Do we have an idea of how much the "adjusted EV" of HM decrease the standard deviation ?

Can you for example compute your standard deviation and your "adjusted" standard deviation ?
I think it's not very hard. You have to export your HM stats to excel and then use a formula (I don't remember sorry).

This way, we can now have smaller confidence intervals when you make some estimation about your true winrate.
Let's say for example, you want to know the probability that you're a winning player knowing that you ran at 1BB/100 over 100k hands. If we have a lower standard deviation, the probability will be bigger (how much ? I don't know) and you understand why it oculd be useful to know the adjusted SD.

I try to compute the standard deviation using excel.

I exported 3500 hands and I used the =ECARTYPE formula to the column containing the profit of each hand.
It gives me 8.5. I play 100NL 6max so 8.5$ is 4.2BB but I know that we talk about standard deviation as BB/100 and not BB.

So, I should compute the standard deviation of my winrate right ?

But how exactly ?

Because if I compute my winrate as the mean of the profit of the n first hand, then my winrate will converge (not exactly but the more hands I have, the more slowly he will move after a certain number of hands) and then my standard deviation will converge to 0 right ?
So, I don't understand something there.

I'm supposed to find something like 40-50BB/100 for my standard deviation.

Your winrate converging shouldn't cause the standard deviation to converge? It sounds like you are trying to use the deviation of the new mean from the previous mean or something similar (that's the only way I can see why your standard deviation would converge to zero).

To work out your standard deviation is pretty easy on a "per session" or "per hand" basis, but I'm not entirely sure how PT does it in terms of BB/100 (the book "Gambling Theory and Other Topics" will tell you IIRC). My guess is that is does something like this:

A) Find your winrate from all your data (ie: the mean).
B) Break up your data into 100 sample "blocks" somehow and find the winrate for each "block" (perhaps with resampling - this is the bit I'm unclear on).
C) Work out the sum of squared deviations of each block's winrate from the mean winrate calculated in (A).
D) Normalize by dividing by the number of blocks and then take the square root.

This is only a guess though and I'm not really 100% sure if it's correct - your best bet is to ask this in the Probability forum (or else buy/borrow a copy of "Gambling Theory and Other Topics").

Juk

PS: I'd be quite interested in seeing any results you get on this. Knowing just how much the luck adjusted version reduces the variance would be pretty interesting (plus if you search the 2+2 archives you'll find some good posts about how to use the info to generate confidence intervals on true winrate estimate and thus you should be able to work out how how many non-luck adjusted hands it takes to get the same level of confidence as the luck adjusted version).

Just to reformulate my question : How HM does compute the "standard deviation" stat ? And how can I compute that using excel and a column with my profit of each hands in my database.

Juk, You choose to take 100 hands by block but you're not too sure it's the "convention" ? Because, the SD you'll find at the end, depend directly from the size of each blocks. The bigger the blocks, the lower the SD since the winrate will be closer than the true winrate.

Unfortunately, I don't have the excel skillz do that what you ask :?

And I'm sure it will be very interesting to know how much the adjusted EV decrease the variance and then improve the confidence's intervals.

Yep, the SD will be lower as you increase the block size, but I'm not even 100% sure if my method is correct (I just thought about how I would go about measuring the SD of the data if I was asked to and could be completely off).

Yep, I don't really know how you would go about it in Excel either. I'm guessing that there must be a better way to do it with some maths that doesn't require breaking it up into blocks (I'd definitely re-post this in the Probability forum as they are the peeps who will know if there is).

Juk

anyone ?