What is Deserved Run Average (DRA)? Definition, Formula, and Example
Deserved Run Average (DRA) is a pitcher evaluation metric created by Baseball Prospectus that estimates the runs a pitcher deserves to allow per nine innings using a mixed model that accounts for count, base, and quality of contact.
What is Deserved Run Average?
Deserved Run Average (DRA) is a comprehensive pitching metric developed by Baseball Prospectus that estimates the number of runs a pitcher truly deserves to allow per nine innings. Unlike ERA, which is heavily polluted by defense and luck, DRA uses a complex statistical model to isolate the pitcher's specific contribution to run prevention. It factors in the count, the base-out state, the type of contact allowed, and the quality of the opposition, returning a single number scaled to look like ERA.
How DRA is Calculated / Measured
DRA cannot be calculated with a simple pocket formula; it relies on a proprietary mixed-level statistical model. The calculation process works as follows:
1. Modeling Run Expectancy: DRA looks at every pitch outcome and assigns a run value based on the count (e.g., a line out in an 0-2 count is less damaging than a line out in a 3-0 count) and the base-out state.
2. Pitcher Isolation: It separates the pitcher's contribution from the defense, the catcher, and the park environment using a mixed-effects model (similar to those used in advanced academic research).
3. Contact Quality: DRA incorporates exit velocity and launch angle data to evaluate how hard balls were hit, rewarding pitchers who induce weak contact.
4. Scaling: The final output is translated into a per-nine-innings number calibrated to the league-average run environment so it reads like a traditional ERA.
Worked Example
Consider Kevin Gausman's 2023 season. Gausman posted an excellent traditional ERA of 3.16. However, his DRA that year was 2.96.
The gap between the two numbers tells a specific story: Gausman's underlying peripherals—his massive strikeout rate, low walk rate, and the quality of contact he allowed—indicate he actually *deserved* to give up even fewer runs than his ERA showed. Conversely, a pitcher like Jordan Montgomery had a 3.20 ERA in 2023, but his DRA was 4.41. The DRA model saw that Montgomery relied heavily on his defense to convert balls in play into outs, and his true, isolated skill level predicted much worse run prevention moving forward.
Why DRA Matters
DRA is one of the most predictive pitching metrics available. Because it accounts for count context and contact quality, it is more stable year-over-year than FIP or xFIP. Front offices use DRA to identify pitchers whose traditional ERAs are propped up by elite defense or sheer luck. For fantasy baseball managers, DRA is a leading indicator: a pitcher with a low DRA but a high ERA is a prime buy-low target, as his underlying skills suggest positive regression.
Limitations and Common Misconceptions
The main limitation of DRA is its opacity. Because it relies on a proprietary mixed model rather than a public formula, users cannot calculate it themselves or easily audit the math. It is also a "black box" metric, meaning analysts must trust Baseball Prospectus's internal weighting. DRA is often confused with FIP, but FIP ignores contact quality entirely, while DRA embraces it. Additionally, DRA can sometimes undervalue elite groundball pitchers who specialize in inducing double plays, as the model struggles to perfectly assign credit for defensive execution.
Related Terms
In Legends Deck: Legends Deck pitcher cards leverage DRA-style principles to separate true stuff from defensive luck. Our simulation engine models count leverage and contact quality independently, so a pitcher with an elite DRA translates into a dominant Legends Deck card, even if his real-world ERA was artificially inflated by a poor defense.