Messages

Forum Index : Pro Golf : Messages
Author: [Pro Golf] Topic: Heritage Discussion
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#1
Posted: 4/10/2012 2:18:05 AM
Been working on some new statistical correlations to determine key stats for the week.  For right now I just want to see how well I can copy and paste this, if it works, tomorrow I'll explain the numbers and see if I can run some other years too.  This is just based on last year's tourney.

0.60
0.31 1
0.56 0.33 2
0.55 0.30 3
0.52 0.44 4
0.45 0.25 5
ppr 0.45 0.18 6
0.37 0.12 7
0.34 0.14 8
-0.02 0.00 9
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#2
Posted: 4/10/2012 2:20:05 AM
Well garbage.  It didn't work at all.  Anybody have any tips how can cut and paste a small block out of spreadsheet and put in here?  It's only 10 rows by 4 colums, so there's plenty of room.

quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#3
Posted: 4/10/2012 2:29:43 AM
Looks like I can paste them into my covers blog.  I'll use that, I'll also put some explanation up there tomorrow.
quote
MrBogey send a private message View Space | Blog | Friends | Playbook |
MrBogey
Participation Meter
Veteran
Joined: Aug 2011
Posts: 3042
Location: Florida
#4
Posted: 4/10/2012 3:34:43 AM
confused as garbage here sir
quote
Bluefin send a private message View Space | Blog | Friends | Playbook |
Bluefin
Participation Meter
Captain
Joined: Nov 2006
Posts: 8044
Location: Florida
#5
Posted: 4/10/2012 8:34:18 AM
good luck bud hopefully you'll still have some hair left by the time you get it..
quote
Chus send a private message View Space | Friends | Playbook | My Sportsbook: 5Dimes |
Chus
Participation Meter
Prospect
Joined: Nov 2006
Posts: 431
Location: Illinois
#6
Posted: 4/10/2012 2:04:36 PM
cut and paste into notepad first.  then cut and paste into your reply
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#7
Posted: 4/10/2012 7:06:05 PM
QUOTE Originally Posted by Bluefin:

good luck bud hopefully you'll still have some hair left by the time you get it..

Haha Blue.  I'm already bald as cue ball!!!
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#8
Posted: 4/10/2012 7:08:12 PM
Slope   R2     Rank
par4 0.60   0.31 1
sc 0.56   0.33 2
gir 0.55   0.30 3
par5 0.52   0.44 4
pa 0.45   0.25 5
ppr 0.45   0.18 6
par3 0.37   0.12 7
da 0.34   0.14 8
dd -0.02   0.00 9
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#9
Posted: 4/10/2012 7:20:16 PM
Ok, So how this works:

1) I've used ALL scores from the first 2 rounds, including players that have missed the cut.   This is different from just about any other site that I've seen doing statistical correlations between key stats and score - many of them even only use the top 5 or 10 or at best all players that make the cut.  We've discussed in the past that it's often more effective to find "fades" than "fors", so I think it's important to have the bottom end of the correlations in there.

2) First thing I do is normalize all scores and stats to be between 0 and 1.  0=sucks, 1=the best

3) Then, the numbers in the table above are a simple linear regression between the Y (score) and the X (different stats).  

Slope= the slope of the regression line.  The closer this value is to 1, the better indicator that stat is of a players score.

R2= this is R-squared.  A pure statistician probably wouldn't like my simplification, but you can think of it as the confidence you have in the slope number.  R2 also has a max of 1.  In other words, the closer it is to 1, the better the data fits the regression line or the fewer "outliers" we have.

The first column is the different stats:
sc=scrambling
gir=greens in reg
pa=putting avg
ppr=putts per green in reg
da=driving accuracy
dd=driving distance
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#10
Posted: 4/10/2012 8:18:22 PM
I added a couple more stats:  BA=bogey avoidance and PB=par breakers(birdies+eagles).


Slope R2 Rank
BA   0.75 0.59 1
par4   0.60 0.31 2
sc   0.56 0.33 3
gir   0.55 0.30 4
par5   0.52 0.44 5
PB   0.51 0.43 6
pa   0.45 0.25 7
ppr   0.45 0.18 8
par3   0.37 0.12 9
da   0.34 0.14 10
dd  -0.02 0.00 11
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#11
Posted: 4/10/2012 8:35:31 PM
To help get a better feeling for the impact of these numbers, I ran this year's masters through the spreadsheet.  I think you'll find that the numbers are telling you what you're gut probably felt and this should serve as a good gage going forward to tournaments when maybe we don't have such a good gut feel for it:

Slope R2 Rank
gir   0.97 0.45 1
BA   0.91 0.67 2
PB   0.79 0.60 3
pa   0.67 0.41 4
par5   0.65 0.51 5
dd   0.58 0.20 6
sc   0.53 0.24 7
par4   0.52 0.19 8
ppr   0.35 0.12 9
par3   0.31 0.08 10
da   -0.04 0.00 11
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#12
Posted: 4/10/2012 8:58:17 PM
From the 2011 Malaysian Open:

Slope R2 Rank
BA   0.74 0.73 1
par4   0.66 0.45 2
gir   0.60 0.35 3
PB   0.55 0.41 4
sc   0.46 0.23 5
par5   0.44 0.38 6
par3   0.39 0.17 7
pa   0.33 0.19 8
ppr   0.25 0.08 9
da   0.19 0.05 10
dd   0.07 0.01 11
quote
blairaj send a private message View Space | Blog | Friends | Playbook |
blairaj
Participation Meter
Rookie
Joined: Dec 2007
Posts: 969
Location:
#13
Posted: 4/10/2012 11:21:19 PM
Full tourney leans for Heritage, I'll probably play them all, assuming I get something close to these lines:

Z Johnson-109 o Els
Howell-127 o Fowler
Cauley-128 o Bae
Molder-105 o Petterson
happy-114 o Barnes
O'Hern-108 o Immelman
quote
Forum Index : Pro Golf : Messages
You have entered the forum as a GUEST. 
You must login/register to post or reply.