Last week, I wrote an article about optimizing a player model for the specific golf tournament being played. Here’s what I wrote in the intro:
Often, PGA DFS players hear something along the lines of, “Course X is a distance course.” On our PGA Daily Fantasy Flex podcast, we make similar statements. I believe we provide context and degree to those statements, but some analysts don’t. That’s a problem, because degrees matter when you’re building models to predict player performance for a golf tournament. Course X favors distance — but how much? Should you weight Long-Term Driving Distance 20/100 in Models? 30/100?
Thankfully, we can (sort of) deal with this problem, if we’re willing to be a little creative. Our Models and Trends tool have the same metrics, and thus we can use the latter to test for weighted metrics to use in the former. And that’s exactly what I did: I took every metric within Models, calculated how golfers in the top 20 percent of that metric performed historically at this week’s course, Quail Hollow, and then assigned weights. This is an imperfect science for many reasons, one of which is that some metrics have negative value. For example, golfers with a Long-Term Bogey Percentage in the top 20 percent have historically scored a -11.00 Plus/Minus. That is less than the baseline for all golfers at Quail Hollow (-6.23), and thus I eliminated it from the Model. Ideally, we would weight metrics both positively and negatively; since I can’t do that, I used only metrics that tested above baseline and built a model accordingly.
Here’s the Plus/Minus and metric/baseline differential for each positive metric at Quail Hollow:
- Pro Trends Rating: +3.28, +9.51
- Recent Birdie Percentage: +2.61, +8.84
- Course Greens in Regulation: +0.14, +6.37
- Long-Term Adjusted Round Score: -0.85, +5.38
- Course Driving Distance: -1.26, +4.97
- Adjusted Round Differential: -1.42, +4.81
- Recent Adjusted Round Score: -2.03, +4.20
- Long-Term Driving Distance: -2.14, +4.09
- Recent Missed Cut Percentage: -2.25, +3.98
- Course Adjusted Round Score: -2.28, +3.95
- Long-Term Adjusted Round Score: -2.43, +3.80
- Long-Term Field Score: -2.51, +3.72
- Recent Driving Accuracy: -2.61, +3.62
- Long-Term Eagle Percentage: -2.94, +3.29
- Greens in Regulation Differential: -3.04, +3.19
- Recent Driving Distance: -3.27, +2.96
- Course Count: -3.36, +2.87
- Long-Term Birdie Percentage: -3.42, +2.81
- Putts Per Round Differential: -3.79, +2.44
- Recent Scrambling: -4.35, +1.88
- Recent Eagle Percentage: -4.42, +1.81
- Recent Bogey Percentage: -4.45, +1.78
- Recent Greens in Regulation: -4.87, +1.36
- Long-Term Missed Cut Percentage: -5.05, +1.18
- Recent Field Percentage: -5.07, +1.16
- Long-Term Driving Accuracy: -5.49, +0.74
- Long-Term Count: -6.17, +0.06
From here, I translated all of those Plus/Minus values into percentages. For example, with a +2.61 Plus/Minus, Recent Birdie Percentage is given a weight of 9/100 in the model because of how that value compares to the Plus/Minus values of the rest of the metrics. With a Plus/Minus of -3.27, Recent Driving Distance will be given a weight of 3/100 in the model. And so forth; using the positive metrics, we not only have an idea of what has historically been important at Quail Hollow but also how important it has been relative to other metrics.
This model backtested well with a Plus/Minus of +5.50 (screenshot via Models):
Quail Hollow vs. Other PGA Championship Venues
This week is particularly unique because, while some tournaments like the U.S. Open and the PGA Championship routinely rotate courses, we still have a ton of data on Quail Hollow since it has hosted another PGA Tour event, the Wells Fargo Championship, since 2003. Thus, people can research two different data points in terms of historical course data: Quail Hollow versus the last couple years of the PGA Championship at other courses. I assumed people would research data on the Wells Fargo and Quail Hollow specifically, but I’ve actually seen quite a mix among golf analysts online. If that’s the case, there could be a big edge.
To look at this, I performed the exercise above with PGA Championship data. Here’s how the metrics fared and how they compared to Quail Hollow:
As a general theme, while the U.S. Open is known for highlighting ball-strikers and accurate players, the PGA Championship has historically rewarded longer players. And that’s still true: In both models, Long-Term Driving Distance backtested above baseline to the tune of a +4.09 Plus/Minus at Quail Hollow and a +3.28 Plus/Minus at other recent PGA Championship venues.
That said, I think that ball-strikers — if they are long enough — still fit the course. Quail Hollow is going to favor bombers since it plays at nearly 7,500 yards, but degrees matter, so while distance still reigns supreme at Quail Hollow it isn’t as if ball-striking is unimportant. As shown in the above table, bogey avoidance and accuracy off the tee did not backtest well at recent PGA Championship venues, but it’s been important historically at Quail Hollow. You want bombers here, but mostly you want guys who have upside in both distance and ball-striking.
For example, take Paul Casey, a mid-priced golfer: He isn’t an especially long player: His 295.1-yard Long-Term Driving Distance is a little above average in this field. Still, he is one of the highest-rated golfers in this model because of the degree to which we weight distance versus other metrics. It’s one of the most important metrics, but Casey is long enough and strong enough in other ways that he rates as one of the best plays of the weekend per the Quail Hollow model I built. Hideki Matsuyama is another example of someone who doesn’t have the distance of, say, Dustin Johnson, but is long enough and dominant enough in other metrics, especially his Greens in Regulation and Birdie Percentage. Having a well-rounded golfer will be key.
But I don’t want to overstate accuracy versus distance this week. This model is flooded at the top with guys who are long off the tee; the ones who are also accurate just get a bump up, while guys who are slightly above average in distance and elite at ball-striking stand out as well.
For whatever it’s worth, this specific model’s top golfers currently are Tony Finau and Matsuyama.
