Custom postflop stats, new wheel invention needed or not?

[Sopuli]

Hi,

Iam trying to create a postflop pop-up and have questions regarding some stats for 2 players on the flop.
I don't want to invent the wheel twice, so perhaps some of those are already existing.

Case A) he has position

A1) Difference between "% float flop" and "% flop probe bet"? Both are flop bets in position after pf aggressor does NOT cbet, but:
% float flop: is counting only situation where one preflop bet is called (no pf 3bet situations)
% flop probe bet: is counting all situations where any preflop bet is called (also pf 3bet situations)
Is that correct?

A2) If so, then "%float turn" is not taking into account 3bet pf situations. Does that mean I have to build a "%turn probe bet" from scratch?

A3) Stat for how often opponent checks flop in position after I did NOT cbet. Can I build this with 100 - "% flop probe bet"?
Same questions for turn..

A4) pf aggressor checks, opponent bets, pf aggressor check-raises. Stat: how often opponent folds to this check-raise?
If I understood, the check-raise stats which are already in PT3 embedded don't depend on preflop action, right?


Case B) he is out of position

B1) He called flop cbet and leads out on the turn, is this "donk turn"?

B2) He called flop cbet and checks the turn, does this exist already?

B3) we both check the flop (I didn't cbet), he leads out on the turn, does this exist already?

B4) we both check the flop (I didn't cbet), he
- check-calls the turn
- check-raises the turn
- check-folds the turn

I'm aware that I need to edit the stats/colums to be valid for 2 players on the flop, can you give me one example so I know what to put?

Hope I didn't confuse you too much...

Thanks for the support.

Kiwi

[kraada]

Replies in bold below:

Hi,

Iam trying to create a postflop pop-up and have questions regarding some stats for 2 players on the flop.
I don't want to invent the wheel twice, so perhaps some of those are already existing.

Case A) he has position

A1) Difference between "% float flop" and "% flop probe bet"? Both are flop bets in position after pf aggressor does NOT cbet, but:
% float flop: is counting only situation where one preflop bet is called (no pf 3bet situations)
% flop probe bet: is counting all situations where any preflop bet is called (also pf 3bet situations)
Is that correct?
I'm not sure what you're using as the Flop Probe Bet definition; if you give me the definition you're using, I can tell you if this is accurate. The Turn Probe and River Probe are for bets out of position after the person who could have cbet the previous street was in position and did not cbet. It's also called bet vs missed cbet OOP in some places.

A3) Stat for how often opponent checks flop in position after I did NOT cbet. Can I build this with 100 - "% flop probe bet"?
Same questions for turn..

You can build this column fairly easily: sum(if[holdem_hand_player_statistics.flg_f_cbet_opp and lookup_actions_f.action = 'X' and not(holdem_hand_player_detail.flg_f_has_position), 1, 0]) for the times they checked it through and sum(if[holdem_hand_player_statistics.flg_f_cbet_opp and lookup_actions_f.action LIKE 'X%' and not(holdem_hand_player_detail.flg_f_has_position), 1, 0]) for the opportunities column.

A4) pf aggressor checks, opponent bets, pf aggressor check-raises. Stat: how often opponent folds to this check-raise?
If I understood, the check-raise stats which are already in PT3 embedded don't depend on preflop action, right?

[b]Correct, but you could add new columns checking for cbet opportunities (flg_f_cbet_opp) along with the check/raise stuff - just adding that check should be sufficient to get what you want.

Case B) he is out of position

B1) He called flop cbet and leads out on the turn, is this "donk turn"?

Yes

B2) He called flop cbet and checks the turn, does this exist already?

No, but you could build it pretty easily by looking for turn donk opportunities and checking the turn.

B3) we both check the flop (I didn't cbet), he leads out on the turn, does this exist already?

That's Probe Turn.

B4) we both check the flop (I didn't cbet), he
- check-calls the turn
- check-raises the turn
- check-folds the turn

These don't exist, but you should be able to build them using the Probe Turn columns as a starting point.

I'm aware that I need to edit the stats/colums to be valid for 2 players on the flop, can you give me one example so I know what to put?

These should work for multiway flops too even if they get a little more complicated, but you can also check for holdem_hand_summary.cnt_players_f = 2 if you want to limit to strictly heads up flops.

Hope I didn't confuse you too much...

Thanks for the support.

Kiwi

[Sopuli]

Thanks for the quick reply, you make it sound so simple.. Here my comments & follow up questions.

Case A) he has position

A1) Difference between "% float flop" and "% flop probe bet"? Both are flop bets in position after pf aggressor does NOT cbet, but:
% float flop: is counting only situation where one preflop bet is called (no pf 3bet situations)
% flop probe bet: is counting all situations where any preflop bet is called (also pf 3bet situations)
Is that correct?
I'm not sure what you're using as the Flop Probe Bet definition; if you give me the definition you're using, I can tell you if this is accurate. The Turn Probe and River Probe are for bets out of position after the person who could have cbet the previous street was in position and did not cbet. It's also called bet vs missed cbet OOP in some places.

Sorry, noticed I didn't explain well enough.
In my mind "probe flop" means pf aggro is not cbetting oop and I bet right away IP on the flop, probing for information...? If that is not probe bet, what is this called?
Now Iam totally confused
Can you check my definitions below for probe and float? Are they both oop?? Not sure what that "val_p_raise_aggressor_pos" means..

Float Flop:
(cnt_f_float / cnt_f_float_opp) * 100

cnt_f_float=
sum(if[ holdem_hand_player_statistics.flg_p_face_raise and
holdem_hand_player_statistics.cnt_p_raise = 0 and
holdem_hand_player_statistics.cnt_p_call = 1 and
holdem_hand_player_detail.flg_f_open_opp and
holdem_hand_player_statistics.flg_f_bet and
holdem_hand_player_detail.val_p_raise_aggressor_pos > holdem_hand_player_statistics.position, 1, 0])

cnt_f_float_opp
sum(if[ holdem_hand_player_statistics.flg_p_face_raise and
holdem_hand_player_statistics.cnt_p_raise = 0 and
holdem_hand_player_statistics.cnt_p_call = 1 and
holdem_hand_player_detail.flg_f_open_opp and
holdem_hand_player_detail.val_p_raise_aggressor_pos > holdem_hand_player_statistics.position, 1, 0])

vs.

Flop Probe Bet:
100 * cnt_f_probe_bet_raised_pot / cnt_f_probe_bet_raised_pot_opp

cnt_f_probe_bet_raised_pot=
sum(if[flg_f_bet and flg_p_face_raise and not(flg_f_cbet_opp) and
holdem_hand_player_detail.val_p_raise_aggressor_pos > holdem_hand_player_statistics.position, 1, 0])

cnt_f_probe_bet_raised_pot_opp=
sum(if[flg_f_open_opp and flg_p_face_raise and not(flg_f_cbet_opp) and
holdem_hand_player_detail.val_p_raise_aggressor_pos > holdem_hand_player_statistics.position, 1, 0])


---------------------------------------

A3) Stat for how often opponent checks flop in position after I did NOT cbet. Can I build this with 100 - "% flop probe bet"?
Same questions for turn..

You can build this column fairly easily: sum(if[holdem_hand_player_statistics.flg_f_cbet_opp and lookup_actions_f.action = 'X' and not(holdem_hand_player_detail.flg_f_has_position), 1, 0]) for the times they checked it through and sum(if[holdem_hand_player_statistics.flg_f_cbet_opp and lookup_actions_f.action LIKE 'X%' and not(holdem_hand_player_detail.flg_f_has_position), 1, 0]) for the opportunities column.

Thanks for that, sounds logical. Couldn't test it yet, my cache needs long time to update...


---------------------------------------

A4) pf aggressor checks, opponent bets, pf aggressor check-raises. Stat: how often opponent folds to this check-raise?
If I understood, the check-raise stats which are already in PT3 embedded don't depend on preflop action, right?

Correct, but you could add new columns checking for cbet opportunities (flg_f_cbet_opp) along with the check/raise stuff - just adding that check should be sufficient to get what you want.

I put the flag in, test it later

---------------------------------------

Case B) he is out of position

B2) He called flop cbet and checks the turn, does this exist already?

No, but you could build it pretty easily by looking for turn donk opportunities and checking the turn.

Ok, I created the stat
Check T after calling Flop Cbet oop=
100 * (cnt_t_donk_opp / cnt_t_check_even_having_donk_opp)
with
cnt_t_check_even_having_donk_opp=
sum(if[holdem_hand_player_statistics.flg_t_check and holdem_hand_player_statistics.flg_t_donk_opp, 1, 0])

Damn, just started the cache updating, I think defined the stat the wrong way around, outcome will be always over 100%...

---------------------------------------

B3) we both check the flop (I didn't cbet), he leads out on the turn, does this exist already?

That's Probe Turn.

Is there a stat "probe turn" somewhere?? Can you attach it? Is there also "probe river"?Thanks!
Seems probe and float is a mystery to me...

---------------------------------------

B4) we both check the flop (I didn't cbet), he
- check-calls the turn
- check-raises the turn
- check-folds the turn

These don't exist, but you should be able to build them using the Probe Turn columns as a starting point.

Ok, I'll try once I have the "probe turn" stat.

---------------------------------------

I'm aware that I need to edit the stats/colums to be valid for 2 players on the flop, can you give me one example so I know what to put?

These should work for multiway flops too even if they get a little more complicated, but you can also check for holdem_hand_summary.cnt_players_f = 2 if you want to limit to strictly heads up flops.

Ok, thanks.

[kraada]

Your probe flop is pretty close to the float flop as defined above, as you say correctly, 3bet and 4bet pots are counted in that setup.

Apparently Probe Turn and Probe River are going to be new built in stats for Beta 30.9 - I lose track sometimes of what was in the last version as we get close to release. So you'll have those as soon as the next upgrade is out; sorry for the confusion

[Sopuli]

Good news, so I'll wait for the next beta.

One thing regarding A3, I think the first column is wrong and/or I explained wrong.
Looking at the columns you suggested, looks like he is the one who checks instead of cbetting, right? If so, that's good, because I need those too
Does "not(holdem_hand_player_detail.flg_f_has_position)" mean out of position? Iam still learning..

Idea is that I will not cbet and opponent has chance to bet in position but checks.

So I tried to build these, please check them, I got a syntax error and was not able to find the problem.

cnt_f_check_IP_after_opponent_didNOT_cbet=
sum(if[holdem_hand_player_statistics.flg_p_face_raise AND
holdem_hand_player_statistics.flg_p_fold=0 AND
holdem_hand_player_statistics.flg_p_raise=0 AND
holdem_hand_player_statistics.flg_f_check AND
holdem_hand_player_detail.flg_f_has_position, 1, 0])

How can I build the other column which shows the times he could bet in position after I don't cbet?
I couldn't find a column like cnt_f_bet_opp ...

Last, one general question, there are many colums with "def" in it, like cnt_f_bet_def_opp, what does that mean?

Thanks!

[kraada]

Yes, you are correct, I did A3's columns in the other direction.

"def" in a column name tends to mean "defense" so flg_f_3bet_def_opp is whether or not a player faced a flop 3bet (he had a 3bet defense opportunity)

[Sopuli]

Can you check my line for the "right" direction? I can't figure it out..

[kraada]

Your column for the times he bets looks good, for the opportunities you want flg_f_open_opp - if he has the opportunity to open the pot, he can check or bet.