The 2016 US Open
While the US Open is similar in grandeur to our first Major of the year, The Masters, it is quite different in almost every other regard. The biggest difference is in location: The Masters is played at Augusta National and has been since 1934, whereas the US Open rotates every year. The 2016 US Open takes place at Oakmont Country Club in Pennsylvania, which last hosted the tournament in 2007.
For all of our series and talk on course history, you simply have to disregard the idea of course history this week because we don’t have sufficient data. While we have the basic stats from 2007 for some golfers — Driving Distance, Greens in Regulation, Driving Accuracy, Putts Per Round, etc. — we have no newer stats and obviously nothing related to Plus/Minus or DraftKings salary. The course is important. The history there probably isn’t.
Using the Clubs You’ve Got
And yet I still want to provide this weekly article and consider the data we do have, however incomplete, scarce, and/or unreliable it may be. If we are hesitant about the data, we can always choose to ignore it as needed.
In researching, I looked at the US Open from 2007 (the last tournament at Oakmont) and then correlated the stats listed above (as well as Save Percentage) with average score per round. I initially wanted to retrodict the salaries and create theoretical Plus/Minus numbers, but I couldn’t find reliable 2007 odds information.
So instead of doing what we normally do, we’ll instead look only at correlations — but even those are a little tough to trust, given that PGA data in 2007 wasn’t recorded as meticulously as it is today.
Sometimes you’ve just got to use the clubs you’ve got.
Correlation Central
Here are the correlations:
Driving Distance: 0.0470
Driving Accuracy: -0.1887
Greens in Regulation: -0.1377
Putts Per Round: 0.1577
Save Percentage: -0.2117
Since a lower score is better in golf, negative correlations are better here. For example, putting was a ‘negative’ factor for golfers in 2007: Overall, the better a golfer was at putting, the worse he performed at Oakmont. That seems to be the general consensus from the golf community analyzing the US Open this week, so I trust its veracity a little more.
Save percentage is the most highly correlated of the statistics measured, although correlation is fairly low across the board. But even with the low correlation, what we can see in the data is that at Oakmont it is much more important to be accurate than long, as distance hurt golfers while the accuracy stats (both off the tees and hitting greens) helped golfers in 2007. Again, in this instance the data matches the common narrative, so it might be reliable.
The 18th Hole
While it was helpful to gather the data from 2007, it is important to discount course history this week. There has been some work done to Oakmont in the last nine years, and while the changes may not affect how the course plays the larger point is this: A sample size of one tournament per decade is incredibly small. Take all data from Oakmont with a grain of salt. Is it a very hard course? It’s probably safe for us to say that it is. How will recent play or some niche-y stats impact the tournament? We probably shouldn’t attempt to make such a prediction.
Focus on accuracy, but focus the most on talent and pricing. Because it’s a Major, there is a lot of line movement — and a lot of Vegas information in general — so adjust the weights of your research accordingly. I’ve claimed before that the best DFS players aren’t the ones with the most data: They’re the ones who adjust the best. Oakmont is throwing you a curveball this week. How will you adjust?
