Beat: I forget what a set looks like - setometer

[Das Budrick]

where on the curve is the norm?

[tubasteve]

i have no idea what the graph actually means. anyone?? whats the pecentage, when does it turn blue, how is the damn graph generated???

[Jiganti]

Give me more programs like this. Srsly. Now.

[blackize]

Since the 30th

[ocdscale]

The more blue it is, the better you're running.
Having it exactly 50% blue 50% red is what you would expect on average.

[creative]

really? 50/50

im not really thinking right now but that doesnt make too much sense to me

[TurtlePiss]

Yes 50/50. Either you flop a set or you don't.

[pa3lsvt]

[ QUOTE ]
really? 50/50

im not really thinking right now but that doesnt make too much sense to me

[/ QUOTE ]

The curve is the probability curve of flopping a set. The std dev is probably calculated using your sample size.

If you are 50/50 on the graph, you have flopped exactly as many sets as you should have (i.e. results = mean probability).

[LionelHutz00]

[ QUOTE ]
i have no idea what the graph actually means. anyone?? whats the pecentage, when does it turn blue, how is the damn graph generated???

[/ QUOTE ]
the x axis is the number of sets. the y axis is the probability that you flop a set that number of times.

Look at blackize's graph. given the number of hands that he's seen the flop with a pocket pair, he can expect to flop exactly 5 sets 6% of the time. He can expect to flop 10 sets ~13% of the time. and so forth.

The line represents the number of sets you've flopped in real life.

So if the blue area > the red area, you run good.

[bfc]

[ QUOTE ]

I know it counts quads. Not sure about boats.

[/ QUOTE ]

Working backwards from the example given on the home page it appears it counts sets, boats, and quads.

At the risk of being flamed for posting theory in BBV, this is what I believe it does.

It defines "flopping a set" as "the flop contains at least one of the two remaining Xs when you hold XX".

The probability of this occurring in any given hand is 1-(48/50*47/49*46/48)=0.11755102

In the author's example he saw 1042 flops with pocket pairs so expected to flop 1042*0.11755102=122.5 "sets or better". This is the expected number of flopped sets. In actuality he only flopped 101 sets (the variance we know and love).

As others have said, the curve is the probability distribution of number of flopped sets (the expected value (i.e., mean value) is the peak). If we assume a binomial distribution then the standard deviation is sqrt(n*p*(1-p))=10.4.

Approximately 95% of the distribution should be within 2 standard deviations of the mean (assuming a normal approximation to the binomial). That is, between 102 and 143. If your actual number of flopped sets is in the tails (less than 102 or greater than 143) then you have been running either extremely hot or extremely cold (at least when it comes to flopping sets).