How does PT calculate standard deviation/100 hands?

[corpcd]

Hello,

I've been searching the forum, and I found nothing.

I've been testing different ways of calculating the standard deviation for the gains, but I cant get to reproduce the exact number that PT2 says it is the standard deviation per 100 hands.

Do you know how PT calculate it?

Any formulae for clarification will be appreciate.
Thanks in advance,
corp

[ptrack pat]

I used the calculations detailed on 2+2 by Mason Malmuth but the link no longer works so I am attaching screen prints of the pages I printed out a long time ago.

In the case of SD/100 hands - the "result" on page 2 below would be the BB/100 hands for that session and the "time" would be the number of hands/100 instead of the actual time spent playing.

http://s217900936.onlinehome.us/mmessay_pg1.jpg
http://s217900936.onlinehome.us/mmessay_pg2.jpg
http://s217900936.onlinehome.us/mmessay_pg3.jpg

[corpcd]

I used the calculations detailed on 2+2 by Mason Malmuth but the link no longer works so I am attaching screen prints of the pages I printed out a long time ago.

There must be another "trick" in the calculations.

¿Do you consider ALL sessions? Or only those sessions with a given number of hands.

I have a session with 2 hands and -100BB (the second hand i suffered a bad beat, losing my stack and i did exit), and when using the BB/100 hands it has X=-100/2*100= - 5000BB, and this number make the result wrong.

Probably you use only sessions with a given number of hands.

Also, this is what i'm using:

X=100*BB_per_session/number_of_hands_in_the_session
T=number_of_hands_in_the_session/100.

corp

[ptrack pat]

I literally have not looked at this section of SD code for 4 years. It took me an hour of searching the old forum to even find the links to the pictures I gave you above. So I don't know and I'd have to go back and dig through the code to see what it's doing and I don't have the time for that now. I'll try to get to it sometime later.

[corpcd]

I literally have not looked at this section of SD code for 4 years. It took me an hour of searching the old forum to even find the links to the pictures I gave you above. So I don't know and I'd have to go back and dig through the code to see what it's doing and I don't have the time for that now. I'll try to get to it sometime later.

Ok, dont worry. It was no my intention you to waste so much time on this.

If PT authors plan to implement it on PT3, I will ask them for formulation. What a pity that 2+2 posts were deleted.

corp

[corpcd]

Got it!!!!

This is the way PT2 calculates standard deviation:

Xi=net_won_in_the_ith_session.
Ti=number_of_hands_in_the_ith_session/100.

And then, it uses the formulae of the article "ptrack pat" posted above, using sessions.

Thanks for information!
corp