#11
|
|||
|
|||
Re: Aces vs. Joker
[ QUOTE ]
40 hands isn't exactly note-worthy. [/ QUOTE ] Yes it is, if the result is 31-9. It is common to tell people they need more data just to get rid of them, but it's not correct here. The one-tailed significance level is 0.034%, i.e., if it's a toss-up, Joker-2 would win that much or more about one time in 3000. This is a highly significant result. |
#12
|
|||
|
|||
Re: Aces vs. Joker
[ QUOTE ]
just dealt out 40 hands real quick. Joker hand won 31-9. [/ QUOTE ] To add to this, I dealt out 36 hands, 18 suited and 18 unsuited. Joker-2 won 27-9. These were the actual hands made, with the hands won by AA starred: flush vs. pair quads vs. 2 pair 3oak vs. 2 pair straight vs. 3oak straight vs. pair *pair vs. pair flush vs. 2 pair *3oak vs. full house 3oak vs. two pair full house vs. 2 pair flush vs. straight 3oak vs. 2 pair 3oak vs. pair flush vs. pair *pair vs. pair 3oak vs. 2 pair 3oak vs. 2 pair *straight vs. quads 3oak vs. pair 3oak vs. 2 pair straight vs. pair full house vs. 2 pair *pair vs. pair straight vs. pair 3oak vs. 2 pair *3oak vs. full house *3oak vs. 3oak 3oak vs. 2 pair 3oak vs. pair quads vs. 2 pair *pair vs. pair 3oak vs. 2 pair full house vs. 2 pair *pair vs. 3oak full house vs. 2 pair straight vs. pair I was surprised by how frequently Joker-2 made a straight or a flush, 10 times out of 36. Combining the two sets of data gives Joker-2 58 wins out of 76, 76.3%. A 95% confidence interval is 65.1%-85.4%. From 76 hands, we don't have a very tight estimate on how frequently Joker-2 beats AA, but we can conclude that it wins much more than 50%. I'm not sure I want to test this myself, but perhaps just a joker would beat AA. |
#13
|
|||
|
|||
Re: Aces vs. Joker
</font><blockquote><font class="small">In risposta di:</font><hr />
18 suited and 18 unsuited [/ QUOTE ] was anyone else confused by this? if you have a joker then your hand is always suited. |
#14
|
|||
|
|||
Re: Aces vs. Joker
I took "suited" to mean the 2 was the same suit as one of the aces.
|
|
|