Filtering For Player Type

[pat3392]

Is there a way to do ^? So let's say I want to know how often a TAG folds his steal to a 3bet.

[kraada]

Not at this point in time, but there are plans to have this kind of thing eventually.

[pat3392]

Any idea how this will likely be implemented? And very approximately, in much time do you think it'd take for this to be implemented? > 1 year/< 1 year?

[kraada]

Well the first step is reintroducing tags for players - that should happen over the next few months ideally. After that we'll look into what we can do filter-wise. At first I'll definitely be able to work up something for you for the My Reports area (once the tagging is implemented), as for something more automated it may take quite a bit longer depending on how things end up working, just because filters like that can be extremely inefficient.

[pat3392]

Okay awesome, this is defintely a big plus for PT4 for me, as I'm suprised HM2 have not implemted said feature, nor even thought about it before and have no idea how to do so.

[pt4pt4pt4]

This is along the same lines of what i was referring too while back. i would like to filter all hands that i was involved where the villian or one of the villians had a VPIP range of 10-20 percent Lifetime (for example). Is this possible without player tagging feature?

[kraada]

You can do it with manual My Reports filtering, but filters like this are very slow. To filter for hands played where a player with VPIP between X and Y put money in the put use the following Filter Expression:

player.id_player in (SELECT p.id_player from player p, cash_hand_player_statistics chps where chps.id_player = p.id_player GROUP BY p.id_player HAVING (((( (CASE WHEN ( (sum((case when(chps.id_hand > 0) then 1 else 0 end)))) <> 0 THEN (((sum((case when(chps.flg_vpip) then 1 else 0 end))) * 1.0 )/( (sum((case when(chps.id_hand > 0) then 1 else 0 end))))) ELSE 0 END) ) * 100) BETWEEN X AND Y)))

(Note: this uses the PT3 definition of VPIP - I don't have one with the PT4 definition handy but I can work it up for you if you'd like - this will probably be pretty close regardless.)

[pat3392]

How slow is very slow? Would there be anyway to apply this to a database of 25 million hands +?

[pt4pt4pt4]

You can do it with manual My Reports filtering, but filters like this are very slow. To filter for hands played where a player with VPIP between X and Y put money in the put use the following Filter Expression:

player.id_player in (SELECT p.id_player from player p, cash_hand_player_statistics chps where chps.id_player = p.id_player GROUP BY p.id_player HAVING (((( (CASE WHEN ( (sum((case when(chps.id_hand > 0) then 1 else 0 end)))) <> 0 THEN (((sum((case when(chps.flg_vpip) then 1 else 0 end))) * 1.0 )/( (sum((case when(chps.id_hand > 0) then 1 else 0 end))))) ELSE 0 END) ) * 100) BETWEEN X AND Y)))

(Note: this uses the PT3 definition of VPIP - I don't have one with the PT4 definition handy but I can work it up for you if you'd like - this will probably be pretty close regardless.)

In order to get this to work with tourney i change the one cash to tourney statement and you are saying that i need to change a VPIP expression?

[kraada]

Change cash to tourney and it'll work for tourney. This is just the same as the VPIP Legacy stat (because that's what I have - I don't have the case statement for cnt_walks handy and it'd require joining to the lookup_actions table so I can't do it off the top of my head). It'll be pretty close to real VPIP.

On a 25 million hand database, expect queries to take an hour and upwards using this kind of thing. Though that's based on my experience with PT3 and a 14M hand database on my machine so YMMV.