Analytics have been taking over the sports world over the past couple decades. Statisticians have been employed by most major sports franchises to crunch numbers and calculate stats such as Wins Above Replacement (WAR) in baseball, point differentials when a player is on/off the court in basketball, and when it makes sense to go for it / kick a field goal / or punt in football.
Golf is no different. Thanks to the addition of Shotlink data to the PGA Tour, every shot hit in a tour event is tracked in a detailed manner. Shotlink answers all of the following questions:
What lie was the ball hit from (Fairway, Rough, Tee, Sand, Putting Green, Recovery)? What kind of lie did the shot end up in?
How far was the shot from the hole? How far did the shot go? How far off target was the shot?
All of this data can be aggregated and used to study golf decisions and golf shots from a new perspective. The trending statistics being used on PGA Tour broadcasts are Strokes Gained. If you aren’t familiar with Strokes Gained, you can check out my primer here. The basic gist of Strokes Gained is quantifying the quality of a shot vs. a reference. In this case, the reference is how close the PGA Tour averages from a similar distance / lie. If you want an in depth explanation of Strokes Gained, I highly suggest reading Mark Broadie’s book Every Shot Counts.
If you aren’t into numbers or stats, then ignore the words and look at the pretty plots. Otherwise, its about to get nerdy…
For the past couple months I have been using Game Golf to track my golf game. It allows me to geo-tag every golf shot that I hit. It records how far I hit each shot, which club I used, and my gps location. This can effectively be used as Shotlink data for an amateur. The only place where it falls short is for putting. GPS is only accurate for 4-5 yards or 12-15 feet. This is not accurate enough to calculate Strokes Gained for putting, where a difference of a couple feet could be significant in terms of expectations to make the putt. For an idea of how distance affects the number of shots needed to hole out, check out the plots in my other post here. Due to the lack of accuracy for putting, I typically pace off my putts and just record how many feet I had for each putt. It becomes part of my pre-shot routine and I can typically pace it off while other players are reading a putt in order to minimize anyone waiting on me.
The first plot in this post shows my total Strokes Gained for the 9 rounds I have played thus far with Game Golf. The lines are split by rounds I have played in GCAT tournaments and those that were just for fun. It is obvious from the plot that I have played better in non-tournament rounds. There could be lots of reasons why I might play worse in a tournament… Nerves, playing golf courses I’m less familiar with, more difficult courses, poor weather, etc. There could be lots of reasons why I might play better or worse in any particular round, however aggregating the data allows us to look for trends. In this case, the question I want to answer is.. why am I playing worse in tournaments? More specifically, which part of my game is struggling?
The easy way to answer this is instead of aggregating the Strokes Gained data together for a Strokes Gained Total, splitting up the Strokes Gained to look at various parts of my game. For this first analysis, I will split my Strokes Gained into 4 categories. The categories are:
Strokes Gained Tee -> All shots hit from a tee area.
Strokes Gained Midgame -> All shots greater than 70 yards, not hit from a tee.
Strokes Gained Shortgame -> All shots less than 70 yards, not on the green.
Strokes Gained Putting -> All shots from the green.
I drew the line at 70 yards for shortgame vs. midgame as I hit my highest lofted wedge ~85 yards on a full swing. Therefore, anything less than 70 yards is some sort of reduced swing or short game type of shot. Let’s start with shortgame. How does my short game compare in tournaments vs. non-tournaments?
The first thing I notice is that my shortgame is pretty good. In fact, I am gaining strokes on the average PGA Tour shot in this category which is great. The second thing that I notice is that my tourney play is actually better in this area. That may be because in a tourney I am more likely to grind over these shorter shots to try to save par. It matters more, so I pay more attention. Next lets look at Midgame, or basically iron play.
The data for Midgame looks very consistent across all the rounds. The exception is my latest round at The Cliffs where I did not strike the ball particularly well. Otherwise, the differences between Tourney and Non-Tourney play are subtle. I do not think there is quite enough data yet to be concerned with my midgame play. So let’s move on to my play off the tee.
The round at Kiawah – Turtle Point certainly shows a difference off the tee. It seems early in the season I struggled off the tee regardless of Tourney / Non-Tourney, but things are improving. Not overly concerned here. That leaves putting…
Bingo! From the first plot that showed the difference in Total Strokes Gained we could see that there was a 3-5 stroke difference between my tournament and non-tournament performance. 3-4 of those strokes can be directly attributed to putting! In my non-tournament rounds I am putting almost as good as a tour pro, however in my tournament rounds I am losing 3-4 strokes. Being that the sample size is small, I can say for sure that this is because of poor putting from close range and not making enough in the 6-10ft range. I have missed several putts inside 6 feet in each of these rounds, which referencing the chart showing what a typical pro would do, means I am losing over 0.5 strokes each time that happens. I will give myself somewhat of a break for now as it is early in the season the greens haven’t been spectacular. Even at Kiawah Island – Ocean, the greens were bumpy this early in the year and I felt like I had several good putts bounce off line from short range.
In any case, this cursory glance is already giving insight into the difference between tournament and non-tournament rounds. The data can be manipulated and further broken down to draw further conclusions on how I do with certain clubs or certain distances. We will tackle that another day.