should be BANNED. This study just confirms what I (and many others) have thought for years, coins are too easy to manipulate. Even if you can only get say 55%, that's not fair, that's not random.


I may have lost worlds on a dice roll (my own). If I had lost worlds on an opponent's coin flip, wow....I would always have to wonder. (To be clear, I do not think David would do this, but I would distrust anyone who insists on using a coin.)

Even if an opponent wasn't trying to cheat, what if their grabbing the coin to flip instead of you changed the odds ever so slightly? You would always wonder. There are so many implications if you have anything but random. A clear, plain dice is much more trustworthy than a coin. Plain and simple. Just as that study shows, there is nothing inherently 'pokemon' about a coin. A coin is the same old, lazy method that has been used in disputes forever. A dice is better in all practical ways.

 

[FunnyBear]

All participants achieved more heads than tails results, with 7 of the 13 participants having significantly more heads results (p ≤ 0.05). The highest proportion of heads achieved was 0.68 (95% confidence interval 0.62–0.73, p < 0.001).

All more heads does not seem like a random result to me, even if only 7/13 had statistically significant more heads. The question is, can you take the guy that got 68% and have him reproduce it multiple times?

 

[PokePop]

All participants achieved more heads than tails results, with 7 of the 13 participants having significantly more heads results (p ≤ 0.05). The highest proportion of heads achieved was 0.68 (95% confidence interval 0.62–0.73, p < 0.001).

All more heads does not seem like a random result to me, even if only 7/13 had statistically significant more heads. The question is, can you take the guy that got 68% and have him reproduce it multiple times?

That's the thing.
That study is a good first step. It indicates something.
However, as a scientist, one study with 13 participants doing 300 flips is a small data set.

More study would need to be done.

How about we set up a controlled experiment at Nationals, using good condition theme deck coins, a lot of players, and a lot of flips, and gather a lot of data?
We could run it at the PokeGym Mutant Draft.

I wonder if there is a Guinness World record for the number of people flipping a coin at the same time?

 

[Raen]

That's the thing.
That study is a good first step. It indicates something.
However, as a scientist, one study with 13 participants doing 300 flips is a small data set.

More study would need to be done.

How about we set up a controlled experiment at Nationals, using good condition theme deck coins, a lot of players, and a lot of flips, and gather a lot of data?
We could run it at the PokeGym Mutant Draft.

I wonder if there is a Guinness World record for the number of people flipping a coin at the same time?

This. You need thousands of people doing thousands of flips before you really have a good, solid set of data. 13 people and 300 flips is a very, very small sample in the scheme of things. Generally, the higher the number of flips, the closer you get to 50/50.

 

[JandPDS]

Anyone who has ever played this game for any time knows that the coins can be manipulated and are not truly random. IF that was not the case then die would never have become the randomizer of choice. For one thing the coin as it can be used in tournaments is not a true randomizer. The person using it gets to select which side that they get to put on top. IF it was truly random then they should have no prior knowledge of if the heads or tails side of the coin was on top or bottom when they began the flip. In order to be truly random a coin would have both sides covered until the coin landed the outcome would be revealed after the coin lands. But the the flipper gets to pick which side they choose it is not random at all. When coin flips are used in sporting events or to settle an election or something like that the flipper is never the caller. So the person choosing heads or tails does not control the flip. That is a random outcome as the coin could outcome could be heeds or tails and the correct choice could also be declared heads or tails. Pokémon is not random. The flipper knows the winning choice (it is always heads) and they have direct stake in the outcome. So it is in their best interest for them to create as many positive (heads flips) outcomes as possible. For coins to be random them a judge would have to flip it and the person would have to call heads or tails in the air. But having the person flipping and calling is not a random or fair.

The die used to be manipulated in this way. That is why it the die must now be rolled in the open palm of the hand and then rolled. A die can be made to be less random as well as a coin. Yesterday day I was playing a durrant player who picked up a jumbo die for crushing hammer and dropped it a few inched off the table. It landed completely flat and did not bounce, landing on the 6. Now if I was not already 2-3 at the time and cared about the outcome of the match then I would have made her reflip as this was an obvious manipulation of the die (Jumbo die should not be allowed anyways) But my deck cant beat a durant even if they have no crushing hammer head flips so it did not matter.

But if it is well accepted that a die with 6 sides can be manipulated in such way that the outcome is not random, and we had a rule change as such, the how can anyone disagree that a coin with only two sides is not subject to being made less then random as well with its outcomes.

Coins should be banned from Tournaments. I do not see any reason for there being allowed. I know they are because in Japan a die = gambling and they did not want gambling to be a part of Pokemon. But there is no such association here so we should not use them. And in Last Chance Qualifiers in the past only die have been allowed. That was a good idea that I wish would be standardized for all official Pokemon Play events.

 

[PokePop]

And in Last Chance Qualifiers in the past only die have been allowed.

That is not a correct statement. As I noted above, POP would never disallow coins as it would not be tolerated by Pokemon Japan. They did supply dice to be used as randomizers, and if you were to use a die as a randomizer, it had to be one they supplied. But they did not ban coins.

 

[JandPDS]

That is not a correct statement. As I noted above, POP would never disallow coins as it would not be tolerated by Pokemon Japan. They did supply dice to be used as randomizers, and if you were to use a die as a randomizer, it had to be one they supplied. But they did not ban coins.

Excuse me I was there. We were all told before the LCQ started the first time it was in Hawaii that the only radomizer allowed to be used was the event provided die that was at each tabel. We were forbidden to use our own die or our own coins for randomizing outcomes. Ture I am unsure if this has happened again after this event. But it did at this LCQ. ( I will never forget that LCQ as I went 6-2 and finished in 9th place when they tool the T7 into worlds) So it for certain happen once before. And this was before clear die were mandated for coin flips.

 

[PokePop]

I was there as well.
And I am positive you are mistaken.
Yes, there were many players who misunderstood the directive that if they wanted to roll a die, they could only roll the die provided and heard it as "randomizer", but that is not the case.
I corrected that misunderstanding a few times for players at the event.
So, the supplied die was mandated if you wanted to use a die.
I am 100% sure about that since I was not only there, but was involved in the planning meetings for the event.

 

[jaws_is_ra]

Even thought it has been say many times by many people I think its easier to put it like this

A Judge can ban an official coin if they have a reason to believe the fairness of said coin is in question

A Judge can not ban all Official Pokemon coins in genral from an event

A Judge can ask you to use a dice over a coin but can not mandate it (Asking and mandating it are 2 diffirent things)

A Player can humblling decline a request to use a dice if the coing is a Pokemon official coin and the coin in question meets the guide lines to stay legal

A Judge can mandate that a special kind of dice is used but can not denied official coins from being used.

A Player can get a penilty if flipping a coin becomes a problem (Ie the coin going all over the place or trying to to do trick flips) but can not be peniltized for using an Official Coin.

If ordered to use Dice over Official Coins is a Big no no (As long as the coin meets all the guide line to remain legal) but Asking to use a Dice over coins is oked

I think that sums it up a lot better then all the examples people are trying to give and all the debate what if this happens and a Judge should and should not do.

If I missed something in summing it up to an easy way to read I am sorry

 

[Ditto]

Even thought it has been say many times by many people I think its easier to put it like this

A Judge can ban an official coin if they have a reason to believe the fairness of said coin is in question

A Player can get a penilty if flipping a coin becomes a problem (Ie the coin going all over the place or trying to to do trick flips) but can not be peniltized for using an Official Coin.

I'm sure you probably have the right idea, but just so others don't get confused, I wanted to clarify the quoted parts.

The judge can NOT ban a coin because they don't feel it is fair. They can only ban a coin if it has been PHYSICALLY ALTERED. A judge is to believe an official coin is ALWAYS fair unless it has been altered.

A player will not get a penalty for flipping a coin, or even for not knowing how to flip a coin in your second quoted statement. A player could get a penalty for slow play, or for stalling, or for something like that, but not for flipping a coin, and not for being unable to flip a coin. The coin has absolutely nothing to do with any penalty in this case.

 

[Cyrus]

The last time this debate got so big, I asked if anyone was aware of PCL's explicit intent when introducing the coin-flipping mechanic.

Maybe I missed something, but it seems like nobody was able to provide an explicit answer to this question. So I'll ask it here again:

PCL explicitl

 

[Ditto]

The last time this debate got so big, I asked if anyone was aware of PCL's explicit intent when introducing the coin-flipping mechanic.

Maybe I missed something, but if I remember correctly, nobody was able to provide an explicit answer to this question. So I'll ask it here again:

PCL explicitly wants its players to use coins; otherwise the cards wouldn't say "flip a coin." But what is their reason for this?

*Do they want random outcomes?
*Do they want to test our skill at flipping coins?
*Do they do it just because it's cool?

If someone can provide a sufficient answer for their intent, then I (and maybe others) would find that massively valuable in figuring out just where to stand on this. The same goes for Rock-paper-scissors.

Some of you may think I'm splitting hairs, but this is a serious question. It's awfully suspicious that they've never, ever requested players to use dice.

As I understand it, dice are considered a sign/form of gambling in Japan. PCL doesn't want Pokemon being associated with gambling in any way. So that really only leaves flipping a coin left as a randomizer, other than things like RPS, which we see they use as well.

 

[Cyrus]

Well, if they don't want Pokemon being associated with gambling, then they've failed at it miserably (given our use of dice + the video game creators' constant spamming of gambling centers in RBY=DPP lol).

Still, I shouldn't let that comment about dice take over the concern about PCL's intent in the mechanics of flips and RPS. If it turns out that, sure, they couldn't car less about randomization, then that puts this discussion in a completely different light.