Over and Under Achieving Golfers

[saevel25]

I was curious after the discussion in this thread about Rory,

I decided to look at wins versus strokes gained averages. I took data from the last three seasons.

If you limit the data to anyone who averaged above .250 strokes gained for the season. You end up with this nice graph. Under .250 strokes gained, the graph goes linear at 1 win average per season from .250 to -1.250 strokes gained.

I used this to see what the expected wins per season a golfer should have.

Henrik Stenson underperformed the most in a single season in 2015.

Paul Casey underperformed the most over the past three years if you total difference between their wins and expected wins over that time frame.

Justin Thomas overachieved the most in 2017.

Jordan Speith has overachieved the most over the past 3 seasons.

Rory has underachieved over the past three seasons, by a value of -2.0 total.

This is just looking at wins, so a bit of luck is in this as well.

 

[iacas]

So the bottom axis is strokes gained, and the left axis is expected wins?

But didn't you map more players than wins are available?

[saevel25]

8 minutes ago, iacas said:

So the bottom axis is strokes gained, and the left axis is expected wins?

But didn't you map more players than wins are available?

The bottom axis is strokes gained. The left axis is expected wins.

I took the top 10 list from the PGA Tour website. So, who ever got at least in the Top 10 ended up as a data point.

[iacas]

2 minutes ago, saevel25 said:

The bottom axis is strokes gained. The left axis is expected wins.

I took the top 10 list from the PGA Tour website. So, who ever got at least in the Top 10 ended up as a data point.

Right, but you end up with a total of how many wins per year based on that chart? How many data points do you have? 55? 60?

And all of those data points have at least 1 win, with the average being closer to 1.5 to 1.7, which means you have to find room for almost 100 wins in a regular season. There aren't 100 events in a regular season.

I feel like you had to then apply a multiplier (or divisor). If someone is 3.5/96 in your chart, and there are 42 PGA Tour events, then 3.5/96*42 =~ 1.5 expected wins. Find a winning percentage and then apply that to the number of PGA Tour events.

That's before you even account for the number of events in which someone plays (which you should do much further upstream, not toward the end). No?

(If two players have an expected win rate of 0.7 and one player plays only 15 PGA Tour events while another plays 25… odds are the 25-event guy will win more events, because strokes gained per round doesn't take into account how often that person plays.)

[Dr. Manhattan]

John Daly wasted most of his incredible talent. 

[saevel25]

17 minutes ago, iacas said:

Right, but you end up with a total of how many wins per year based on that chart? How many data points do you have? 55? 60

What I did was took the Top 10 data from 2015, 2016 & 2017. I also took the same years from the Strokes Gain Total. I cross-referenced the two data sets. I then used BIN ranges to average the strokes gained and wins. That is were I got that chart. If a golfer averages 1.5 strokes gained, they should expect to get 1.85 wins for the season.

I took their actual wins for the season and subtracted the expected wins.  The total win value is the total expected wins over the past three seasons. I was curious to see who overachieved the most over that span.

Stenson, at 2.2 strokes gained, should have gotten close to 3.5 wins for the season. He got no wins.

Paul Casey had a higher total difference between his actual wins and expected wins over the past three seasons.

 

[iacas]

1 minute ago, saevel25 said:

What I did was took the Top 10 data from 2015, 2016 & 2017. I also took the same years from the Strokes Gain Total. I cross-referenced the two data sets. I then used BIN ranges to average the strokes gained and wins. That is were I got that chart. If a golfer averages 1.5 strokes gained, they should expect to get 1.85 wins for the season.

I took their actual wins for the season and subtracted the expected wins.  The total win value is the total expected wins over the past three seasons. I was curious to see who overachieved the most over that span.

Stenson, at 2.2 strokes gained, should have gotten close to 3.5 wins for the season. He got no wins.

Paul Casey had a higher total difference between his actual wins and expected wins over the past three seasons.

Matt, I feel like you're missing my point.

How many wins per season does your chart predict? It seems like significantly more than events are available to win.

[saevel25]

1 minute ago, iacas said:

Matt, I feel like you're missing my point.

How many wins per season does your chart predict? It seems like significantly more than events are available to win.

I see your point on that.

Yea, I think they are inflated due to how the excel function works. It does not consider a blank cell as a zero. So it averaged only the players who actually got a win.

[iacas]

2 minutes ago, saevel25 said:

I see your point on that.

Yea, I think they are inflated due to how the excel function works. It does not consider a blank cell as a zero. So it averaged only the players who actually got a win.

So what's the better version of the chart?

[MuniGrit]

52 minutes ago, Dr. Manhattan said:

John Daly wasted most of his incredible talent. 

I'd raise you a Fred Couples. I just was watching the 90 PGA & 98 Masters and how he blew those. I feel he should have won 5 majors in his career. Now he did have a very successful career and made a boat load of money.

[iacas]

4 minutes ago, MuniGrit said:

I'd raise you a Fred Couples. I just was watching the 90 PGA & 98 Masters and how he blew those. I feel he should have won 5 majors in his career. Now he did have a very successful career and made a boat load of money.

Fred battled back injuries and troubles his whole career.

John Daly battled himself.

[mchepp]

I think this is a pretty cool concept to think about @saevel25. It would be interesting to see if strokes gained putting from a certain distance helped with that over performance or driving or if there are any trends related to individual strokes gained statistics. 

[MuniGrit]

3 minutes ago, iacas said:

Fred battled back injuries and troubles his whole career.

John Daly battled himself.

He still blew those two tournaments I just listed and probably a couple more where he was right there on the back nine last day.

[saevel25]

Ok, the latest and greatest. I expanded it thru 2013.

The most underachieving season was Henrik Stenson's 2015 season, with 0 wins, and 2.85 expected wins.

PLAYER NAME		AVERAGE		Year		Victories		Exp Wins		Diff
Henrik Stenson		2.21		2015		0		2.8		-2.8
Steve Stricker		2.19		2013		0		2.8		-2.8
Sergio Garcia		1.98		2014		0		2.2		-2.2
Jim Furyk		1.80		2014		0		1.7		-1.7
Sergio Garcia		1.52		2013		0		1.2		-1.2
Hideki Matsuyama		1.46		2015		0		1.1		-1.1
Paul Casey		1.45		2017		0		1.0		-1.0
Keegan Bradley		1.40		2013		0		1.0		-1.0
Jim Furyk		1.39		2013		0		0.9		-0.9
Rickie Fowler		1.38		2016		0		0.9		-0.9

The most overachieving season is Justin Thomas in 2017, with 5 wins, and a 1.3 expected win value.

 

PLAYER NAME		AVERAGE		Year		Victories		Exp Wins		Diff
Justin Thomas		1.62		2017		5		1.3		3.7
Tiger Woods		2.06		2013		5		2.4		2.6
Hideki Matsuyama		0.99		2017		3		0.5		2.5
Jason Day		2.11		2015		5		2.5		2.5
Jordan Spieth		2.15		2015		5		2.7		2.3
Jimmy Walker		1.25		2014		3		0.8		2.2
Dustin Johnson		1.91		2017		4		2.0		2.0
Martin Kaymer		0.24		2014		2		0.1		1.9
Patrick Reed		0.33		2014		2		0.1		1.9
Russell Knox		0.51		2016		2		0.2		1.8

Here is the golfers who has underachieved the most over the  past 5 seasons

PLAYER NAME		Total Victories		Total Expected Wins		Diff
Jim Furyk		1		4.3		-3.3
Paul Casey		0		3.2		-3.2
Steve Stricker		0		3.1		-3.1
Sergio Garcia		2		4.6		-2.6
Keegan Bradley		0		2.3		-2.3
Henrik Stenson		4		6.1		-2.1
Justin Rose		3		5.1		-2.1
Ryan Palmer		0		2.0		-2.0
Graham DeLaet		0		1.8		-1.8
Charl Schwartzel		1		2.6		-1.6

Here the golfers who has overachieved the most over the past 5 seasons

PLAYER NAME		Total Victories		Total Expected Wins		Diff
Justin Thomas		6		1.9		4.1
Jimmy Walker		6		2.2		3.8
Jordan Spieth		11		7.4		3.6
Dustin Johnson		10		6.4		3.6
Patrick Reed		5		1.5		3.5
Billy Horschel		4		1.2		2.8
Tiger Woods		5		2.4		2.6
Brandt Snedeker		4		1.8		2.2
Hideki Matsuyama		5		3.0		2.0
Jason Dufner		3		1.0		2.0

Maybe recently it has been more apparent, but Rory has had 7 victories since 2013, and he was expected to get 7.3 victories.

I took Rory's Strokes Gained Putting and subtracted it from his Strokes Gained total. I wanted to see how much is putting was contributing to his wins.

Rory with his putting stats: 7.3 expected wins
Rory with PGA Tour average Putting: 9.2 expected wins

If Rory had a PGA Tour average putting stats, he would be expected to win nearly 2 more times from 2013 thru 2017 seasons.

[MuniGrit]

Rickie Fowler?