True Runs Allowed

Mark J. Rebilas-USA TODAY Sports

Diving in to how the Rays pitchers have truly performed this season.

There's been a lot of dismay for the Rays pitching staff the last few days, especially after Chris Archer's forearm cramp handed seven innings to the bullpen that lost the game by one run.

Moreover, some pitchers have ebbed and flowed in how "hittable" they seemed on the mound. Hellickson hasn't had his stuff, David Price has, Fausto's a mystery, and it's easy to get lost in what it all means.

I'd like to take a step back and look at pure pitcher performance, and more than what FIP can tell you...

True Runs Allowed (tRA)

Imagine if FIP also knew how batters were hitting the ball. That's tRA.

In case you don't know much about the stat, here's a brief primer:

This fascinating statistic was developed by Graham MacAree in 2009 and, like FIP, isolates the pitcher from his defense by using strikeouts, walks, HBPs, and home runs. But it doesn't stop there. Values are then assigned for groundballs, line drives, fly balls, and infield flies, based on the average values of those batted ball types across the league.

The allure of tRA comes from the fact that where FIP throws out a ton of (noisey) information about the pitcher's performance, with it's own set values and weights based on pitcher performance alone. tRA uses that information, but in an intelligent, nuanced manner that is aware of the basic dynamics of DIPS theory.

tRA is measured as the number of runs a pitcher allows through 27 plate appearances, the same scale used as ERA.

What are the implications?

A great performance by tRA's standards would be a score of 3.8, and anything 3.5 or less is considered extraordinary. This season's average performance has been 4.76 tRA, which is about twenty points lower than usual.

You might have noticed tRA before in FanGraph's Batted Ball section as tERA, which correlates these "runs allowed" to an "earned" run average. tERA can be easily converted back to tRA if you divide tERA by 0.92 (re: David Appleman).

The implications are that a pitcher who allows a ton of line drives will have a much higher tRA than, say, a ground ball pitcher, because league average performance on line drives shows that more runs would be scored (on average).

This is a very similar process to FIP, but with a dose of reality. The result is a projected past performance under normal circumstances. tRA tells it like it is, showing the "true" performance of the pitcher against the standard hitter.

How it this different from SIERA?

SIERA is best used to make ERA predictions, and is the most accurate statistic in search of future performance. tRA and tERA reflect on expected previous performance.

tRA can be regressed as a predictor for immediate future performance, based on the pitcher's past performance and the league-average rates, and is available on StatCorner as xRA+ under the more detailed player pages.

***

So, how are the Rays doing?

When you take a step back and look at the whole season, the Rays starters have been absolutely incredible, sporting a 4.11 tERA this year (second only to the Tigers in the American League).

Let's dive deeper to the player level, but first, a few reminders:

- Remember that our goal is to show how many runs a pitcher should have allowed.

- If you're looking for an expected earned run average, use tERA.

- If you're looking for an expected number of runs allowed, use the final column for tRA.

Here are the numbers for this season's starting pitchers, in order of tRA. Reading across the line, you will find:

Player's name | ERA: Performance thus far | FIP: Pitcher performance under normal circumstances | tERA: What the pitcher should have been responsible for with league average hitters and defense | tRA: What should have happened with league average hitters and defense | SIERA: Expected future performance

Name ERA FIP tERA tRA SIERA
David Price 3.36 3.53 3.48 3.78 3.45
Matt Moore+ 3.41 3.70 3.69 4.01 4.16
Jeremy Hellickson 4.77 4.04 4.02 4.37 4.00
Chris Archer+ 2.71 4.21 4.04 4.39 4.52
Roberto Hernandez 4.75 4.42 5.05 5.49 3.65
Alex Cobb+ 3.01 3.52 3.97 4.32 3.16
Jake Odorizzi 6.00 4.65 5.04 5.48 4.42
Alex Colome+ 2.25 5.04 5.42 5.89 5.02

The latter three players are separated out due to their limited appearances this season. 
+ indicates a current injury, although Archer is not expected to miss any starts.

What we read with these stats is that the majority of Rays pitchers are allowing hitters to put the ball into play with weaker results than should be expected, if not for the Rays defense.

Of course there are other factors that play into ERA -- just look at Archer's meager (but excellent) average, or Hellickson's last few years -- but considering the league average pitcher performance this season is a 4.76 tRA, the Rays are showing well above average performance from every arm in the starting rotation.

The bullpen has likewise been impressive. Their 3.54 tERA is second only to the Athletics (3.33) in the American League, even with the struggles from Rodney (inconsistent delivery), McGee (pitch selection), and Peralta (sometimes ineffective).

Here are the player numbers for relief pitching:

Name ERA FIP tERA tRA SIERA
Alex Torres 0.98 1.96 1.64 1.13 2.28
Joel Peralta 3.19 3.42 3.25 3.52 3.66
Jamey Wright 3.42 3.07 3.45 3.71 3.19
Fernando Rodney 3.69 3.11 3.79 4.05 3.19
Jake McGee 4.40 3.47 3.89 4.18 2.68
Cesar Ramos 4.53 3.35 4.25 4.66 3.77
Kyle Farnsworth 5.76 4.33 5.03 5.49 3.96
Jesse Crain+ 0.74 1.51 2.18 2.37 2.50
Brandon Gomes 5.40 3.05 2.24 2.45 2.33
Josh Lueke 3.75 3.88 3.16 3.43 4.17

Major League Numbers only

This is where tRA and tERA are most telling.

Whether Farnsworth has been 5 runs bad or 5.5 runs bad is not the real discussion. He's the only pitcher in the pen that's below average over the course of the season, and despite having some use as an enforcer, he hasn't carried his weight.

The Professor was credited with two runs in Wednesday's loss, and his pitches are incredibly hit-able. Even without the implications of defense, he has the worst FIP on the team by nearly 1.00.

The Rays did their part to go out and get Jesse Crain for peanuts, but he's still on the disabled list with shoulder soreness, and the Rays could use a stop gap until he's ready.

That's what makes Lueke's possible promotion this weekend so intriguing. He's shown strong numbers in the majors this season on his brief appearance, and has torn up Durham to the tune of a 1.56 FIP with a 33.5% K-rate. Even in his brief performances this season (12.0 innings), Lueke averaged two runs better than Farnsworth when considering batted ball tenancies.

Other Observations

- Only using the Rays pitching staff's numbers from this season, I did not find any correlation between tRA and the player's K%, BB%, WHIP, BAbip, BAA, or LOB%.

- Perhaps adding SIERA to this article just creates noise, but I found it interesting to show just how poor of a predictor ERA, FIP, and tRA stats are for future performance when considering the past.

In the end, I hope you found all of this encouraging. Even when the Rays lose a pitching duel. The offense has been so dominant this season (still second in wRC+ at 112) that it's been easy to get down on the pitching staff.

As a team, the Rays hurlers have not only the second best FIP (3.74), but a more telling second best tRA (4.27), and second best tERA (3.93) in the American League.

There's a lot to appreciate in just how great the Rays have been.

More from DRaysBay:

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join DRaysBay

You must be a member of DRaysBay to participate.

We have our own Community Guidelines at DRaysBay. You should read them.

Join DRaysBay

You must be a member of DRaysBay to participate.

We have our own Community Guidelines at DRaysBay. You should read them.

Spinner

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker