Determinants of Scoring at the Wyndham Championship
This analysis is based on scores and stats from individual rounds in the last 10 Tour events at the Wyndham Championship: 4,615 rounds in total.
Section 1: Absolute Correlation Coefficients with Score
SGTee: The correlation between Score and SGTee remained relatively low across the years, suggesting that performance off the tee does not have a strong direct impact on the final score at the Wyndham Championship. This could indicate that players who score well do not necessarily excel due to tee performance but through other aspects of their game.
SGApp: The correlation for SGApp shows some variation but generally has a moderate impact. In particular years, a higher correlation indicates that approach shots more significantly influence the scoring when accuracy into the greens can be pivotal.
SGATG: The Strokes Gained Around the Green displays a varied correlation across the years, sometimes peaking, which highlights that in specific conditions, the ability to save shots around the green can impact scores greatly.
SGP: This metric has consistently shown a significant correlation with scores, particularly in recent years. This suggests that putting is a critical component of scoring well in the Wyndham Championship.
Driving Distance: The correlation of Driving Distance with Score is relatively low, indicating that merely hitting the ball far does not correlate strongly with better scores here.
Driving Accuracy: Over the years, Driving Accuracy shows a varying correlation, with some years demonstrating a stronger impact on scoring. This indicates that in these years, keeping the ball in play off the tee was more crucial, possibly due to course setup or weather conditions affecting play.
Greens In Regulation (GIR): Consistently one of the more significant traditional metrics, GIR's correlation with Score suggests that reaching greens in the regulation number of strokes is a strong indicator of better performance at the Wyndham Championship.
Scrambling: This metric, representing the ability to save par after missing the green, varies year by year but can significantly impact scores, especially in years where GIR is more challenging to achieve.
PPGIR: This metric indicates a fairly consistent correlation, reinforcing that once on the green in regulation, the ability to putt well is essential to maintaining or improving scores.
Par 3: The correlation with Par 3 performance shows variation, suggesting that scoring on these holes can significantly impact overall performance in specific years. This may have been due to changes in course setup, such as pin positions or wind conditions.
Par 4: Consistently showing a stronger correlation, Par 4 performance often dictates scoring outcomes at the Wyndham Championship, reflecting their importance and prevalence on the course.
Par 5: The correlation here tends to be moderate..
Section 2: Partial Dependence Plots against Score
Partial dependence plots (PDPs) are a tool used in machine learning and statistical modeling to illustrate the relationship between a target variable and one or more feature (e.g. SGApp, SGATG, DrivingDistance, GreensInRegulation). They show the marginal effect of a feature on the predicted outcome of a model. PDPs are particularly useful for understanding how individual features impact the target variable, allowing for better interpretation and insights from the model.
In determining the value of Score, PDPs can help visualize how changes in each feature impact the predicted score, holding other features constant. This can provide insights into which features are most influential and how they affect the score.
Section 2 - Partial Dependence Plots
SGTee: The PDP for SGTee shows a slight negative relationship with score, indicating that as players gain more strokes off the tee, their scores tend to improve (lower scores). However, the effect is not strong, suggesting that while important, SGTee is not the primary factor affecting scores.
SGApp: This metric shows a more pronounced negative relationship with score. Players with higher strokes gained on approach shots generally achieve better scores. This highlights the importance of siding with players with strong approach play at the Wyndham Championship.
SGATG: The PDP for SGATG shows some variability but generally a slight negative trend, indicating that strokes gained around the green can contribute to better scores.
SGP: There is a noticeable negative relationship between SGP and score, underscoring the importance of putting. This reinforces the idea that good putters have a significant advantage in this tournament.
Driving Distance: The PDP shows a minimal impact of driving distance on scores, aligning with the correlation findings. Focus less on this metric when predicting the scores at the Wyndham Championship, as it doesn't strongly influence scoring.
Driving Accuracy: A slight negative relationship exists, suggesting that higher accuracy can marginally improve scores. This indicates that while not the most critical factor, players who keep the ball on the fairway may have a slight edge.
Greens In Regulation (GIR): The PDP indicates a clear negative relationship with score, highlighting the importance of hitting greens in regulation. This a key focus for score prediction.
Scrambling: This plot shows a variable impact, but better scrambling tends to relate to better scores.
PPGIR: A negative trend is evident, indicating that players with fewer putts per green in regulation tend to have better scores
Par 3: The PDP for Par 3s shows a moderate relationship, where better performance on these holes is associated with improved scores.
Par 4: A strong negative relationship indicates that performance on Par 4s is crucial for scoring well.
Par 5: The PDP shows a moderate impact, with better performance on Par 5s leading to improved scores. While not as critical as Par 4s, these holes offer scoring opportunities that players can leverage.
Section 3: Importance of Each Metric in Determining Score
Random Forest Regressor and Feature Importance
Random Forest Regressor is an ensemble learning method that constructs multiple decision trees during training and outputs the average prediction. It combines the predictions of several models to improve accuracy and robustness.
Feature importance is a technique used to interpret a machine learning model. It refers to the score that quantifies the contribution of each feature to the prediction made by the model.
In a Random Forest, the importance of a feature is computed by looking at how much the feature decreases the impurity (e.g., variance for regression tasks) across all the trees in the forest. The more a feature decreases the impurity, the more important it is considered.
The calculated importance scores for all features are then normalized to give relative importance as a percentage. This shows the relative contribution of each feature to the prediction task.
Interpreting Feature Importance
Features with high relative importance percentages have a strong impact on the model's predictions. They are crucial for accurate predictions and indicate key areas where performance matters most.
Features with low relative importance have a minimal impact on the model's predictions. While they can still contribute, they are less critical.
SGTee (17.99%): The importance of SGTee is lower compared to the PGA Tour average of 24.82%. This suggests that while tee performance is crucial, it may not be as decisive at the Wyndham Championship, where other skills are more impactful.
SGApp (36.24%): This metric holds the highest importance among SG metrics, significantly higher than the PGA Tour average of 26.77%. It highlights the significance of approach shots, indicating that players who excel in this area are more likely to perform well, aligning with the course's demand for good ball-striking skills.
SGATG (25.16%): The importance is similar to the PGA Tour average of 24.44%, reinforcing the relevance of the short game in scoring well, especially in situations requiring recovery around the greens.
SGP (20.60%): The importance of putting is slightly lower than the PGA Tour average of 23.98%, yet it remains a crucial component for success. Good putters can still have a substantial advantage, though approach play appears more decisive here.
Driving Distance (10.17%): This is slightly higher than the PGA Tour average of 9.31%, indicating some importance in driving distance, yet not a primary factor.
Driving Accuracy (8.08%): The importance is higher than the PGA Tour average of 3.77%, suggesting that keeping the ball in play has a greater influence at the Wyndham Championship.
Greens In Regulation (GIR) (15.07%): This is significantly lower than the PGA Tour average of 29.77%, indicating that while reaching greens is important, it is less critical compared to PPGIR at this event.
Scrambling (10.45%): Similar to DrivingAccuracy, its importance is lower than the PGA Tour average of 27.02%.
PPGIR (56.23%): This metric dominates in importance, far exceeding the PGA Tour average of 30.13%. This indicates that once players reach the green, putting becomes the most decisive factor for scoring well.
Par 3 (16.78%): The importance is similar to the PGA Tour average of 17.32%, indicating that performance on Par 3s is relatively consistent with broader trends.
Par 4 (68.93%): Significantly higher than the PGA Tour average of 67.12%, Par 4 performance is crucial at the Wyndham Championship. These holes often dictate scoring outcomes, aligning with their prevalence and difficulty.
Par 5 (14.29%): The importance is slightly lower than the PGA Tour average of 15.56%, suggesting that while scoring on Par 5s can be advantageous, it is less critical compared to Par 4s.
Top 5 Ranked Players - 2024 Wyndham Championship
The table below shows the top-5 ranked players and their average estimated scores from the three different Random Forest models above.
| Player | Score |
|---|---|
| Luke Clanton | 67.18 |
| Davis Thompson | 68.05 |
| Mark Hubbard | 68.07 |
| Erik Van Rooyen | 68.08 |
| Michael Kim | 68.09 |
Estimated scores for all players can be found here.