Effective Stack vs Actual Stack

[carl91]

What's the difference between the eff. stack filter and actual stack filter exactly?

Say, I ran a filter where I am active player 8 (Big Blind) and a filter for position of first raiser 0 (button). I then want to look at pre-flop frequencies 40-60BB deep (e.g. fold to steal, 3-bet pre-flop, call pf 2bet). I apply the "effective" stack size filter 40-60BB and compare with the "actual" stack size filter 40-60BB. Often the results are kinda similar but just slightly different. What's the difference here exactly?

[Flag_Hippo]

Effective stack is the amount you could potentially lose as of your first action in the hand so it will be influenced by the stack size of players who have acted before you (and those that are still to act after). For example:

BTN (20BB)
SB (30BB)
BB (50BB)

If BTN raises and SB folds then the effective stack of BB is 20BB. If SB VPIPs in-between then the effective stack of BB would be 30BB.

[carl91]

Ok, still a little confused but think I get what you're saying.

I've attached 2 screenshots, one with eff. stack size filter and the other with actual stack size filter. All other filters are the same for both.

Effective Stack Size Filter:

https://imgur.com/wU2IPPW

Actual Stack Size Filter:

https://imgur.com/s7T5zsC

You can see there's a marginal difference in call 2bet freq + 3bet freq though ...

[Flag_Hippo]

With those filters no player other than the BTN can be involved in the hand so only the BTN can influence the effective stack of the BB in those hands. With the actual stack filter you will see all hands where the BB had between 40-60BB (BTN could have any stack between 1BB and ∞). With the effective stack filter there would be a slightly different sample as hands where the BTN had less than 40BB would be filtered out and hands where the BB had more than 60BB with the BTN having 40-60BB would now be included.

[carl91]

Ok got it, thanks.

On a side note, what's the reason the call pf 2bet stat and call any pf raise stat would give a different (albeit only marginal) frequency with these filters on??

[Flag_Hippo]

Those statistics will not have the same times/opportunities as long as there are hands where BB 3Bets and BTN 4Bets. 'Call Any PF Raise' is defined as the 'Percentage of the time that a player called any raise preflop, given that he had a chance to do so' so, for example, if there was one hand where the BB 3Bet and called a BTN 4Bet then:

'Call PF 2Bet' would be 0% (0/1)
'Call Any PF Raise' would be 50% (1/2)

[carl91]

Ah right, that makes sense. Thanks.