Quote:
Originally Posted by mojon
Thanks for your detailed reply - much appreciated.
I'm sorry to go on about the results so much but I can't use the app if I dont have full faith in the results :
If you have A9o, the chances of missing the flop completely is
44/50 x 43/49 x 42/48 = .68
Yours shows .538
And shouldnt the high card hit/make figure differ - one would be lower when the flop pairs? What am I missing ?
By all means pls. do keep going on re. the results - imo just a single knowledgable, interested customer is worth more than a dozen possibly disinterested testers I can ever hire : ). But a good tester is worth his weight in gold.
I've added pics below for your A9o example, the "hit hand" stats and the "make hand" stats.
I don't see a problem here and think this is just a matter of how to interpret the stats. Yes the app shows 54% but this is for "High Card". This doesn't mean your "the chances of missing the flop completely" because if the flop is 223, then A9o has missed this flop completely but A9o has made OnePair here, not HighCard, so this won't be counted in the HighCard stat.
To get to your "the chances of missing the flop completely" using the app's set of stats, we would need to take the 54% and add to it the cases where the flop *alone* makes a pair, or 3OfAKind, etc. From the pics, A9o makes OnePair 40.3% of the time, but hits OnePair 28.9% of the time, so subtracting these, the flop alone makes a pair 11.4% of the time. Adding 54% and 11.4% gives 65.4%, which is taking us closer to your 68% (I haven't added in the diff for 3OfAKind).
We still won't get all the way to 68% by adding in these diffs because of the meaning of the "hit OnePair" stat, e.g. on a flop of 922, A9o gets credit for hitting OnePair but has made TwoPair, so there's some overlapping of stats going on here. But I think this clears up the main 68% vs. 54% discrepancy.
Re. "And shouldnt the high card hit/make figure differ" -
The rationale/meaning of the "hit hand" feature is that a stat's "hit hand" prob. would always be lower than its "make hand" prob. because hitting a particular hand is a stronger condition than making the hand. If the board pairs, we're automatically out of the HighCard stat and are into OnePair or better, so the HighCard stat wouldn't be affected. The OnePair stat is affected however, and we can use the "hit" vs. "make" difference as above. So I prefer not changing the HighCard stat and staying with the current 4 bold-blue stats in the pics as the "hit hand" stats.
Do let me know of course if anything doesn't sound like the best way to go. I'm trying to balance adding in new good but maybe marginally-useful features vs. reaching a stability point and moving on to new work. Thanks.
A9o, Deal-To-Flop, Hit Hand Stats:
A9o, Deal-To-Flop, Make Hand Stats:
Last edited by rj999; 06-16-2011 at 02:59 PM.