Bullpen Misuse Leads to Another Loss

Game is tied 1-1 going into the bottom of the ninth inning. You have the following relievers left:

That was the situation the Phillies faced tonight against the St. Louis Cardinals.

Charlie Manuel went with Baez to start the inning. Baez, who has not shown the ability to strike out Major League hitters at an acceptable rate. The first three Cardinals made contact, reaching on three singles to load the bases, something that happens against pitchers that can’t miss bats. If Manuel had not intended to use Madson to start the inning, he should have at least considered using Madson as a fire extinguisher — he should have been warming before Baez threw his first pitch to Ryan Theriot. In a situation where you’re pitching to the home team in the bottom of the ninth inning with runners on first and second with no outs, you want to achieve one of two goals: A) get a strikeout, or B) get a double play. Madson, with his elite strikeout and ground ball rates, can accomplish both better than nearly anyone in baseball, let alone on his own pitching staff.

Baez breathed a sigh of relief when he had the fortune of generating a ground ball from Matt Holliday that was hit directly at Jimmy Rollins, who threw to home for the force out. Lance Berkman, a switch-hitter, was due up next, so of course Manuel went with the most logical option: Ryan Madson J.C. Romero. Romero, who struggles mightily against right-handed hitters. Against Berkman, who has an OPS approaching 1.100 against lefties this season. Naturally, Berkman hit a screaming line drive to center field that went over Michael Martinez‘s head to end the game.

This isn’t just Monday-morning quarterbacking. Matt Swartz (@Matt_Swa) was doling out the strategy on Twitter before the inning even started:

With Pujols due up 3rd, Madson should be pitching the bottom of the 9th regardless of what happens in the top

Manuel’s misuse of the bullpen has been noted previously here, but to add a cherry to the dessert of analysis, check out this graph from Chasing Utley (@Phylan) from last week (click to enlarge):

If Madson is the best reliever in the bullpen (and he is, there is no counter-argument), he certainly isn’t being used as such.

I realize that it’s easy to pick on the manager and it’s even cliche at this point, but these are objective reasons why Manuel’s in-game decision-making has been unsatisfactory. His affable personality and clubhouse management are certainly great attributes to have in a manager; if Charlie would be a bit more discerning with his strategy, he could be a great manager.

Like Crashburn Alley on FacebookFollow Crashburn Alley on Twitter

Charlie Manuel Unnecessarily Taxing His Pitchers’ Arms

The Phillies played their 40th game of the season last night against the St. Louis Cardinals. Cliff Lee threw 122 pitches, marking the 11th time a Phillies starter has thrown 110 or more pitches in a game thus far in 2011. For those of you keeping score, that’s 27.5 percent. Only the Houston Astros have been more taxing of their starters’ arms. The average Major League team has six 110+ pitch performances on record; the Phillies are at nearly twice that total.

Player Date Tm Opp Rslt App,Dec IP H R ER BB SO HR Pit GSc aLI
Roy Halladay 2011-04-24 PHI SDP W 3-1 GS-9 ,W 8.2 5 1 1 1 14 0 130 83 .852
Cole Hamels 2011-04-22 PHI SDP W 2-0 GS-8 ,W 8.0 4 0 0 3 8 0 126 79 1.245
Roy Halladay 2011-04-13 PHI WSN W 3-2 CG 9 ,W 9.0 6 2 2 2 9 0 123 74 1.553
Roy Halladay 2011-05-15 PHI ATL L 2-3 CG 8 ,L 8.0 8 3 3 2 7 1 119 59 1.339
Cliff Lee 2011-05-06 PHI ATL L 0-5 GS-7 ,L 7.0 9 3 3 1 16 0 117 62 .724
Roy Halladay 2011-05-10 PHI FLA L 1-2 CG 8 ,L 8.0 5 2 1 2 9 0 115 73 1.225
Cliff Lee 2011-05-01 PHI NYM L 1-2 GS-7 7.0 8 1 1 2 5 0 113 60 1.203
Roy Halladay 2011-04-07 PHI NYM W 11-0 GS-7 ,W 7.0 6 0 0 1 7 0 113 71 1.016
Cliff Lee 2011-05-16 PHI STL L 1-3 GS-7 ,L 6.1 6 3 3 6 4 0 112 53 1.416
Roy Halladay 2011-04-19 PHI MIL L 0-9 GS-7 ,L 6.2 10 6 6 2 3 1 112 31 .790
Cliff Lee 2011-04-02 PHI HOU W 9-4 GS-7 ,W 7.0 4 3 3 0 11 1 111 68 .710
Roy Halladay 2011-05-05 PHI WSN W 7-3 GS-7 ,W 7.0 6 2 2 0 10 0 110 67 .757
Provided by Baseball-Reference.com: View Play Index Tool Used
Generated 5/17/2011.

The pitch count debate is anything but finished. Current information is limited in scope and conclusions are very hazy. Some pitchers, like Roy Halladay, seem to be well-conditioned to throw as many pitches as necessary while others are gasping for air before they ever reach 100. And, of course, the 100-pitch marker is itself arbitrary. Why 100? Why not 103 or 98 or 112? You get the point.

Still, I think we can all agree that each pitch is riskier than its precedent. By exactly how much is anyone’s guess, but it is non-zero. If a manager has the opportunity to allow his starter to throw fewer pitches, he should take it, generally speaking.

That was very clear yesterday when Lee was having a lot of trouble locating his pitches, setting a career-high with six walks in six and one-third innings. Lee, of course, is known for his pinpoint control, leading the Majors in strikeout-to-walk ratio last year at 10.3, and led the Majors for this season, going into last night’s game at 9.1.

Lee needed 25 pitches to escape the first inning, and was at 86 pitches through four innings, at which point he had allowed five of his six walks. He should not have taken the mound for the fifth inning. Baseball purists and macho men reading this should experience heightened blood pressure after reading that sentence. Taking out your $100 million starter after 86 pitches and four innings?!

Yes. Eliminate risk. As it turns out, Lee had fairly easy fifth and sixth innings, allowing him to pitch into the seventh, but situations like that are why you have arms in the bullpen capable of pitching multiple innings. It is exactly the reason why the Phillies still carry Kyle Kendrick on the roster. If you are not comfortable using your relievers in that situation, then when are you comfortable using them? Why are they taking up a roster spot, instead of a more useful bench player?

One need only go to this page and search for “(P)” to get an idea of the risk involved. Is getting an extra two or three innings out of Lee, rather than the rarely-used bullpen, worth risking losing him to injury? Even if the risk of injury is one percent rather than, say, 40 percent, the answer is still a resounding “no”.

In the fifth inning until he was pulled in the seventh inning, Lee faced sub-1.00 leverage index situations with seven of ten batters, and of course the other three situations were his own doing — a result of his lack of stuff. Going into the seventh, the Phillies were facing four-to-one odds to win the game. Manuel either has a remarkable lack of confidence in his bullpen or was not cognizant of how much he was asking from his starting pitcher.

The average leverage index, for the games in which Phillies starters accrued 110 or more pitches, was 1.04. As the FanGraphs Saber Library explains, an average LI is 1.00, so it isn’t as if these starters are in super-important situations. And, lest we forget, it is May — we are just now arriving at the one-quarter mark.

Even when we look at the peak leverage index, the decision-making isn’t justified. The average max-LI for the 11 110-plus-pitch games is 3.07, with a max of 7.13 in Halladay’s start against the Washington Nationals on April 13. The rest fell under 4.00, with four registering under 2.00. The two most egregious over-uses both involved Halladay: on April 7 against the Mets, when Halladay pitched seven innings as the Phillies won 11-0; and April 19 against the Milwaukee Brewers, when Halladay went six and two-thirds innings as the Phillies lost 9-0.

If the starters being overworked were Joe Blanton and Roy Oswalt, to whom Phillies owe nothing in the long term, that would be somewhat justifiable. But the Phillies owe Halladay as much as $60 million from 2012-14, and Lee as much as $124 million from 2012-16. Winning regular season games in May is nice, but protecting long-term investments is more important.

Going into last night’s game, the Phillies led the league in average innings pitched per start by starting pitchers at 6.5 (roughly six and two-thirds innings). As a result, the Phillies had also called upon the bullpen the least, at 100 and one-third innings, beating the Los Angeles Dodgers by about eight innings.

We expected this situation to occur, given the hype around the starting rotation going into the season. When you have four legitimate aces, the bullpen will end up used less and less. There is, however, the smart way to ration innings and there is the dumb way. Thus far, I’m not so sure Manuel’s use of his starting rotation falls under the former category. Oswalt has his own health problems (recently, his back), while Halladay and Lee have not had the cleanest bills of health over their respective careers — both heading into their mid-30′s as well.

There’s a reason why you keep your Porsche(s!) in the garage and your Toyota parked on the street. One represents an investment; the other, convenience. You use the Porsche only on special occasions, not for everyday driving. Manuel should use his rotation and bullpen accordingly.