Question about the RFI stat

Forum for users that want to write their own custom queries against the PT database either via the Structured Query Language (SQL) or using the PT3 custom stats/reports interface.

Moderator: Moderators

Question about the RFI stat

Postby TheZebra » Fri Sep 03, 2010 7:25 pm

Have a question about the RFI stat, know this is not the "learn SQL forum" so appologize in advance since this is 110% me just not quite understanding SQL and the PT3 DB. But trying to learn.

Anyways the RFI stat value expression goes: (cnt_p_raise_first_in / cnt_p_open_opp ) * 100

The column nt_p_raise_first_in reads:
sum( if[tourney_holdem_hand_player_statistics.flg_p_first_raise, 1, 0])

The column cnt_p_open_opp reads:
sum(if[tourney_holdem_hand_player_statistics.flg_p_open_opp, 1, 0])

As far as i understand and could check with SQL queries is that the flg_p_first_raise is true when a player raises even if there has been limps previous,
than as far as i understand the flg_p_open_opp is false if their has been a previous limp (or a previous raise).

So than basically as i see it if PlayerX play 5 hands and each hand raise limpers his cnt_p_raise_first_in would be 5 and cnt_p_open_opp would be 0 making the RFI = 0% of times he raised first in? I dont understand how it can get right if the cnt_p_open_opp dont add in the times PlayerX is just opening a limped pot.

Appologize, known somewhere im so wrong here but just have to ask what it is that im not getting.
TheZebra
 
Posts: 10
Joined: Sat Aug 28, 2010 11:33 am

Re: Question about the RFI stat

Postby WhiteRider » Sat Sep 04, 2010 1:36 am

"cnt_p_raise_first_in" is actually defined as:
sum( if[holdem_hand_player_statistics.flg_p_first_raise AND holdem_hand_player_statistics.flg_p_open_opp, 1, 0])

You're probably looking at "cnt_p_first_raise".

(Feel free to ask questions - that's what we're here for!)
WhiteRider
Moderator
 
Posts: 54025
Joined: Sat Jan 19, 2008 7:06 pm
Location: UK

Re: Question about the RFI stat

Postby TheZebra » Sat Sep 04, 2010 12:04 pm

Follow up question,

The RFI and Steal stats than seem to work exactly the same than? Only that with RFI u can actually filter by any position u want? Is that a correct observation?

Thanks Whiterider for the help.
TheZebra
 
Posts: 10
Joined: Sat Aug 28, 2010 11:33 am

Re: Question about the RFI stat

Postby kraada » Sat Sep 04, 2010 12:24 pm

Attempt to Steal is the same as RFI in the CO, BTN or SB - RFI cannot normally be filtered to "Blinds or CO or Button" hence this needs its own stat.

You can use RFI to filter for any one of the following positions though: EP, MP, LP, Blinds.
kraada
Moderator
 
Posts: 54430
Joined: Wed Mar 05, 2008 2:32 am
Location: NY

Re: Question about the RFI stat

Postby TheZebra » Sat Sep 04, 2010 1:02 pm

kraada wrote:Attempt to Steal is the same as RFI in the CO, BTN or SB - RFI cannot normally be filtered to "Blinds or CO or Button" hence this needs its own stat.

You can use RFI to filter for any one of the following positions though: EP, MP, LP, Blinds.


out of curiousty, why cant RFI normally be filtered to "Blinds or CO or Button"
TheZebra
 
Posts: 10
Joined: Sat Aug 28, 2010 11:33 am

Re: Question about the RFI stat

Postby kraada » Sat Sep 04, 2010 2:06 pm

Because that's not how the HUD filtering system works - you can filter any individual statistic to one position type, not a combination of position types.

If you wished to make a custom statistic, you could make any combination you desired, though.
kraada
Moderator
 
Posts: 54430
Joined: Wed Mar 05, 2008 2:32 am
Location: NY

Re: Question about the RFI stat

Postby TheZebra » Sat Sep 04, 2010 4:24 pm

kraada wrote:Because that's not how the HUD filtering system works - you can filter any individual statistic to one position type, not a combination of position types.

If you wished to make a custom statistic, you could make any combination you desired, though.


I think this is what i mean and im attempting to do. The sentence above i wrote,

"The RFI and Steal stats than seem to work exactly the same than? Only that with RFI u can actually filter by any position u want? Is that a correct observation?"

was a bit unclear. What i meant was that the RFI would give me the possibility to create a custom made steal stat for any position that i could use in the HUD. Not that i could use the RFI and than use the HUD to filter it somehow for each position.

U probably remember u gave me some help in creating a steal stat based on villian stack in BBs in a previous thread. I wanted to add the HJ too, but than flg.steal_att is always false for position=2 since position=2 per default isnt a steal position right?, ran some SQL against the DB aswell to check that this assumtion was correct asweel(so as i father the flg_steal_att can than only be used for pos 9,0,1).

I could obv use flg_steal_att for SB,But and SB than but since i want HJ and that i might want to input another steal pos or combine for instane a steal stat for MP pos etc i thought i might aswell make a custom RFI stat that works as a Steal stat since it is seemed more simple (where i can just dup it and alter the position)

STEAL(CAPS) below uses the flg_steal_att and the steal (low caps) is the RFI setup to work as a steal stat (and can be altered any position) I run SQL queries to double check that it came up with the same numbers and as far as i can tell it does, i input how i wrote it below so u can double check.

STAT: STEAL button <12BB stack:
((Z_STEAL_att_P0_12BB) / (Z_STEAL_opp_P0_12BB)) *100

COLUMN: Z_STEAL_att_P0_12BB
sum(if[tourney_holdem_hand_player_statistics.flg_steal_att AND (tourney_holdem_hand_player_statistics.position)=0 AND (tourney_holdem_hand_player_detail.amt_before / ( tourney_holdem_blinds.amt_bb )) <12 , 1, 0])

COLUMN: Z_STEAL_opp_P0_12BB
sum(if[tourney_holdem_hand_player_statistics.flg_steal_opp AND (tourney_holdem_hand_player_statistics.position) =0 AND (tourney_holdem_hand_player_detail.amt_before / ( tourney_holdem_blinds.amt_bb )) <12, 1, 0])

Instead i plan to use:

STAT: steal button <12BB stack:
((Z_steal_att_P0_12BB) / (Z_steal_opp_P0_12BB)) * 100

COLUMN: Z_steal_att_P0_12BB
sum(if[tourney_holdem_hand_player_statistics.flg_p_open_opp AND (tourney_holdem_hand_player_statistics.position) =0 AND (tourney_holdem_hand_player_detail.amt_before / (tourney_holdem_blinds.amt_bb)) <12, 1, 0])

COLUMN Z_steal_opp_P0_12BB
sum(if[tourney_holdem_hand_player_statistics.flg_p_open_opp AND (tourney_holdem_hand_player_statistics.position) =0 AND (tourney_holdem_hand_player_detail.amt_before / (tourney_holdem_blinds.amt_bb)) <12, 1, 0])

Is this all cool or is there something im not quite getting.

Once again, thank to u both for all the help.
TheZebra
 
Posts: 10
Joined: Sat Aug 28, 2010 11:33 am

Re: Question about the RFI stat

Postby kraada » Sat Sep 04, 2010 5:44 pm

That looks good :)
kraada
Moderator
 
Posts: 54430
Joined: Wed Mar 05, 2008 2:32 am
Location: NY

Re: Question about the RFI stat

Postby WhiteRider » Sun Sep 05, 2010 5:40 am

It's probably just a typo or copy/paste error, but both of those columns are the same - you need:

COLUMN: Z_steal_att_P0_12BB
sum(if[tourney_holdem_hand_player_statistics.flg_steal_att AND (tourney_holdem_hand_player_statistics.position) =0 AND (tourney_holdem_hand_player_detail.amt_before / (tourney_holdem_blinds.amt_bb)) <12, 1, 0])

..but I guess you probably have that right in PT3 already!
WhiteRider
Moderator
 
Posts: 54025
Joined: Sat Jan 19, 2008 7:06 pm
Location: UK


Return to Custom Stats, Reports, and SQL [Read Only]

Who is online

Users browsing this forum: No registered users and 0 guests