luke123abc1 wrote:cash_hand_summary.str_aggressors_p = '830
%'
luke123abc1 wrote:However, I am still seeing 0 as result.
You inserted a '%' into the expression I suggested which isn't going to work when using '=' since '%' isn't a valid character for the aggressors string. You can only use % when pattern matching with LIKE or SIMILAR TO - for more information on pattern matching in PostgreSQL see
this guide.
luke123abc1 wrote:I have fixed it. Somehow the statistic was not loading properly and a restart of PT4 has it loaded. I had to add the line for playerposition = 3 to fix the stat FYI.
I don't follow this. You were already using
cash_hand_player_statistics.position = 3 in the expressions you posted and you wouldn't have been getting any results for the stat due to using
cash_hand_summary.str_aggressors_p = '830%' in the actions column.
luke123abc1 wrote:Now, it is show 61%. Is this normal for microstakes populations? UTG is calling 61% of BTN 3 Bet. Seems crazy to me.
luke123abc1 wrote:So I did this for every spot IE UTG vs BTN, UTG vs BB, UTG vs SB etc. I posted examples of my columns below. Then, I calculate the statistic Call/Oppurtunity for each spot. For every spot, the value of this statistic is greater than the value of 2Bet PF & Call 3Bet for UTG. Is there a mistake somewhere in my columns or should this be impossible?
It's not impossible. It might be unlikely on a large enough sample but I don't have your population data although when UTG is calling vs the blinds they will have postion postflop so those might be higher. There is also an adjustment I've made (see below) which would lower the call percentages you are seeing. If you think you are still getting unusual results then make sure you have rebuilt your custom database cache via 'Database -> Database Management -> Rebuild Cache -> Custom Cache Rebuild'. It's also good practice to rebuild the custom cache when you've finished a batch of changes so that all of the new/changed custom columns can use the
database cache.
luke123abc1 wrote:cnt_utg_vs_bb_3bet_opp
sum(if[cash_hand_player_statistics.position=3 AND
(cash_hand_summary.str_aggressors_p LIKE '838'), 1, 0])
While it doesn't affect the function of the column the name isn't accurate since this is not a 3bet opportunity for UTG.
cnt_utg_vs_bb_3bet_def_opp would make more sense.
luke123abc1 wrote:cnt_utg_vs_btn_3bet_call
sum(if[cash_hand_player_statistics.position=3 and
cash_hand_player_statistics.enum_p_3bet_action = 'C' AND
(cash_hand_summary.str_aggressors_p = '831'), 1, 0])
This isn't correct as you are testing the aggressors string for a CO 3bet.
luke123abc1 wrote:Statistic looks like:
UTG call BB 3 Bet % after RFI
Another thing you might want to consider is whether you want to include all calls or only certain calls. For example some 3bets are squeezes so you may want to exclude those. Also if there are players to act between the 3bettor and UTG then they can call that 3bet which would give UTG better pot odds to call. If you want to exclude all hands where other players got involved then you can use something like this for UTG vs BTN:
cnt_utg_vs_bb_3bet_call- Code: Select all
sum(if[cash_hand_player_statistics.position=3 and
cash_hand_player_statistics.enum_p_3bet_action = 'C' and
cash_hand_summary.str_aggressors_p = '830' and cash_hand_summary.str_actors_p = '303', 1, 0])
cnt_utg_vs_bb_3bet_def_opp- Code: Select all
sum(if[cash_hand_player_statistics.position=3 and
cash_hand_summary.str_aggressors_p SIMILAR TO '(830|8303%)' and cash_hand_summary.str_actors_p SIMILAR TO '(30|303%)', 1, 0])
I've also changed the opportunities column since when UTG has an opportunity to call a 3bet it's also an opportunity to 4bet (if the 3bet wasn't an all-in) so the opportunites column needs to count the hands where this happened otherwise it would inflate UTGs call percentage.