Royals' bad bullpen is too expensive

Wherefore art thou Soria?Relief pitching is the most volatile commodity in modern big-league baseball. Year in, year out, teams struggle to find reliable arms to fill the various bullpen roles that have sprouted like weeds in the contemporary version of the game. This year, the Royals have found themselves devoid of quality middle relief performances. Who is to blame? Dayton Moore? Trey Hillman? The players? The easy answer is all the above, but asking the question may help to deflect attention from what has emerged as an open sore over the last few days.

You can't look at individual middle reliever performance using conventional statistics without including holds.* That metric has its issues, though. A pitcher gets credit for a hold if he enters the game with a lead in a save situation and then departs with his team still in the lead. If he loads up the bases while he's in the game, no matter. He still gets a hold. Then if the subsequent reliever comes in and allows those inherited runners to score, he gets tagged with the blown save. So we're talking about less than perfect statistics. To better guage reliever performance, the mathematically-inclined can turn to a couple of places.

One excellent metric is the WX system at Baseball Prospectus, which is defined as such:

"Expected wins added over an average pitcher. WX uses win expectancy calculations to assess how relievers have changed the outcome of games. Win expectancy looks at the inning, score, and runners on base when the reliever entered the game, and determines the probability of the team winning the game from that point with an average pitcher. Then it looks at how the reliever actually did, and how that changes the probability of winning. The difference between how the reliever improved the chances of winning and how an average pitcher would is his WX."

B-Pro also has offshoots of that, WXL, WXR and WXRL that add additional adjustments, such as for quality of batters faced.

Another great stat, especially for relievers, is Wins Probability Added. This metric is based upon research that appeared in "The Book: Playing the Percentages in Baseball", by Tom Tango, Marshall Lichtman and Andrew Dolphin, all of whom I believe are currently employed by big-league teams. (Not the Royals.) WPA is tracked at Fangraphs.com and is summed up like this:

"WPA is the difference in win expectancy (WE) between the start of the play and the end of the play. That difference is then credited/debited to the batter and the pitcher. Over the course of the season, each players’ WPA for individual plays is added up to get his season total WPA."

Win expectancy is "The percent chance a particular team will win based on the score, inning, outs, runners on base, and the run environment."

So these are two different systems, both based on play-by-play data, expressed as wins, that give us a nearly ideal picture of how a relief pitcher does in the context of how he is used. Currently, there are 274 big-league pitchers that have thrown at least 10 relief innings this season. Here is the top 10 in WXRL:

RANK NAME, (WXRL)
1. David Aardsma, (3.927)
2. Mariano Rivera, (3.802)
3. Jonathan Broxton, (3.683)
4. Heath Bell, (3.557)
5. Rafael Soriano, (3.492)
6. Ryan Franklin, (3.471)
7. Ramon Troncoso, (3.464)
8. Jonathan Papelbon, (3.457)
9. Joe Nathan, (3.024)
10. Francisco Cordero, (2.929)

Here are the Royals' relievers with at least 10 relief innings:

RANK NAME, (WXRL)
23. Joakim Soria, (2.015)
123. John Bale, (0.287)
149. Roman Colon, (0.126)
183. Horacio Ramirez, (-0.043)
191. Robinson Tejeda, (-0.078)
203. Juan Cruz, (-0.159)
206. Ron Mahay, (-0.202)
244. Jamey Wright, (-0.504)
269. Kyle Farnsworth, (-1.05)

Here are those same relievers and their 2008 WXRL figures:

4. Joakim Soria, (5.444)
12. Ron Mahay, (3.594)
87. Juan Cruz, (1.158)
104. Doug Waechter, (0.918)
108. Kyle Farnsworth, (0.869)
118. John Bale, (0.788)
229. Jamey Wright, (0.105)
233. Robinson Tejeda, (0.099)
394. Horacio Ramirez, (-0.226)
--. Roman Colon (did not pitch in majors)

I include Doug Waechter because when he was signed in the offseason, he was considered to be a key bullpen addition, but has been injured this season. The Royals put together a staff of relievers that, by and large, got the job done last year. Ron Mahay was one of the 15 best relievers in baseball. This year, he's 206th.

On one hand, you can say that these lists absolve Dayton Moore and Trey Hillman from blame because the pitchers have not lived up to expectations. To a certain extent, that is certainly true. However, this only underscores my point about the volability of relief pitching. My biggest complaint with this year's middle relief staff is not that it's been bad. It's that it's been expensive and bad. Here's what some of these guys are making this season:

Kyle Farnsworth ($4.25 million)
Ron Mahay ($4.0)
Juan Cruz ($2.25)
John Bale ($1.2)
Horacio Ramirez ($1.8)
Doug Waechter (.64)

That's $14.14 million for what is quite probably baseball's worst middle relief crew. It's like investing in a blackjack player. He might hit 21 a few times, but sooner or later, he's going to go bust. This is precisely why it's not a good idea to overspend on middle relievers. Not only can you get stuck with a horrific performance-to-pay ratio on your roster, but you're going to be prone to stick with the underachievers simply because you're paying them so damned much money.

The Royals rank 27th in team WXRL this season, but even that is inflated by Soria's outstanding performance. (A side issue: Where the hell was Soria this weekend while the rest of the bullpen was imploding?) The best bullpen in baseball has been that of the Dodgers, followed by the Red Sox and the Reds.

Here are the salaries of the Dodgers' five-best relievers, keeping in mind that LA is one of the few franchises that could actually afford to squander huge sums on middle relief:

Guillermo Mota, 2.35 million
Jonathan Broxton, $1.825
Ramon Troncoso, .4 (league minimum)
Ronald Belisario, .4 (league minimum)
Brent Leach, .4 (league minimum)
Total: $5.375 million

How depressing is that? You just don't know where middle relief is going to come from one season to the next. You have to be willing to move guys in and out of specific roles and up and down from the minors. Try a minor-league starter in a short relief role. Mix and match according to platoon advantages, or flyball/groundball tendencies, or whatever. Smart managers find a way to get through those crucial middle innings -- if they have the parts to work with. Just don't make them expensive parts.

Again, the problem with the Royals' bullpen isn't that it sucks. It's that it costs too much, so they feel compelled to keep running these guys out there. If there were a bunch of $600,000 pitchers out there, they wouldn't be in this predictament. And that $14 million currently going towards Farnsworth, Mahay and company could have paid for a much-needed power bat.

That's my takeaway from the Tampa series. Sure, the bullpen blew three straight leads. But the offense scored six runs in the last 24 innings of the series. That leaves no margin for error and with this group, circa 2009, you need plenty of breathing space. Of course, if you bring back the same group next season, they might be a lot better. That's middle relief. Let me leave you with a random sampling of some non-Royals relievers and their WXRL scores this season:

11. Jeremy Affeldt, (2.793)
39. Kiko Calero, (1.565)
41. J.P. Howell, (1.517)
45. Leo Nunez, (1.403)
65. Ramon Ramirez, (0.968)
94. Shawn Camp, (0.502)
97. Mike MacDougal, (0.474)
110. Dennys Reyes, (0.378)
118. Octavio Dotel, (0.308)

* - I didn't really address this holds thing. You can't assess middle relief using saves and blown saves because you're not giving them a chance to record a positive stat. While middle relievers may enter a game in a save situation, they realistically aren't going to get the save no matter how well they pitch. It's not their job. Even the three saves the middle relievers have converted where there to be had only because Soria was on the disabled list. If you remove closer saves from the equation, every single team in baseball is going to have a save percentage not far from zero. Holds are meant to alleviate part of that problem by giving the middle staff some credit. I love the idea in theory, but the specifics of the stat need some work. That said, holds aren't useles. Also, through the play-by-play analysis involved in WPA, I figured out that, on average, a middle reliever appearance in a save situation has approxiately one-half the impact on a game's outcome as that of a closer. So that's why I developed HPS% (holds-plus-Saves conversion percentage), where holds carry 1/2 the weight of saves.