More Fun With Rotation Numbers

My article on Wednesday, How Good Is Your #4 Starter, presented some counterintuitive findings on starting rotations, such as—big surprise—how good an average #4 starter is. Due to injuries and the related fact that the guy who starts the year as your #5 starter often ends up in the middle of the rotation, teams rarely get the production they predict out of each rotation spot.

These calculations don’t hold the key for any breakthrough new approach to roster construction, but they do illustrate some of the ways in which good (or lucky) teams are different from bad ones. With that in mind, let’s look at a few ways in which teams set themselves apart last year.

Rotation Spot Rankings

Here are the teams who had the best and worst performances from their “#1 starters” last year. Remember that in some cases, a team’s #1 starter is a composite: it’s defined as 32 starts from the team’s one, two, or possibly three best starters.

Team    #1 ERA
MIN     2.47
HOU     2.55
FLA     2.96
LAA     2.97
STL     3.09
…
SEA     4.22
CHA     4.28
TEX     4.41
WAS     4.64
KC      4.96

Both of our top two teams got substantial help from part-time pitchers: the Twins benefited from Francisco Liriano‘s 16 starts, while the Astros took advantage of their half-season from Roger Clemens. On the bottom of this list, there’s one surprise: while the White Sox had one of the most balanced rotations in baseball (as we’ll see in a moment), no one stepped up as their 2006 ace.

Now for the best and worst rotation middles:

Team    #3 ERA
SD      3.78
DET     3.85
LAA     3.91
FLA     3.99
HOU     4.20
…
TB      4.95
CHN     5.02
STL     5.10
WAS     5.27
KC      5.70

Even after accounting for the pitcher-friendliness of Petco, it’s impressive that the Padres got nearly 100 starts with a total ERA under 3.80. It may be a bit more surprising to see Florida on this list, but their core of rookies and Dontrelle Willis came through in a big way.

Finally, the best and worst in the #5 spot. Remember that, especially at the bottom of this list, these numbers are often the result of mixing and matching after a manager loses a front-line starter or two to injury.

Team    #5 ERA
DET     4.48
SD      4.91
CHA     4.99
OAK     5.16
SF      5.18
…
PHI     6.92
BOS     6.95
KC      7.32
CHN     7.40
BAL     8.45

Hayden Penn, we’re looking at you. It amazes me that there’s a four-run difference between the best and worst teams on this list; if you figure that each outing from a fifth starter lasts about five innings (a tenuous assumption, when a pitcher is giving up a run an inning), that’s a difference of about 70 runs over the course of the season. Seventy runs!

On another note, what was particularly impressive (or lucky) about the Tigers’ performance here is that it was patched together from several pitchers, most notably Zach Miner and Wilfredo Ledezma. It’s one thing to get 32 strong starts from your #5 starter if no one gets hurt; it’s yet another to patch together a group that outperforms the majority of #3 starters.

Distribution

It seems to me that an “even” rotation is a good one, both in the sense that your team is better off with #4 and #5 guys in the vicinity of league average, and because it would more consistently distribute the burden on your bullpen. One way to measure that is to take the standard deviation of each team’s five rotation-spot ERAs. The best and worst are as follows:

Team    STDEV
CHA     0.28
DET     0.32
OAK     0.54
SD      0.59
WAS     0.61
...
FLA     1.37
HOU     1.39
CHN     1.55
MIN     1.67
BAL     1.78

Indeed, most of the teams at the top of this list are good ones; however, the Nationals sneaked onto the board with their thorough mediocrity. The bottom of the list is a reminder that the distribution doesn’t matter much: many of those teams have a high standard deviation because their #1 pitcher is so good, not just because the back end is iffy.

What About Good Teams?

Even when speaking in the most general sense, people differentiate between a #3 starter on a good team and an equivalent rotation slot on a bad team. To try to pinpoint just how different those numbers are, here are the averages for the top half of pitching staffs in baseball last year:

Lg      #1      #2      #3      #4      #5
MLB     3.30    3.83    4.30    4.87    5.94
AL      3.27    3.81    4.26    4.84    5.76
NL      3.32    3.85    4.33    4.88    6.09

It continues to amaze me just how bad these #5 slots are. We all watch our favorite teams eke out one too many starts from, say, Ben Hendrickson or Jeff Karstens, but it has never seemed to me like back-end starters are so thoroughly horrible. What also comes as a surprise is that there’s no single place or two in which good teams outperform the average. The better staffs have slightly better back ends and somewhat superior front ends as well.

Another way to look at this would be to figure the averages for last year’s playoff teams. Of course, playoff teams needn’t have great pitching staffs; some of them aren’t even in the top half of their league. Here are the results for those eight squads:

Lg      #1      #2      #3      #4      #5
MLB     3.40    3.81    4.34    4.91    5.80
AL      3.36    3.75    4.27    4.93    5.65
NL      3.43    3.87    4.41    4.89    5.95

There’s not much of a difference between this table and the previous one; the one slight exception is for #5 starters. While the rotations, overall, are a bit worse than the average of the top half of MLB teams, the worst 30 starts are a bit better. I don’t know if there’s anything to make of that, or if it’s the result of a small sample and a tremendous Tigers rotation.

By Way of Conclusion…

Over the last few years, we’ve heard more and more general managers acknowledge the need to go into the season six, seven, or more starters deep; a great example of that is the Brewers, who just signed Jeff Suppan.

One could make an argument that the Brewers’ rotation was set, with Ben Sheets, Chris Capuano, Dave Bush, Claudio Vargas, and Carlos Villanueva. It wouldn’t have been a great rotation, but if everybody stays healthy, those hurlers could give you a credible fivesome for 150 or more games. But, of course, the odds of that are tiny. When Villanueva (or Darrell Rasner, or Brad Hennessey) becomes your sixth man, your chances of suffering a 6.00 ERA over 30 starts goes down substantially.

If there’s one thing this analysis suggests, it’s just how common that painful outcome is. If, as I suggested on Wednesday, your #4 starter isn’t as good as you thought, that goes double for the guys who pitch the following day.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Ten Things I’ve Learned This Offseason
Next: Secrets Are Not To Be Fehred »

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current ye@r *