PokerTracker

I was playing about trying to make a custom stat for tracker. I wanted to get the percent of times the player has open shoved the button pre-flop.

I have made 2 stats..

cnt_times_player_could_have_shoved_button

Code: Select all

sum( if[ tourney_hand_player_statistics.flg_p_open_opp = '1' AND position = '0', 1, 0] )

cnt_times_player_could_have_shoved_button

Code: Select all

sum( if[ amt_chips - amt_bet_p = '0' AND position = '0', 1, 0] )

Then I made the stat
Open Shoves Button

Code: Select all

(cnt_times_player_open_shove_button / cnt_times_player_could_have_shoved_button) *100

When I try to use the stat I get an error..

Code: Select all

Unable to execute query......... ERROR: coulmn 'tourney_hand_sumary.id_hand" must appear in the GROUP BY clause or be used in an aggregate function.

Anyone have any ideas why this is happening?

Cheers

You need to use the full table reference in order for your columns to work right, so for your first column you want:

sum( if[ tourney_hand_player_statistics.flg_p_open_opp AND tourney_hand_player_statistics.position = 0, 1, 0] )

I presume your second column was supposed to be times the player did open shove the button? This definition will work better for you:

sum(if[tourney_hand_player_statistics.flg_p_open_opp and tourney_hand_player_statistics.position = 0 and lookup_actions_p.action = 'R' and tourney_hand_player_statistics.amt_p_raise_made = tourney_hand_player_statistics.amt_p_effective_stack, 1, 0])

This says you had a chance to open on the button and you made exactly one action (raise) and the size of your raise was equal to your effective stack size (aka a shove).

kraada wrote:You need to use the full table reference in order for your columns to work right, so for your first column you want:

sum( if[ tourney_hand_player_statistics.flg_p_open_opp AND tourney_hand_player_statistics.position = 0, 1, 0] )

I presume your second column was supposed to be times the player did open shove the button? This definition will work better for you:

sum(if[tourney_hand_player_statistics.flg_p_open_opp and tourney_hand_player_statistics.position = 0 and lookup_actions_p.action = 'R' and tourney_hand_player_statistics.amt_p_raise_made = tourney_hand_player_statistics.amt_p_effective_stack, 1, 0])

This says you had a chance to open on the button and you made exactly one action (raise) and the size of your raise was equal to your effective stack size (aka a shove).

Thanks Kraada. That make sense. I come from c/c++ so haven't done much in SQL.

Unfortunately I'm still getting the same error. Do I have to use the column 'tourney_hand_summary.id_hand' somewhere in my stat as it says it must appear in the GROUP BY clause or an aggregate function?

I get the same error whether or not I use the stat function posted above, or just output either of the columns.

Cheers

What other stats do you have in this report? The above stats don't have any data from tourney_hand_summary in them . . .

kraada wrote:What other stats do you have in this report? The above stats don't have any data from tourney_hand_summary in them . . .

Sorry, I had done it as a hand stat not a player stat. All works perfectly. Thanks kraada