Okay this might need to be in the noob section so bear with me here.

From another thread I know WHAT standard deviation is, as far as a definition goes.

However what do the numbers in the stats in PT actually mean? If I'm looking at my SD/100 and SD/Hour what exactly am I looking for, and how would I apply it when looking at my other stats?

For example my SD/100 at 36,600 hands is 15.84...when I look at this I should be saying to myself.....what?

What about SD/100...is/are there any comparisons I should be making between the 2 numbers?

Is there a point when you should look at those numbers and say "whoa this is crazy" whan comparing the 2?

I'm not a math whiz by any stretch of the imagination so if anyone can lay it on me in simple terms I'd appreciate it.

__________________
If you're looking to borrow money or have someone give you money in order to play poker online please don't contact me... Click Here for help , and explain your situation.

Standard deviation is the average difference from the average...
So if you make $10/hr playin poker and your standard deviation is $3, half the time you'll make between $7 and $13/hr, the other half you'll be outside that range.

Fish, the first standard deviation (and correct me if I'm totally wrong, but in my research this is what I have learned, since a few years ago, I had the identical questions as Pok7's) is around 80-90% I believe, not 50%.

Somebody is goign to have to link a good statistics page so we can all learn.

I can say this though, The 300bb rule for limit bankrolls is based on a 3bb/100 WR with a SD of 20/100 (this leads me to believe the 90% is the first SD, since very rarely do we play outside of +23 to -17 WR range). So if you have a WR around 3/100 a lower SD below 20 is good, while higher is 'bad'. So if you have a WR of 4/100 and a SD of 15/100, you could be playing with a bakroll of 150-200 bb's (somebody can crunch the numbers). Personally, my 3.5/100 and 17 SD allows me to play limit with a bank of 180 bb's, but I choose not too, I have gone broke too many times.

SD by defintion is a measurement of short term luck. If you have them, Malmuth goes into some analysis of SD in his Poker Essays books.

__________________

"Standard Deviation:
The standard deviation is the square root of the average squared deviation from the mean. Ex - The standard deviation of the Year 2 Internal Medicine scores was 6.9; therefore, we can conclude that 68% of the class fell within plus or minus 6.9 points of the mean."

I'm kinda ashamed I got this one wrong, I'm taking a Probability and statisitcal inference class right now. I did some reading my in my book and the exact % of your sample within the range [mean +/- standard deviation] will actually vary somewhat, but on your basic bell curve seems to be 68%

i knew i shoulda paid attention during math class

__________________
If there wern't luck involed i guess id win everyone

If I'm not mistaken, I believe that 96% of the time, X falls within 2 standard deviations of the mean, and as FoD said, X falls within 1 SD 68% of the time.

__________________
"There is a good chance I gave you a very bad description of something that doesn't work."

Lets say that you're average earnings/hour is $20. After 5 hours you've made $20 in 1, $22 in 2, $25 in 3, $17 in 4 and $16 in 5.

So 1 SD would = 3.3 (using calculations, i didnt jst make that up)

That means that if your SD in 1 hour is 1 then you can vary from $3.3 +/- of the average about 68%% of the time.

If you're SD = 2 then you vary from $6.6 +/- of the average 96% of the time and:

If you're SD = 3 then you will vary from $9.9 +/- the average 99.7% of the time.

It basically tells you the likelyhood of variance from your average score.

So in this example, his variance would be +/- 15.84 BBs, 68% of the time?

what is the SD/100 i don't get that.. is that you're standard deviation for money made per 100 hands? if it is then mayb it is every 100 hands he makes +/- $15.84 68% of the time which is the variance. That would make sense.



ask me about graph if confused

This seems to imply that his variance is the same in the negative direction as in the positive. I've found that I generally win larger hands(albeit less frequently) than I lose.

So(making these numbers up) if I tend to vary between -$5 and +$20 per 100 hands, 68% of the time, then what would be my standard deviation?