I have been working on a custom HUD stat, for determine the following player types (+ donkey as a fallback)
- Code: Select all
TABLE #1: PLAYER TYPE REFERENCE DATA
╔═════════╦═══════╦══════╦═════╦═════════════════╗
║ TYPE ║ VPIP ║ PFR ║ AF ║ VPIP/PFR Ratio ║
╠═════════╬═══════╬══════╬═════╬═════════════════╣
║ NIT ║ 15 ║ 12 ║ 2 ║ 80 ║
║ TAG ║ 21 ║ 18 ║ 3 ║ 86 ║
║ LAG ║ 26 ║ 23 ║ 3 ║ 88 ║
║ WHALE ║ 55 ║ 9 ║ 2 ║ 16 ║
║ MANIAC ║ 55 ║ 38 ║ 4 ║ 69 ║
╚═════════╩═══════╩══════╩═════╩═════════════════╝
This is the actual stats I have been using
HANDS = cnt_hands
VPIP = (cnt_vpip / (cnt_hands - cnt_walks)
PRF = (cnt_pfr / cnt_pfr_opp) * 100)
PFR/VPIP ratio = (cnt_pfr / cnt_pfr_opp) * 100) / (cnt_vpip / (cnt_hands - cnt_walks) * 100
Here is a simplified stat script using these stat variables
(I wish there was a way to use variables in PT4 stat scripts)
- Code: Select all
if (HANDS > 20,
if (VPIP < 17, format('Nit'),
if (PFR / VPIP > 0.75 and VPIP <= 35,
if (VPIP < 17, format('Nit'),
if (VPIP > 17 and VPIP < 25, format('TAG'),
if (VPIP >= 25, format('LAG'),
format('Donkey')))),
if (PFR < 18, format('Whale'),
format('Maniac')))),
format('Unknown'))
Original working code
Spoiler: show
Based on this custom stat I have created a report with statistic from a total of 859 players (all with a hand sample over 20). From this 859-sample I created a new table with the average stats based on each player type to see how the numbers stack up compared to the reference data (table #1 above)
- Code: Select all
TABLE #2: AVERAGE BASED ON REAL DATA SAMPLE
╔════════╦══════╦═════╦════╦════════════════╗
║ TYPE ║ VPIP ║ PFR ║ AF ║ VPIP/PFR Ratio ║
╠════════╬══════╬═════╬════╬════════════════╣
║ Nit ║ 13 ║ 9 ║ 2 ║ 72 ║
║ TAG ║ 21 ║ 18 ║ 3 ║ 85 ║
║ LAG ║ 29 ║ 24 ║ 3 ║ 84 ║
║ WHALE ║ 32 ║ 12 ║ 2 ║ 43 ║
║ MANIAC ║ 43 ║ 27 ║ 3 ║ 65 ║
╚════════╩══════╩═════╩════╩════════════════╝
When comparing these two tables you can tell that for the first three player types (Nit, TAG, LAG) the numbers add up quite accurately. I realize that this is no exact science, and that I cannot expect the numbers to match 100%. Still, for player types WHALE and MANIAC the number do not add up as nicely.
I would appropriate any input, advice or suggestions on how to improve my script to determine the player types (especially WHALE and MANIAC) more accurately