tRA: Because you pitched better, and you deserve better.

Have you ever watched a pitcher pitch a game, and walk out with an ERA of like 6, and you said to yourself "He did not pitch all that badly, what on earth happened there?" You figure he let a couple baserunners on via walk, struck out a good number, only sprayed a few singles, how on earth did he let 5 runs cross?

Well, we turn to FIP then, a pitching metric I introduced in an earlier Counting Rocks, which accounts for the elements of a game that a pitcher directly impacts, strikeouts, walks, and home runs. It's a pretty simple formula, I can do the math on my phone calculator.

Based on ERA, K9, BB9, HR9, and FIP, here are the pitchers we've played this season:

Now just looking at these numbers, we could say "But Glendon Rusch had a below-league-average FIP, why would we let him go?!" Well, the pure stats guy might say that his FIP says he's good, and just make the assumption that defense let him down, BABIP screwed him, etc etc. But we all know that Rusch was a very hittable pitcher this year, despite his good K rate and low BB rate. How hittable though? Did he really deserve that ERA?

The problem is that a pitcher really can control more than just his K's, BB's, and HRs, but to a lesser extent than those 3 true outcomes. FIP still has a lot of use in looking at the more fundamental aspects of what a pitcher does, but it might not be the fairest indicator of a pitchers' performance.

So FIP clearly doesn't have all the answers. If it did, Manny Corpas would still be a downright solid pitcher for us, and Franklin Morales would be in the Springs still. Leaving a ball up in the zone is most definitely controllable by a pitcher, and while the end result is a far more variable event than a "true outcome," it's pretty obvious that the hittable pitchers are going to give up more runs than the not-as-hittable ones.

Because that variability is still going to be a factor in ERA, I still think that it's an inadequate measure of a pitcher's performance in a game. For example, Jason Hammel got more or less screwed by poor defense on Tuesday night, but it's not as if every ball in play past the Barmes error was an error as well. So all the runs were unearned, but as jrockies and Silverblood were commenting, some runs were more unearned than others.

The problem with ERA is that some pitchers are being rewarded or punished for events that aren't their faults. Should we say "good job Marquis" of Dexter dives for a blooper and makes the final out with the bases jacked? Should we curse Alan Embree if he gives up a similar blooper that falls between the two fielders? Probably not. ERA seems to almost be a team statistic, because a run that isn't a home run will involve the team rather than just the pitcher alone. Aaron Cook's runs and outs will come differently than those from Jorge De La Rosa, based on how they pitch and how the team reacts to how they pitch.

There needs to be a "should have been" ERA so we can compare what he actually did and what actually happened.

Welcome to tRA.

Join me after the jump and we'll get into it.

tRA was formulated by Graham MacAree, who is a writer for the SBNation blog Lookout Landing and a moderator at USS Mariner.

tRA is based on the concepts that wOBA is based on. Each offensive event is worth a certain amount of runs, and each defensive event is worth a certain numbers of outs. Tom Tango gave us this data by poring through years of game data and determined how many outs and how many runs occurred after each event. He also formulated how many runs each base/out situation was worth. Using that matrix there will give you a good idea on what run expectancies you should anticipate based on how many are on and how many outs are remaining.

For example, a strikeout is worth 1 out. A home run is worth 0 outs. But how many outs is, say, a fly ball worth? Based on 2008 data, a fly ball (to the outfield) resulted in an out 83% of the time, so we can infer that an outfield fly ball is worth 0.83 outs. However, just because an OFB is worth 0.83 outs does not mean that it's worth 0.17 runs. An OFB, in fact, is worth only 0.046 runs.

So now that we've established that each play is worth a certain number of runs and outs, we can go from there to begin to calculate the tRA.

Using the expected number of runs that a pitcher gives up, based on the run expectancies, and the number of expected outs the pitcher records, we can calculate an analogue to ERA by taking the expected number of runs per expected out (runs/outs) and multiply by 27.

So you're thinking "well that's great, you found ERA again" - and I don't blame you. But here's the thing about tRA that makes it great - it puts the pitcher in a neutral environment (because there are many park-adjustment factors to account for), and puts a neutral defense behind them (because all the run/out expectancies are based on the entire league's performances, so therefore it displays as an average rating), and rather than being the report card for how many runs the team gives up during a pitcher's outing, it actually rates the pitcher on what they let happen.

So similar to FIP, tRA is good as an indicator of how lucky/unlucky a pitcher is getting, in terms of bad defense, bloops, etc etc. But this time, we're not just nullifying hits, but rather, we're accounting for everything that happens, and giving credit to the pitcher for everything he does during the game. It's a better report card of what happened during a pitcher's outing.

So let's look at what tRA means to some of the Rockies' pitchers.

Again, what does this all mean?

Well, the thought is that while Ubaldo Jimenez is sporting a decent 4.30 ERA, were he pitching for the Neutral Planet Neutrals in Neutral Park with Neutral Defense behind him and the Neutral Neutralities batting against him, he'd have a 3.47 ERA. What Ubaldo has allowed to happen through his pitching will allow 3.47 runs per 27 outs. You'll also notice that tRA thinks Matt Daley is a stud so far, and it further solidifies the excellent start that Jorge De La Rosa has had.

So what this breaks down to is we've quantified the concept of "the box score may not show it, but [Pitcher] pitched a solid game".

How can we use it going forward? When we see a pitcher go through some struggles, we can still look at his tRA and see how it corresponds with his FIP and his ERA. We can now say "well, what he's done would give him an ERA of X.XX, but instead he actually has an ERA of Y.YY" and then we can all shake our hands miserably at the defense or the BABIP or whatever we want to blame.

tRA and wOBA won't ever be as easy to use as FIP and OPS, just because they involve a spreadsheet or internet access to get a handle on, while FIP or OPS can be eyeballed or at least figured out on that cheesy calculator in your cellphone. However, with the extra effort, they are numbers that can be taken more at face value rather than having to explain all the caveats and details that need to be considered in OPS and FIP.

There are a good number of metrics on the market anymore, and deciding which to use is difficult. I'm partial to Tango's work, because it's all based on actual occurrences, actual run totals, and real numbers. The metrics based on his studies are the most encompassing and include the most detail, and I'm pleased to be able to present them here on Purple Row.

A lot of my information was taken from http://www.statcorner.com, where you can get the current tRA data for not only major league pitchers, but also for the minors down through A ball (Asheville as well as Modesto).

That's all for this week, fair RowBots. I hope to see us citing wOBA's and tRA's sooner than later when complaining about how hittable Aaron Cook has been this season.