Expected Birdie Percentage

Expected Birdie Percentage
By Mark Sweeney and Jamie Donaldson

I tend to spend a fair amount of time looking at a player’s statistics in an attempt to really understand where their strengths and weaknesses are, particularly in the putting and scoring arena. The primary stats we have to work with are GIR’s, Scrambling, Putts Gained, and Approach Shot Proximity, but I still often find I can’t quite summarize a player’s round from these.

For example, a player can hit 14 greens and shoot even par or shoot 6 under par, but what was the difference in performance that caused the difference in scores? Putts Gained certainly helps to clarify his putting performance, but not his approach proximities. And the Approach Proximity stat is just an average and doesn’t give birdie opportunities. For example an Approach Proximity of 25 feet doesn’t tell whether a player had only 25-foot birdie putts, or whether he had five 10-foot birdie putts and one 100-foot putt.

What I really need to understand comes from an idea I got from Dr. Bhrett McCabe called “In Position”, meaning how often when you hit the green were you in position to make a birdie (generally inside 20 feet). If a player is hitting the green more than 20 feet from the hole, then he can’t be expected to make a lot of birdies. In Position is a great stat to look at but I still want more—how close is a player hitting it and what are his expected number of birdies given his ball striking performance.

Lately Jamie Donaldson (Senior AimPoint Instructor for Europe) and I came up with some new metrics called the Expected Birdie Percentage (EBP) and the Actual Birdie Percentage (ABP). The idea is simple: when a player hits the green in regulation, what were the odds of him making a birdie (EBP), and how did he perform relative to those odds (ABP)?

What we did is look at the probability of making birdies given the length of the player’s first putt after hitting the green in regulation, or his second putt if hitting a par 5 green in two shots (for now I’m ignoring chips-ins, hole-outs, etc.). For example, a 3 foot birdie putt should be made 99% of the time, and a 15 foot birdie putt should be made about 24% of the time. A player’s average make probability is his EBP. A player’s actual make percentage of those putts is his ABP.

The results were interesting to say the least[1]. The players with the top 5 highest EBP’s are:

Rank	Player Name	EBP	Driving Dist. Rank	Money Rank
1	Watson, Bubba	39.5%	1	1
2	McIlroy, Rory	38.3%	5	28
3	Holmes, J.B.	36.2%	4	27
4	Johnson, Dustin	35.8%	2	3
5	Woodland, Gary	35.8%	15	25

All are very long hitters who can hit par 5 greens in two shots. This finding supports Mark Broadie’s research in Every Shot Counts suggesting that long shots into greens are highly important, particularly the range of 175-225 yards. What we may be seeing is that these are the big boy’s second shot into the par 5 greens, setting up eagles and easy birdies, versus long approaches into par 4’s which will generally lead to a two-putt par.

We can also look at how many birdies a player actually made after a GIR (ABP), and the top 5 players are:

Rank	Player Name	ABP	Money Rank
1	McIlroy, Rory	41.8%	28
2	Schwartzel, Charl	37.8%	68
3	Woods, Tiger	37.8%	n/a
4	Watson, Bubba	37.8%	1
5	Walker, Jimmy	37.7%	2

No big surprises there, but why is Woods ranked low in EBP and high in ABP?

Then we compare the expected birdie percentage (EBP) and the actual birdie percentage (ABP) and calculated the difference between the two into a metric we called Gain. A players Gain is how many more percentages of birdies the player made above his expectation, due to putting. The top 5 players in Gain are:

Rank	Player Name	Gain	Putts Gained Rank
1	Woods, Tiger	11.5	n/a
2	Every, Matt	6.2	3
3	Walker, Jimmy	5.2	4
4	Donald, Luke	4.7	6
5	Loupe, Andrew	3.9	11

This means these players had the highest level of over-performance in making birdies relative to their ball striking proximities. A Gain of 0.0 means you are meeting the tour average for converting birdie putts, while a negative Gain means you are under-performing and leaving birdies on the table.

These metrics can be very useful in understanding a player’s round because it gives a picture of how well he put himself in position to make birdies and how well he putted for birdies after a GIR. Let’s take a look at Martin Kaymer’s rounds at The Players Championship:

Round	EBP	ABP	Gain
1	44%	53%	8.8
2	32%	38%	6.9
3	26%	18%	-7.7
4	34%	23%	-11.3

His first two rounds were spectacular overall, and on the weekend he actually made fewer birdies than expected, even though every day except round 3 he put himself in great positions.

Remember these metrics are not meant to predict total score, just to give insight into a player’s ability to make birdies. It also doesn’t take into account (at the moment) field putting averages on those days, just overall yearly putting averages. If we look at the field averages for each round at The Players we see the following:

Field	EBP	ABP	Gain
Round 1	33%	32%	-1.4
Round 2	29%	24%	-5.7
Round 3	32%	33%	0.9
Round 4	32%	29%	-3.1

In 3 out of 4 rounds players under-performed on the greens, possibly due to the putting difficulty, pin positions, weather, or faster green speeds at The Players.

The Expected Birdie Percentage is a method of evaluating a player’s chance of making a birdie after hitting a green in regulation–the higher the percentage the better his approach shots are. Actual Birdie Percentage then shows his success in converting those opportunities into birdies. It shows what birdie chances you are creating, how that compares to tour average, and how many of those you actually convert. What the data suggests is that long hitters have a big advantage in scoring on par 5’s; if their birdie probabilities are higher than other players, they will make more birdies over a large number of rounds than the field. The data certainly needs more detailed analysis, but this quick overview shows interesting patterns and can help players and coaches understand scoring a little better.

2014 EBP Rankings

EBP Rank	Player Name	EBP	ABP	Gain
1	Watson, Bubba	39.5%	37.8%	-1.7
2	McIlroy, Rory	38.3%	41.8%	3.4
3	Holmes, J.B.	36.2%	35.1%	-1.1
4	Johnson, Dustin	35.8%	34.9%	-0.9
5	Woodland, Gary	35.8%	31.4%	-4.3
6	Rose, Justin	35.5%	32.8%	-2.7
7	Matsuyama, Hideki	34.9%	37.7%	2.9
8	Glover, Lucas	34.7%	28.2%	-6.5
9	Ishikawa, Ryo	34.2%	34.5%	0.3
10	Schwartzel, Charl	34.1%	37.8%	3.7
11	Peterson, John	34.1%	27.6%	-6.4
12	Scott, Adam	33.9%	36.8%	2.9
13	Streelman, Kevin	33.8%	33.8%	0.0
14	Palmer, Ryan	33.8%	33.5%	-0.3
15	Bradley, Keegan	33.7%	33.6%	-0.1
16	Mickelson, Phil	33.7%	31.7%	-2.0
17	Kirk, Chris	33.6%	33.8%	0.2
18	Kaymer, Martin	33.5%	30.9%	-2.6
19	Dufner, Jason	33.4%	33.8%	0.3
20	Spieth, Jordan	33.4%	33.4%	0.0
21	Kokrak, Jason	33.3%	34.4%	1.1
22	Noh, Seung-Yul	33.3%	30.4%	-2.9
23	Matteson, Troy	33.2%	30.0%	-3.2
24	Watney, Nick	33.2%	28.9%	-4.2
25	Hoffman, Charley	33.1%	33.5%	0.4
26	La’Cassie, Bronson	33.1%	28.5%	-4.6
27	Guthrie, Luke	33.0%	32.0%	-1.0
28	Moore, Ryan	33.0%	32.3%	-0.7
29	Garrigus, Robert	33.0%	34.8%	1.8
30	Bae, Sang-Moon	32.9%	32.8%	-0.1
31	Chappell, Kevin	32.9%	31.4%	-1.5
32	Westwood, Lee	32.9%	32.3%	-0.6
33	Hadley, Chesson	32.9%	32.6%	-0.3
34	Beljan, Charlie	32.6%	31.1%	-1.5
35	Mahan, Hunter	32.6%	30.6%	-2.0
36	Stanley, Kyle	32.6%	25.7%	-6.8
37	Walker, Jimmy	32.5%	37.7%	5.2
38	Roach, Wes	32.5%	30.9%	-1.6
39	Stadler, Kevin	32.4%	30.8%	-1.6
40	Fowler, Rickie	32.4%	31.5%	-0.9
41	Leishman, Marc	32.4%	30.8%	-1.6
42	Wilcox, Will	32.3%	28.9%	-3.3
43	Brown, Scott	32.2%	33.0%	0.8
44	Bowditch, Steven	32.2%	30.1%	-2.2
45	Poulter, Ian	32.1%	32.4%	0.3
46	Fritsch, Brad	32.1%	26.9%	-5.2
47	Stallings, Scott	32.1%	29.5%	-2.6
48	Steele, Brendan	31.9%	32.1%	0.2
49	Reed, Patrick	31.9%	30.5%	-1.4
50	Lovemark, Jamie	31.9%	29.7%	-2.1
51	Jones, Matt	31.8%	32.9%	1.1
52	Austin, Woody	31.8%	29.0%	-2.8
53	Horschel, Billy	31.8%	28.9%	-2.9
54	English, Harris	31.7%	31.6%	-0.1
55	Putnam, Michael	31.5%	27.5%	-4.1
56	Colsaerts, Nicolas	31.5%	30.4%	-1.0
57	Sabbatini, Rory	31.4%	29.7%	-1.7
58	Herman, Jim	31.4%	27.1%	-4.3
59	Cauley, Bud	31.4%	29.0%	-2.4
60	Gainey, Tommy	31.4%	29.5%	-1.9
61	Davis, Brian	31.4%	33.5%	2.2
62	Na, Kevin	31.3%	33.2%	1.8
63	Lee, Danny	31.3%	29.9%	-1.4
64	Lee, Richard H.	31.3%	29.9%	-1.4
65	Villegas, Camilo	31.2%	32.0%	0.7
66	Perez, Pat	31.2%	31.8%	0.6
67	Svoboda, Andrew	31.2%	31.1%	-0.1
68	Simpson, Webb	31.2%	35.0%	3.9
69	Johnson, Zach	31.2%	32.3%	1.2
70	McNeill, George	31.2%	31.7%	0.6
71	Barnes, Ricky	31.1%	30.4%	-0.7
72	Els, Ernie	31.1%	29.1%	-2.0
73	Weir, Mike	31.1%	28.3%	-2.8
74	Loupe, Andrew	31.1%	34.9%	3.9
75	Pettersson, Carl	31.0%	30.4%	-0.6
76	Bohn, Jason	31.0%	31.2%	0.2
77	Chopra, Daniel	30.9%	31.8%	0.9
78	Stroud, Chris	30.9%	31.4%	0.5
79	Harman, Brian	30.9%	31.7%	0.8
80	Blixt, Jonas	30.9%	29.3%	-1.6
81	Henley, Russell	30.9%	32.0%	1.1
82	Laird, Martin	30.9%	29.7%	-1.2
83	Van Pelt, Bo	30.8%	28.2%	-2.6
84	Compton, Erik	30.8%	30.7%	-0.2
85	Hearn, David	30.8%	30.0%	-0.8
86	Hoffmann, Morgan	30.8%	33.7%	2.9
87	Stuard, Brian	30.7%	32.4%	1.7
88	Casey, Paul	30.7%	33.4%	2.7
89	Byrd, Jonathan	30.6%	29.0%	-1.6
90	Ernst, Derek	30.6%	26.1%	-4.4
91	Weekley, Boo	30.6%	25.2%	-5.4
92	McGirt, William	30.5%	29.5%	-1.1
93	Love III, Davis	30.5%	28.5%	-2.0
94	Haas, Bill	30.5%	29.7%	-0.8
95	MacKenzie, Will	30.5%	33.5%	3.0
96	Koepka, Brooks	30.5%	30.7%	0.2
97	Baddeley, Aaron	30.5%	32.6%	2.1
98	Every, Matt	30.4%	36.6%	6.2
99	O’Hair, Sean	30.4%	27.5%	-2.9
100	Teater, Josh	30.4%	28.2%	-2.2
101	Romero, Andres	30.4%	32.0%	1.6
102	Huh, John	30.3%	27.7%	-2.6
103	Oosthuizen, Louis	30.3%	29.5%	-0.8
104	Overton, Jeff	30.3%	31.4%	1.2
105	de Jonge, Brendon	30.2%	28.7%	-1.6
106	Campbell, Chad	30.2%	23.3%	-7.0
107	Potter, Jr., Ted	30.2%	25.3%	-5.0
108	Thompson, Michael	30.2%	32.9%	2.7
109	Snedeker, Brandt	30.2%	29.3%	-0.9
110	Cabrera, Angel	30.2%	31.9%	1.7
111	Lingmerth, David	30.1%	31.5%	1.4
112	Maggert, Jeff	30.1%	24.8%	-5.3
113	Streb, Robert	30.1%	31.9%	1.8
114	Duke, Ken	30.1%	28.4%	-1.6
115	Aragon, Alex	29.9%	29.8%	-0.2
116	Appleby, Stuart	29.9%	33.3%	3.4
117	Castro, Roberto	29.9%	28.7%	-1.3
118	Howell III, Charles	29.9%	30.9%	0.9
119	Van Aswegen, Tyrone	29.9%	30.6%	0.6
120	Harrington, Padraig	29.9%	30.3%	0.4
121	Swafford, Hudson	29.9%	28.0%	-1.9
122	Gardiner, Scott	29.8%	29.0%	-0.8
123	Driscoll, James	29.8%	31.1%	1.3
124	Kuchar, Matt	29.8%	32.9%	3.1
125	Langley, Scott	29.8%	29.0%	-0.8
126	Flores, Martin	29.8%	28.5%	-1.3
127	Prugh, Alex	29.8%	26.8%	-3.0
128	DeLaet, Graham	29.7%	27.7%	-2.1
129	Kelly, Jerry	29.7%	29.8%	0.1
130	Kisner, Kevin	29.7%	30.2%	0.5
131	Senden, John	29.7%	30.2%	0.5
132	Yang, Y.E.	29.6%	29.2%	-0.5
133	Toms, David	29.6%	28.3%	-1.4
134	Thompson, Nicholas	29.6%	26.1%	-3.5
135	Herron, Tim	29.6%	27.1%	-2.5
136	Tringale, Cameron	29.5%	28.1%	-1.3
137	Summerhays, Daniel	29.4%	30.0%	0.5
138	Vegas, Jhonattan	29.4%	26.1%	-3.4
139	Wi, Charlie	29.4%	29.5%	0.1
140	Cink, Stewart	29.4%	28.9%	-0.5
141	Merrick, John	29.4%	25.6%	-3.8
142	Points, D.A.	29.3%	27.3%	-2.0
143	Ogilvy, Geoff	29.3%	29.6%	0.3
144	Hahn, James	29.2%	28.5%	-0.7
145	Choi, K.J.	29.1%	33.0%	3.8
146	Singh, Vijay	29.1%	25.6%	-3.5
147	Chalmers, Greg	29.1%	32.8%	3.7
148	Todd, Brendon	29.1%	31.5%	2.4
149	Knox, Russell	29.0%	28.5%	-0.5
150	Wilkinson, Tim	29.0%	29.8%	0.8
151	Immelman, Trevor	28.9%	26.8%	-2.1
152	Frazar, Harrison	28.9%	26.1%	-2.9
153	Garnett, Brice	28.9%	29.3%	0.4
154	Martin, Ben	28.9%	31.0%	2.1
155	Williams, Lee	28.9%	26.8%	-2.1
156	Mallinger, John	28.8%	25.0%	-3.8
157	Fdez-Castano, Gonzalo	28.8%	32.5%	3.7
158	Rollins, John	28.8%	26.2%	-2.6
159	Merritt, Troy	28.8%	26.0%	-2.8
160	Clark, Tim	28.7%	28.8%	0.1
161	Gay, Brian	28.7%	29.3%	0.6
162	McDowell, Graeme	28.7%	31.3%	2.6
163	Renner, Jim	28.6%	29.5%	0.9
164	Lee, D.H.	28.6%	28.2%	-0.4
165	Levin, Spencer	28.6%	27.5%	-1.1
166	Lynn, David	28.5%	30.3%	1.8
167	Hanson, Peter	28.4%	28.6%	0.2
168	Crane, Ben	28.3%	28.4%	0.1
169	Malnati, Peter	28.1%	30.1%	2.0
170	Baird, Briny	28.1%	27.5%	-0.6
171	Ogilvie, Joe	28.0%	30.3%	2.3
172	Furyk, Jim	27.9%	25.5%	-2.4
173	Jacobson, Freddie	27.9%	30.1%	2.2
174	Collins, Chad	27.7%	28.0%	0.4
175	Donald, Luke	27.7%	32.3%	4.7
176	Clarke, Darren	27.6%	23.3%	-4.3
177	Molder, Bryce	27.6%	27.2%	-0.3
178	Allenby, Robert	27.4%	26.8%	-0.6
179	Goosen, Retief	27.2%	29.0%	1.8
180	Wilson, Mark	27.2%	26.0%	-1.2
181	Durant, Joe	27.1%	24.0%	-3.2
182	Hurley III, Billy	27.1%	26.8%	-0.3
183	Gates, Bobby	27.1%	23.9%	-3.2
184	Loar, Edward	27.0%	27.4%	0.4
185	Slocum, Heath	26.9%	26.1%	-0.8
186	Hicks, Justin	26.9%	26.1%	-0.8
187	Wagner, Johnson	26.8%	25.5%	-1.3
188	Leonard, Justin	26.7%	28.3%	1.6
189	Henry, J.J.	26.6%	23.1%	-3.5
190	Woods, Tiger*	26.3%	37.8%	11.5
191	Curtis, Ben	26.3%	28.3%	2.0
192	Tway, Kevin	25.8%	25.8%	0.0
193	Ames, Stephen	25.4%	25.6%	0.2
194	Goydos, Paul	24.6%	21.1%	-3.5
195	Verplank, Scott	23.5%	21.8%	-1.7