Injury Risk Model
The DVS Injury Risk (IR) Model is a statistical based approach to predicting the chance of injury for a given pitcher.
Now you can add the IR model results to your team’s suite of analytics. One-off analyses can be performed as well. Please contact the DVS Baseball Analytics team to learn more.
1.7 Billion
In the last five years, MLB teams have paid over $1.7B to pitchers while they were on the injured list.
22,000 Days
In 2019, over 22,000 days were spent on the injured list by pitchers. In terms of seasons, this is equivalent to the complete pitching staff of 10 MLB teams, a third of the league, missing the entire season.
696 Innings
Pitchers with a DVS Score of 10 or less have an average time to first major injury of 696 innings.
IR Model
The DVS Injury Risk (IR) Model is a statistical based approach to predicting the chance of injury for a given pitcher. DVS Score alone is a good indicator but we have added more reliability with the additional inputs of injury history, usage, position and other biographical data. Version 1 of this model leverages survival analysis of a Cox proportional hazards model. Machine learning was utilized to train the model with data from over 400 pitchers ranging in birth years from the 1920s to the 1990s. This model is continually learning as it is now comprised with data from over 1,000 pitchers. Version 2 of the model is in development with the addition of several more inputs and more training data.
IR Model Performance
The power of our predictive results are currently measured through two primary statistical measures: The Brier Score and Concordance.
Brier Score
The Brier score is a measure of the accuracy of a probabilistic function (in this case, the probability of sustaining a major shoulder or elbow injury). The closer the Brier score is to 0, the more accurate the model predictions. When assessing the ability of the model to predict major throwing-related injury, the overall Brier score is 0.134, which is 46% better than flipping a coin.
Concordance
The Concordance yields a measure of predictive
performance for making decisions between pairs of
pitchers. The Concordance for the DVS Model is 63%,
which means that when comparing two pitchers the DVS
Model is able to accurately predict the pitcher who will
undergo a major throwing-related injury first 63% of the time. This is approximately 2-to-1 odds of making the right decision between two pitchers.
IR Model Player Analysis
In mid-2018, the IR model identified Michael Fulmer as being ‘high risk’ with a 21.4% chance that he would be injured in his next 200 IP. Fulmer had Tommy John surgery in March, 2019. He last pitched in September, 2018, about 50 innings after the study. Major risk factors that drove the risk high for Fulmer were Position at Foot Strike, injury history, arm swing, and finish.