Call vs open from different positions

[Naeco]

Hello,

I already have a custom stat for: calling against minraise which is the following:

(cnt_p_mr_call2 / cnt_p_mr_call_opp2) * 100

cnt_p_mr_call2 expression: sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 1 AND lookup_actions_p.action LIKE 'C', 1, 0])
cnt_p_mr_call_opp2 expression: sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 1 AND lookup_actions_p.action LIKE '_', 1, 0])

This stat seems to work fine. However I'd like to split it in two stats: one for calling against BTN and one for calling against SB.
I tried adding this in both expressions

tourney_hand_summary.str_actors_p LIKE '9%' for SB and tourney_hand_summary.str_actors_p LIKE '0%' for BB, however it says that this is not valid SQL.

Anyone could tell me what I did wrong?

Thanks

[Flag_Hippo]

I already have a custom stat for: calling against minraise which is the following:

(cnt_p_mr_call2 / cnt_p_mr_call_opp2) * 100

cnt_p_mr_call2 expression: sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 1 AND lookup_actions_p.action LIKE 'C', 1, 0])
cnt_p_mr_call_opp2 expression: sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 1 AND lookup_actions_p.action LIKE '_', 1, 0])

This stat seems to work fine.

Anything using facing is the amount a player needs to call so that isn't going to work correctly except for the BB. If, for example, the CO minraises then the BTN would have to call 2bb, the SB 1.5bb and the BB would have to call 1bb. To calculate correctly for all positions you will need to add the size of any blind posted by the player to the amount they are facing. If you want this specific to 2bets then I'd recommend using tourney_hand_player_statistics.amt_p_2bet_facing instead (and using equal to x is clearer than using between x and x):

Code: Select all

(tourney_hand_player_statistics.amt_p_2bet_facing + tourney_hand_player_statistics.amt_blind) / tourney_blinds.amt_bb)) = 2

Also lookup_actions_p.action LIKE '_' isn't suitable for the opportunites column and it isn't required. You only need to know if they faced the raise for the opportunities column but with lookup_actions_p.action LIKE '_' you are only going to count hands where they took exactly one preflop action (R, C or F) but if they raise (or call and another player squeezes) then the player will be taking subsequent preflop actions and those hands wouldn't be counted for the initial opportunity.

However I'd like to split it in two stats: one for calling against BTN and one for calling against SB.
I tried adding this in both expressions

tourney_hand_summary.str_actors_p LIKE '9%' for SB and tourney_hand_summary.str_actors_p LIKE '0%' for BB, however it says that this is not valid SQL.

Anyone could tell me what I did wrong?

We would need to see exactly how you are adding it into the original expression to see why it would not be validating. If I add tourney_hand_summary.str_actors_p LIKE '9%' to your original column expression it validates:

Code: Select all

sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 1 AND lookup_actions_p.action LIKE 'C' and tourney_hand_summary.str_actors_p LIKE '9%', 1, 0])

however using tourney_hand_summary.str_actors_p LIKE '9%' specifies the first actor and not the first raiser so if the BTN limps and the SB minraises your statistic is not going to work correctly.

[Naeco]

Thanks for the answer.

We would need to see exactly how you are adding it into the original expression to see why it would not be validating. If I add tourney_hand_summary.str_actors_p LIKE '9%' to your original column expression it validates:

Code: Select all

sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 1 AND lookup_actions_p.action LIKE 'C' and tourney_hand_summary.str_actors_p LIKE '9%', 1, 0])

however using tourney_hand_summary.str_actors_p LIKE '9%' specifies the first actor and not the first raiser so if the BTN limps and the SB minraises your statistic is not going to work correctly.

Would this expression work instead if I want to specify that BTN is the first raiser ?

tourney_hand_summary.str_aggressors_p LIKE '80%' AND tourney_hand_summary.str_actors_p LIKE '0%'

Also could you tell me if new stat is correct? I want to see how often BB defends against an open raise from SB from 2 to 4bb:

cnt_p_mr_call: sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 3 AND lookup_actions_p.action LIKE 'C' and tourney_hand_summary.str_actors_p LIKE '9%', 1, 0])
cnt_p_mr_call_opp: sum(if[(tourney_hand_player_statistics.amt_p_raise_facing / tourney_blinds.amt_bb) BETWEEN 1 AND 3 and tourney_hand_summary.str_actors_p LIKE '9%', 1, 0])

[WhiteRider]

Would this expression work instead if I want to specify that BTN is the first raiser?
tourney_hand_summary.str_aggressors_p LIKE '80%' AND tourney_hand_summary.str_actors_p LIKE '0%'

Yes. That excludes anyone limping before the first raise as well. To just specify that the button made the first raise without any other restrictions you only need:
tourney_hand_summary.str_aggressors_p LIKE '80%'

Also could you tell me if new stat is correct? I want to see how often BB defends against an open raise from SB from 2 to 4bb:

Yes, that looks good I think.

[Naeco]

Thanks I have everything I need now.