A ruling question on prizes.

[FireFighter095ReBorn]

I watched many battles while i waited for mines to start in milford when i was at the states and i remember that there were sometimes when prizes were forgooten to be put out but i dont remember seeing the prize swap peanilty used in the first offense. Here is what i think

That player b's penailty should of been a caution and the caution should of been noted on the back of the match slip so they will have it recorded unless it happened again and on the 2nd offense they should of been penalized

In player m's case it is player b's responsibility to let the judge know so the judge can watch what is going on. Max 3 minutes should be allowed per turn to do what you need to do. I think if seen player m should of been penalized and the same action as in player b's case should of been done

I hope both players do well in thier next tournmant and i hope many professor's knoledge can help them.

 

[SteveP]

Taking cards from the top of the deck will thus change nothing, only searching or putting known cards back in your deck can change things.

I disagree. Probabilities change (as I pointed out). Think about it. You've got only 1 of a particular card in your whole deck. If you forget to lay out your prizes then:

- there's a 0% chance that you'll NEVER draw that card.
- there's about a 10% chance that you WILL draw that card in your first 6 draws.

Now, suppose you DO lay out your prizes. Then:

- there's about a 10% chance that you'll NEVER draw that card (if it's in your prizes).
- there's about a 15% chance that you WILL draw that card in your first 6 draws (if it's NOT in your prizes).

Randomness is ALL ABOUT probabilities.

 

[PokePop]

mozartrules is correct about the probabilities. That is exactly how I've heard the WotC MTs describe.

 

[SteveP]

For those who've studied statistics, from what I remember, probabilities are based on:

1. sample size
2. size of the sampling space (from whence the sample is drawn)
3. contents of the sampling space
4. randomess (or distribution) of the items in the sampling space

So, in our example here, the 2nd characteristic changes (and possibly the 3rd item).

Listen, I will accept the argument that card drawing from the top of the deck (without search/shuffling) is random. Likewise, its a random event to place prizes.

HOWEVER, the probabilities in that randomness DO change!

Are there any listeners out there who are Statisticians by trade? I'd love to hear to take on this.

 

[mozartrules]

Let us make a simple example where we have a single BEX in out deck and it wasn't in our opening hand. The chances of it being in the prizes is of course 6/53.

You now draw a card and this card is either BEX (with a probability of 1/53) or some other card. The total chance of BEX being in your prizes can now be calculated as 1/53 (we drew BEX) * 0 (it cannot be in the prizes) + 52/53 (we drew something else) * 6/52 (BEX is in the prizes of the unknown 52 cards). This comes back to the same 6/53. You can make more complex examples, but I am pretty sure they all reduce to the same probability.

I studied this kind of situation a lot in my younger days when I played in the national junior team in Contract Bridge (Denmark). The problem is separating the original probabilities (which in this case hasn't changed) from the probabilities that you calculate as you gain additional knowledge. It is very easy to get confused - it has happened to me numerous times - but I think I am doing it right here.

 

[SteveP]

So, we include the prizes in the sampling space, even though the process of drawing cards is completely different than drawing prizes?

I really think that the sampling space is different in these two situations (laying out the prizes versus NOT laying out the prizes). If you CAN'T take a sample from the prizes, should they really be included in the sampling space?

Oh well...

 

[mozartrules]

If you CAN'T take a sample from the prizes, should they really be included in the sampling space?

Yes, as long as you have no influence (meaning knowing the cards) on which cards go where. Imagine that you have a deck of ordinary playing cards, well shuffled.

Your odds of drawing the ace of spades is 1/52 (formulas are much simpler when looking for a single cards, but the principles are true for any number of cards).

Now let someone take 6 cards (or any other number lower than 52) at random and draw a card. Your odds are still 1/52 of drawing the ace of spades. The higher odds of you taking a specific card out of the remaining cards is exactly offset by the odds of the card not being in there at all.

Say someone takes one card. There is 1/52 chance of it being AoS in which case you have zero change of getting it, there is 51/52 chance of it being something else and you then have 1/51 chance of drawing AoS from the other cards. Events like that are calculated as 1/52 * 0 + 51/52 * 1/52 = 1/52 and you get the same result for all number of removed cards

N/52 * 0 + (52 - N)/52 * 1/(52 - N) is the same as 1/52 for all N in [0; 51].

If that someone looks at the N cards removed he will of course know more about your odds, it will be either 0 or 1/(52 - N) depending on whether he is looking at the AoS or not. It is tempting to think that that knowledge changes the original odds, but it doesn't.