Jonathan Judge, Harry Pavlidis, and Dan Turkenkopf over at Baseball Prospectus have developed a new pitching metric, sort of a replacement for ERA, dubbed DRA or "Deserved Run Average." Taking several cues from ERA itself, DRA hopes to provide added context for the amount of runs that a pitcher truly gives up in any given game. It does so by, among other things, taking into account many more factors than ERA does (which is simply ER*9/IP).
I won't give an in-depth explanation of the stat as Baseball Prospectus has already written two great articles on the subject, but I thought it'd be fun to give a quick and dirty rundown on this new way to look at pitching and just how it rates Indians pitchers so far in 2015.
For starters, let's talk about what DRA is on a very simplistic level.
One of the biggest issues with ERA is that it's a catch-all for runs by a pitcher, no matter the true context. If Trevor Bauer gives up a double, is taken out of the game, then Scott Atchison gives up a two-run home run, is Bauer really responsible for that base runner scoring? In short, not entirely. What should be measured is the fact that Bauer gave up a double, not that the double turned into a run scored. Being that Bauer was taken out of the game in this made-up scenario, there's no way to know that he actually would have given up that runner for a score, or if he would have finished out the inning without damage being caused. In the world of ERA, the runner is Bauer's problem and that's that. But with DRA, things are a bit different.
Deserved Run Average deals with this issue by by weighing the type of hits given up by a pitcher, not just counting a run as a run. Similar to a hitting stat such as wOBA, DRA is about giving each at-bat a specific value and elminating as much irrelevant context as possible. A home run, for example, is usually worth about 1.4 runs. According to Baseball Prospectus, this is because sometimes a home run will be hit with runners on base. Whether or not runners were on base, and whether or not the pitcher on the mound put those runners there, a home run costs a team 1.4 runs on average.
Back to the imaginary Bauer/Atchison scenario, Bauer would be knocked for only 0.75 runs. Why? Because, on average, a double will cost a team 3/4 of a run. Whether or not another batter gets that runner home, the fact of the matter is that Bauer cost the team some kind of run value, and that should be reflected in his Deserved Run Average.
Placing value on different batting events is only a small part of this new proposed statistic, so here's a quick rundown on some of the highlights of what exactly goes into it:
- Adjusting for weather, stadium and other "non-baseball" conditions. Accounting for various factors such as weather and which stadium the game is being played in is a staple in sabermetrics as this point, and DRA is no different. Game time temperature, wind and being the home or away team all account for if a pitcher gave up a home run, strikeout or other hit. While ERA (and other basic stats) skip over this, DRA accounts for all of them.
- Adjusting for the batter, catcher and umpire. What batter is actually being faced, who the catcher is and the kind of strike zone that an umpire has are also important factors to the type of event that could happen on any given at bat. Once again, DRA takes all of these into account.
- Adjusting for base runners. One of the more interesting adjustments that DRA makes is for runners on base. Part of this adjustment are two other new stats - Takeoff Rate Above Average (TRAA) and Swipe Rate Above Average (SRAA). These two stats measure the effectiveness of a pitcher when it comes to holding base runners, both in terms of the effectiveness of base stealing attempts, and how much they runners take off, whether or not they succeed. In essence, a low TRAA and/or SRAA says that base runners are leery about running on any given pitcher, while a higher values points to a pitcher being taken advantage of by base runners.
- Based on BF, not IP. In order to weed out pitchers who allow a lot of base runners then get lucky by leaving them stranded, DRA is based on the number of batters a pitcher faces, not just a flat number of innings pitched.
- Scaling with RA/9, not ERA. Being that Earned Run Average involves some level of variation depending on what the scorekeeper on any given night deems an earned or unearned run, DRA removes this variance by skipping ERA entirely and scaling to runs allowed per 9 innings. This could be up for some debate, but according to Baseball Prospectus, "we understand that ERA is what many of you are used to, but once you get over that, you’ll be much happier."
There's more detail in the BP link above.
So, how does DRA rate Indians pitchers? So far in 2015, Trevor Bauer is by far the best Indians pitcher in terms of his Deserved Run Average. As of April 29, Bauer is 9th out of a qualified 442 pitchers with a 2.99 DRA.
Of the other top-10 DRA pitchers, Bauer has the third highest ERA. In part, this means that he's been unfairly assigned more runs than others in the top 10 (except for Jake Odorizzi, wow!), which would make sense given the frequent collapse of the Indians bullpen, and how many base runners Trevor usually leaves on base.
Corey Kluber is the next Tribe pitcher, down at #32 with a 3.31 DRA, and he has an even worse case of being unfairly assigned runs. His 4.24 ERA is a major stand-out among other pitchers around the same DRA range, most of whom are still right around 2.00 to 3.00 earned runs.
Here are all the Indians pitchers, both starters and relievers, and where they rank amongst the other pitchers in the league (as of April 29). Baseball Prospectus hasn't yet released their SRAA data, but I've included each pitcher's TRAA. Keep in mind that a lower TRAA rating is better, where 0% would be considered average. The number rank is where they place with their DRA rating out of the 442 qualified pitchers.
Deserved Run Average ratings more or less confirm what we already know about our wonderful Tribe staff. The Indians are generally getting good value of their starters, while the bullpen is in total disarray.
It's a little surprising seeing Salazar with such a high DRA compared to his ERA, but when you consider that he has left a staggering 98.7 percent of hitters on base, it makes sense. Remember, DRA takes these abandoned runners into account by assigning a value to all those hits, even though they may never result in a run. Trevor Bauer also leaves a large number of runners on base (roughly 85%), but to this point in 2015 he's given up fewer extra base hits than Salazar, making for a lower DRA.
With a stat such as DRA, it's important to keep in mind that this could prove to be a great indicator of past value, but it will never be a tool to predict a pitcher's future performance, the way a stat like FIP can be used. Deserved Run Average is meant to tell you just how many runs a pitcher has given up for his team, while including as many relevant factors as possible (while also weeding out the ones that aren't important).
All statistics courtesy of Baseball Prospectus and FanGraphs