What is DRA (Deserved Run Average)? Definition, Formula, and Example
DRA (Deserved Run Average) is Baseball Prospectus's run-prevention metric that estimates how many runs a pitcher deserved to allow per nine innings after stripping out the influence of defense, framing, park, and opponent quality.
What is DRA?
DRA, or Deserved Run Average, is Baseball Prospectus's flagship pitching metric. It estimates how many earned runs a pitcher *deserved* to allow per nine innings, scaled to look like ERA so a DRA of 3.50 reads like a 3.50 ERA. The difference is what DRA removes: the quality of the defense behind the pitcher, the catcher's framing, the ballpark, the strength of the opposing hitters, and even the umpire. What is left is the pitcher's own contribution to run prevention. Lower is better, and unlike raw ERA, DRA does not punish or reward a pitcher for the team he happens to play in front of.
How DRA is Calculated
DRA is not a simple closed-form equation like FIP. It is built from a series of mixed (multilevel) statistical models that assign credit and blame for every event on a plate-appearance level. Conceptually:
DRA = runs attributable to the pitcher, per 9 IP, after controlling for defense + framing + park + opponent + umpire + base-out state
Each plate appearance is run through models that estimate the pitcher's responsibility for the outcome, the results are converted into a run value, and the total is scaled to a runs-per-nine figure on the ERA scale. Baseball Prospectus also publishes DRA- (DRA minus), an index where 100 equals league average and every point below 100 is one percent better than average — so a DRA- of 75 means the pitcher was 25% better than league average at preventing deserved runs. DRA- is park- and league-adjusted, making it the cleaner number for comparing pitchers across eras and ballparks.
Worked Example
Take Tarik Skubal's 2024 American League Cy Young season for the Detroit Tigers. He posted a 2.39 ERA with a 2.49 FIP, 228 strikeouts, and elite walk avoidance. Because his peripherals were as dominant as his run prevention, his DRA landed close to his ERA, with a DRA- in the low 60s — roughly 35-40% better than league average. Compare that to a pitcher with a shiny 3.00 ERA propped up by a great defense and a pitcher-friendly park: DRA would peel those advantages away and likely show a number well above 4.00, exposing the ERA as partly a product of context rather than skill. Chris Sale's elite 2024-25 run similarly produced a DRA- in the high 70s, marking him as well above average even when raw ERA fluctuated.
Why DRA Matters
DRA is one of the most context-complete pitching metrics available, which makes it valuable for front offices projecting future performance and for fantasy players trying to separate true talent from lucky run-prevention. Because it is more predictive and less noisy than ERA, it is a strong input for evaluating whether a breakout is real or a mirage created by a great defense and a forgiving park.
Limitations and Common Misconceptions
DRA is a "black box" relative to FIP — you cannot compute it on a napkin, and its values shift as Baseball Prospectus updates its models, so historical numbers can change retroactively. It is descriptive of deserved runs, not a perfect predictor of the future; SIERA and xERA attack a similar problem from different angles. It is also frequently confused with FIP, but FIP uses only strikeouts, walks, hit-by-pitches, and home runs, while DRA models every batted ball and contextual factor.
Related Terms
In Legends Deck: pitcher cards lean on context-adjusted metrics like DRA to set their core run-prevention rating, so an ace who quietly outperforms his ERA earns a higher card value than a pitcher riding an elite defense — your simulated staff is graded on deserved skill, not surface stats.