The Importance of the 6th Starter

facebooktwitterreddit
Prev
1 of 2
Next

There has been an ongoing discussion at FanGraphs, Bless You Boys and elsewhere about the dangers of a lack of rotational depth – and specifically the Detroit Tigers offseason strategy regarding the rotation. At FanGraphs, what we read sums up to this: teams get a lot of starts from guys outside the top 5, and there can be a steep dropoff. The Tigers, specifically, might have a steeper dropoff than others. Kurt Mensching’s response at BYB sums up to “teams that get a lot of starts from guys outside the top 5 tend to lose a lot anyway”. The Detroit Tigers, whether or not they sign Jeremy Bonderman as an insurance policy, still need guys like Scherzer to stay healthy and effective to win. All true, obviously…. but both analyses are going about this a bit wrong.

The problem is that (because it’s easier) they are all looking at the rotational core as being the 5 guys who made the most starts. Therefore, counting Galarraga as ‘core’ it looked like the Tigers rotation had exceptional health/effectiveness last year. Presumably, that should have meant an effective rotation and a top finish. If we haven’t forgotten last season entirely, we know that isn’t the case. Armando Galarraga was the 6th starter, not the 5th. Dontrelle Willis was the one on the 25 on opening day, and woeful ineffectiveness caused him to be moved from the organization. We may be greatly understating the importance of teams’ 6th and 7th starters if they are considered ‘core’ if the team actually needed them.  We may also be overstating the importance of ‘health’ in the rotation: teams that suffered injuries but had no strong backup plan would wind up with many starters with 10 or so terrible starts.

Even if we just look at the American League, we can find other examples of this phenomenon, 6th starters counted as core:  Trevor Cahill won 18 games for the A’s last year, but he did not make the team out of spring training.  The same goes for Toronto’s Brett Cecil, who began 2010 in AAA Las Vegas.  The Royals started 2010 with Gil Meche in the rotation, not Bruce Chen.  If we count the core as consisting of the first 5 pitchers to make starts for the big-league club this has the effect, statistically speaking, of increasing the importance of a team’s 6th (and 7th) starter(s), we see 22% of a team’s starts made by players who began the season outside of the rotation as opposed to 17.4% and in particular we see an increased importance of the 6th starter specifically.  For 9 out of the 14 AL clubs in 2010, the 6th starter made more starts than at least one member of the initial rotation.  This, of course, can be biased by rotation members who began the season on the DL, but I believe this is a small price to pay.   Daisuke Matsuzaka did make more starts than Tim Wakefield & Scott Kazmir made the 3rd most starts for the Halos despite starting the season on the DL.  Even if we exclude the Angels and Red Sox, we still see that 7 of 12 AL teams were forced to go early and often to a 6th starter.  Both Mensching & Nicholson-Smith are understating the importance of rotation fallback plans, though Nicholson-Smith is trying to make the case that rotational depth does matter.

Follow me through the jump for some more detail…

Every situation is different, so any analysis of the importance of 6th and 7th starters should lean heavily on case studies and not averaged statistics.  There are a couple of questions I would like to attempt to address:

1.  Is a team forced to go to 6th and 7th starters often doomed?

2.  Is it better to have veteran talent rather than young pitchers to fill holes in the rotation?

To answer the first question, we need look no further than the Oakland A’s.  In the end, the A’s had by some measures the best pitching staff in baseball last year.  However, they began the year with a rotation that included known injury risk Justin Duchscherer who had made only 3 MiLB starts in ’09 and made only 5 starts in ’10 – though at least those starts were good.  The A’s promptly called up Trevor Cahill, who had started 32 games for them as a rookie in 2009.  Cahill made tremendous strides, relative to his rookie campaign, improved his numbers by just about any measure and made 30 starts for the A’s with a 2.97 ERA.  Two other A’s starters, Brett Anderson and Ben Sheets, missed a couple of months apiece but the A’s made do – getting 18 starts and a 4.29 ERA from current Royal and former 7th starter Vin Mazzaro.  The starters the A’s lost were good, Duchscherer’s career ERA as a starter is 3.01, Sheets 3.79, Anderson 3.57 – but thanks to good options for 6th and 7th starters the As stayed competitive.  In the end, as we all know, it was their offense and not their defense that kept them out of contention in September.

I am not trying to argue by any means that all teams can expect to survive serious injuries to top-of-the-rotation pitchers.  Even the A’s cannot have expected Cahill to pitch so well.  Without a doubt an injury to a Verlander, a Sabathia or an Adam Wainwright stands to cost teams affected 3 or 4 wins on average over the course of a season.  No team has reserves that include pitchers like Verlander, Sabathia and Wainwright obviously – they would be in the rotation already.  However, the way FanGraphs and BYB use the numbers it makes it appear that 6th starters are, by and large, terrible pitchers.  If they are all varying degrees of terrible, what difference does it make which terrible option we plan on using?  If we are forced to call them up at all, we are doomed.  The reality is a little different, if we count a team’s 6th starter as the pitcher who made the most starts without being part of the opening rotation – the AL average in 2010 was an ERA of about 4.25.  That’s almost exactly the same as the AL average of 4.27 for starters.  Yes, that number does include Scott Kazmir who started the season on the DL but his ERA was worse than any other team’s 6th starter at 5.94.

The answer to the second question can be found on page two…