There’s a segment of informed, intelligent baseball fans who would prefer
that performance analysts spend less time criticizing the management of MLB
teams. The notion that "they may know things we don’t" is popular
in some circles, and is used as a defense for decisions that we, as informed
outsiders, find bizarre or even infuriating.
I don’t buy it. The decision-making process in most organizations isn’t a
process at all, but a semi-random collection of individual decisions. Most
of these decisions don’t contribute to an overall plan, and many are
justified by the worst kinds of specious logic, post-hoc rationalizations,
and misuse of statistics.
We saw an example of non-decisionmaking last night in New York, and the
stupidity of it left me practically shaking in anger.
I don’t have a whole lot of credibility left in evaluating Paul
Wilson, about whom I wrote this in Baseball Prospectus 2001:
He finally had his elbow rebuilt in the spring of 1999, sat out a year, and
you see the results above. Paul Wilson is going to be a major national story
by midseason, when Joe Torre names him to the AL All-Star team. It wasn’t
just his good performance after being traded at the July deadline. It was
his effectiveness at every point during his comeback last year. It was the
way in which he walked over the AL in August and September, allowing only
one home run in 51 innings.
The best part is that now he owns a rebuilt elbow and has basically no
mileage on his shoulder. He’s entering what should be his prime without any
concern about his workload history.
I was wrong. Wilson spent most of April and May doing a convincing
impersonation of Charlie Brown, giving up 54 runs and 11 home runs in 57 2/3
innings, and was probably the worst pitcher in baseball in that time.
A funny thing happened, though: Wilson pitched great upon being sent to the
bullpen, with a 3.20 ERA and an 18 to 5 strikeout-to-walk ratio in 19 2/3
innings, and allowed just one run in his last 10 relief innings before
re-entering the rotation July 25. Lo and behold, he was wonderful; entering
last night, he’d allowed no runs in three of his four starts, with the other
being an acceptable 5 1/3-inning, three-run performance. His K/BB ratio had
jumped to 18 to 2 in 21 1/3 innings. Wilson looked like he was ready to
fulfill my expectations of him, especially since Hal McRae was handling him
in a very conservative way, one that appeared to be designed to maximize
Wilson’s opportunity for positive results.
What do I mean by that? As I mentioned, Wilson did not give up a run in
three of his first four post-return starts. In the first, he threw just five
innings, allowing three hits and no walks while striking out five. Despite
his having thrown just 66 pitches, McRae pulled him after those five
innings. He didn’t start the sixth in his next start, either, leaving after
five innings, three runs allowed, four strikeouts, and again, just 66
pitches. He was credited with wins in both those starts.
In the third start of his comeback, Wilson again allowed no runs, this time
in 5 1/3 innings. He was stretched out to 86 pitches, but McRae lifted him
in the sixth during a White Sox rally. McRae seemed quite protective of
Wilson, unwilling to allow him to blow a successful start by pitching too
deep into it.
Wilson’s next start, the one previous to last night’s, saw more of the same:
six shutout innings, 97 pitches, seven strikeouts, one walk. He didn’t come
out for the seventh.
In four starts, Wilson had slowly built his pitch count from 66 twice to 86
to 97. He was basically not permitted to fail, being lifted with a lead in
all four starts, and while still pitching reasonably effectively at the
point of removal. I was pretty impressed, both with Wilson’s work and with
the way McRae handled him to maximize his chance for success, and for
Then came last night.
Wilson was ineffective early, and hurt by the defense behind him. He even
gave up just his second home run since June, in the second inning to Clay
Bellinger. After three innings, he trailed 5-0, and it wasn’t
undeserved. Here are Wilson’s ball/strike counts for his first four starts
after the All-Star break:
Start Pitches Balls Strikes
7/25 66 21 45 7/31 66 18 48 8/5 86 30 56 8/10 97 35 62
Last night, through three innings, Wilson had already made 63 pitches, and
thrown 26 balls (four intentional). It wasn’t going to be his night, and it
would not have been surprising to see McRae remove him after the third. He
didn’t, and Wilson continued to scuffle; while he didn’t allow any runs over
the next three innings, he fell behind most hitters, walked three while
striking out just one, and was generally ineffective.
Through six innings, Wilson had thrown 110 pitches, his longest outing
since…I don’t know. It was definitely his longest major-league outing
since 1996, and I’ll speculate that the number of 110-pitch outings he’d had
in the minors in the interim is safely under five. He was far from
effective–45 balls, 65 strikes–but had managed to salvage some positives
from a night on which he’d scuffled and been let down by his defense, while
continuing the progress of stretching his pitch count to a starter’s level.
Anything positive that was going to be accomplished had been accomplished by
the end of the sixth inning. There was nothing to be gained by sending
Wilson out for the seventh, and doing so put into play the following
- pushing him to a pitch count he hadn’t reached in a long time
- doing so when he wasn’t pitching effectively
- exposing him to the possibility of ruining a decent outing
Having Wilson pitch the seventh inning, when he had been ineffective and was
past his previous high pitch count, was the exact opposite of how McRae had
been handling Wilson–successfully!–up to that point:
- One start before, he’d fallen behind all four Twins hitters in the sixth
inning, allowed two hits, and been removed having thrown 11 more pitches
than he had in his previous start.
- Last night, in the sixth inning, Wilson fell behind two of the four
hitters (a third swung at the first pitch), allowed a walk, and completed
the inning having thrown 14 more pitches than in his previous start.
Note that this isn’t a rant about the pitch count. 130 pitches is probably
too many under these circumstances, but that’s not the point. The point is
allowing a pitcher, still working his way back from various surgeries, to
work too far past his established endurance level on a night when he didn’t
have his best stuff, and in doing so, throwing away the work he had done
just to salvage a decent start.
And the work was thrown away. Wilson got a foulout to start the inning, then
walked consecutive batters. He was now over 120 pitches, with rapidly
declining command. He stayed in the game long enough to get an out and give
up a three-run home run. Then, at 130 pitches, was finally allowed to leave.
The decision couldn’t have been about protecting the bullpen: Bobby
Seay had thrown just six pitches the night before, and neither Victor
Zambrano nor Esteban Yan had pitched since Sunday.
It’s decisions like this on which we’re supposed to go easy, to refrain from
criticizing because "they’re the professionals." Bullspit. This is
the kind of decision that a "baseball person" is supposed
to be able to make better than an outsider, and yet Hal McRae managed to
screw it up, to turn a potential positive next step in Paul Wilson’s
development into an ugly eight-run disaster.
Any coherent plan for the use of Paul Wilson would have had him out of the
game. There was no coherent plan, hell, no plan at all; just the same random
actions that characterize the management of bad baseball teams. It stinks,
and I feel no compunction to soften that assessment just because no team
sees fit to cut me a paycheck.
If this is the application of inside knowledge, then skepticism for that
knowledge is not just deserved, it’s mandatory. Hal McRae made an
inexplicable blunder last night; here’s hoping Paul Wilson doesn’t pay the
price for it.
Joe Sheehan is an author of Baseball Prospectus. You can contact him by