Clarifying PAP: Exploding a Couple of Myths

It has now been more than two years since we created Pitcher Abuse Points
to standardize the measurement of pitcher workloads. By and large, we have
been more successful at that goal than we had any reason to expect. Some
people in baseball now agree with the notion that limiting pitch counts in
an attempt to keep pitchers healthy is one of the most important topics in
baseball. Then again, many more people think we’re full of crap. Or, to
quote Bruce Jenkins of the San Francisco Chronicle:

"Great to see Dusty Baker infuriated over amateur-hour accusations of
him 'overworking' pitchers. Earnest twits like to sit around with their
stats and spectacles trying to pretend they understand the game. Whoa,
somebody went over 100 pitches--can't have it! This calls for an immediate
investigation! Fact is, some guys could throw 200 pitches a game--in a
four-man rotation, if necessary. Some throw 50 and they're done for the
night. Dusty knows the difference, and he doesn't need some knucklehead
trying to manage his rotation.... The nonsense at its worst: Scott Boras,
agent for St. Louis rookie sensation Rick Ankiel, chided the
Cardinals for allowing Ankiel to throw more than 100 pitches in a game.
Manager Tony La Russa said Boras 'stepped over the line,' and privately, he
wanted to strangle Boras. Too bad he didn't..."

"Earnest twits like to sit around with their stats and
spectacles…." What is this, high school? Are we going to get called
"Four-Eyes" next? For the record, I no longer wear glasses, after
having LASIK surgery last year. Keith Woolner does have an eye patch,
though, along with a parrot who insists on calling everyone "Matey."

And doesn’t Dusty Baker wear glasses?

Love it or hate it, the idea that pitch counts matter isn’t going away. You
might think Scott Boras is doing the devil’s work by meddling in the way
the Cardinals use Rick Ankiel. (Actually, you may think Scott Boras is
doing the devil’s work even if you agree with him on this point.) Either
way, the fact is that when Ankiel pitches, everyone is aware of his pitch
count.

But as PAP has taken on a life of its own, it has outgrown the original
boundaries set for it. Worse, some misconceptions about the system have
developed. I wanted to address two of them here:

What’s so terrible about throwing 101 pitches? Why should a pitcher be
penalized for throwing even one pitch above 100?

Every starting pitcher is going to rack up Pitcher Abuse Points. Every
starting pitcher is supposed to pick them up. The point of PAP is not to
mandate that every pitcher, in every situation, must be pulled after he
throws 100 pitches.

A hundred pitches is a benchmark. It’s a nice, round number, and it
correlates well with the stage of a game that a pitcher frequently loses
his effectiveness, i.e., becomes fatigued. Setting 100 as a threshold not
only has the effect of "punishing" a pitcher who exceeds that
total, but it also rewards a pitcher who occasionally has a breather start,
when his managers pulls him after six innings and 93 pitches because his
team is up 8-2.

For an example, let’s take a look at the pitch counts through the end of
June for Shane Reynolds:

Date   Pitches   PAP     Date   Pitches   PAP

4/4         89     0     5/22        89     0
4/9        112    14     5/27        93     0
4/14        88     0     6/2         90     0
4/19       107     7     6/7        120    30
4/25       125    45     6/13       101     1
4/30       109     9     6/18       117    24
5/5        102     2     6/24        78     0
5/10       117    24     6/29       117    24
5/16        94     0     TOTAL            180

Reynolds has 180 PAPs in 17 starts, an average of 10.6 per start. This is
something I can’t stress enough: there is no reason to think that an
average of 10.6 PAPs per start puts a 32-year-old pitcher at any more risk
of injury than if he had spent the first three months of the season laying
on a beach chair on Maui.

We named the metric "Pitcher Abuse Points" because we were trying
to isolate those pitchers that were, indeed, overworked. But for most
pitchers, the term "Pitcher Usage Points" may be more
appropriate. Leading the league in PAP is a bad thing, but that doesn’t
mean that any PAP is a bad thing. Too much fat in your diet is very
dangerous–sorry, Atkins Diet fanatics–but that doesn’t mean that one
slice of pizza a week, by itself, is going to kill you. Some fat in your
diet is necessary for survival, and some work is necessary for pitchers to
stay sharp and keep limber. Major-league teams have been stressing that for
years, and we don’t disagree with the point. We simply disagree about how
much work is enough, and how much is too much.

Getting back to Reynolds, you will notice that he threw more than 100
pitches, technically into the danger zone, in over half (10) of his 17
starts. But in many of those starts, the affect was minimal: In four of
those starts he threw less than 110 pitches, and totaled just 19 PAPs in
those four starts for an average of 4.8, or less than half his overall
seasonal average! No pitcher is going to set off our abuse alarms by
throwing 105 pitches every time out.

Setting the PAP threshold at 100 pitches means that when Reynolds
throws only 78 pitches he gets rewarded for the breather afforded his arm.
If you look back to June 13, Reynolds had a 101-pitch outing in between
two longer outings. Did he cross the 100-pitch threshold? Sure. Did he
really get penalized? Not at all; one PAP means nothing and the outing
significantly decreased his average PAP score for the season.

If anything, you might think that the difference between a massaging
78-pitch outing and a full-length 101-pitch feature should be more than one
PAP. We could set the threshold at 80 pitches, in which case every pitcher
in baseball would have their PAP scores doubled or tripled.

But it’s not the raw numbers that matter: it’s where they are relative to
the rest of the league. You can set the threshold at 80, 120 or anywhere in
between, and Livan Hernandez is probably going to lead the world in
PAP. But until we have better evidence as to where the threshold is for
most pitchers, 100 pitches seems as good a place as any.

Why is everyone throwing a fit because Ankiel threw 121 pitches in one
game? Sure, it’s a lot of pitches, but it’s only one game! Shouldn’t we
give Tony La Russa the benefit of the doubt that he’s not going to make
Ankiel throw that many pitches every time out?

Last September, with the A’s in a desperate wild-card chase, they locked up
in a tight game with the Tigers. Art Howe rode his starter, Tim
Hudson, for 136 pitches as the A’s held on for a one-run win. Was
Hudson abused? Yes, he was: a 23-year-old rookie (he had actually just
turned 24) should not be throwing that many pitches. But was the risk worth
it? Just as emphatically, yes. Not just because of what was at stake, but
because the A’s had, to that point, taken good care of Hudson’s arm. On
only two other occasions did Hudson throw more than 110 pitches in a start,
and nearly half of his season’s PAP (84 of 189) came in that one September
start.

And that is the issue with Ankiel, or any other young pitcher who
occasionally throws 120 pitches. One fairly-high-pitch outing is not going
to do permanent damage to Ankiel’s arm. But the Cardinals are in first
place and they’re going to do whatever it takes to stay there. If La Russa
is willing to let him throw that many pitches on May 25, what is going to
happen in late August when the Reds are making a run and the bullpen is in
shambles and the Cardinals need someone in their rotation to step up? I’ll
tell you what could happen: La Russa could use Ankiel the way Jim Riggleman
used Kerry Wood in 1998. That the Cubs had a lot on the line that
year is undeniable, but if they hadn’t let Wood throw 110 pitches or more
early in the season, then when they had to let him throw that many pitches
late in the year, his arm might not have broken down in August and he might
not have had to miss all of 1999.

Here, start by start, are Rick Ankiel’s pitch counts:

Date   Pitches   PAP     Date   Pitches   PAP

4/5         44  (Rel.)   5/19        76     0
4/9        102     2     5/25       121    33
4/14        91     0     5/30        93     0
4/20       112    14     6/4         98     0
4/26        99     0     6/10        98     0
5/2        116    22     6/20        98     0
5/7         95     0     6/25       110    10
5/13       118    26     6/30        96     0

That’s a total of 107 PAPs in 15 starts, an average of 7.1 per start.
Factoring in Ankiel’s age, that’s a Workload of 21.4; high, but forgivable,
and much lower than Wood’s 58.0 Workload in 1998. And let’s give La Russa
some credit here: Ankiel has yet to throw 100 pitches in back-to-back
starts, and since Scott Boras objected, Ankiel has thrown 100 pitches just
once in six starts.

And here, again, you see the value of setting 100 pitches as the threshold–Ankiel
has thrown 90 to 99 pitches eight times in 15 starts, and has not
received a single PAP in them. Even with a threshold of "just"
100 pitches, if Ankiel had been pulled after 110 pitches in all of his
starts–which would only have required La Russa to pull him earlier than he
wanted on four occasions, forcing a middle reliever to pick up the tab on
just 27 pitches–Ankiel would have just 52 PAPs and a Workload of 10.4, and
no one would be complaining at all.

Rany Jazayerli, M.D., can be reached at ranyj@baseballprospectus.com.