Wednesday, July 16, 2008

All-Star game thoughts

I'm sitting there watching the All-Star game last night. A great pitching match-up between the best of the NL and the AL. 

But, wait, if it is the best of the best, why is it a pitching duel? Couldn't it just as easily be a offensive slugfest? I mean, just the night before I watched Josh Hamilton hit 28 HRs in the first round of the Home Run Derby. I mean look at the AL hitters:

Mauer - .873 OPS
Navarro - .785 OPS
Varitek - .653 OPS
Morneau - .903 OPS
Youkilis - .933 OPS
Kinsler - .945 OPS
Pedroia - .816 OPS
Young - .777 OPS
Jeter - .740 OPS
Rodriguez - .972 OPS
Longoria - .861 OPS
Crede - .792 OPS
Guillen - .797 OPS
Drew - .984 OPS
M. Ramirez - .908 OPS
Sizemore - .913 OPS
Quentin - .900 OPS
Hamilton - .919 OPS
Suzuki - .737 OPS
Bradley - 1.049 OPS

Here's the NL:

Martin - .830 OPS
Soto - .891 OPS
McCann - .940 OPS
Berkman - 1.096 OPS
Pujols - 1.074 OPS
Gonzalez - .848 OPS
Utley - .955 OPS
Uggla - .978 OPS
H. Ramirez - .957 OPS
Tejada - .739 OPS
Guzman - .765 OPS
A. Ramirez - .901 OPS
Jones - 1.086 OPS
Wright - .878 OPS
Braun - .873 OPS
Fukudome - .791 OPS
Hart - .831 OPS
Holliday - .975 OPS
McLouth - .899 OPS
Ludwick - .962 OPS

These are some offensive juggernauts.  Yet, they only scored a total of seven runs. In comparison, looking at the league totals, the average OPS is .740 and 9 runs per game have been scored by teams.

Only four hitters (Jeter, Tejada, Suzuki and Varitek) were at the league average for OPS or lower, and yet they scored two fewer runs than in a major league game.

What's the difference?

My hypothesis - it has something to do with the All-Star game format, in particular the way pitchers are used. But, before we get to that, let's look at the total number of runs for the past 30 games. (Why 30? Statistically, I remember that being a "magical" number for a normal distribution. Plus, it keeps us in more recent times.)

Total runs over 15 - 83, 92, 98 (Coors Field)
Total runs 13 - 15 - 79, 94, 02, 03, 04
Total runs 10 - 12 - 93, 05
Total runs 7 - 9 - 81, 85, 89, 00, 07, 08
Total runs 4 - 6 - 80, 82, 84, 86, 91, 95, 96, 97, 99, 01, 06
Total runs below 4 - 87, 88, 90

So, let's look at the distribution:

Over 15 - 3
13-15 - 5
10-12 - 2
7-9 - 6
4-6 - 11
Below 4 - 3

So 20 games have 9 runs or less and 10 games have 10 runs or more. (Obviously here I'm assuming that 9 runs per game is the average throughout this time period - will go with it for now, but could use more research on that point.) More than half of the games had between four runs and nine runs.

(Interesting sidebar - 02-05 have larger scores - interesting possibility for research on steroids and the ability to make elite players, or that elite players were using. Maybe a way to pinpoint widespread usage - especially if we look at Barry Bonds 2001 season as a spark to go for the needle. Also - how long do the effects of training combined with steroid use last?)

So, the difference is obviously that pitchers only have to pitch one (maybe two) innings at a time. My question is: why don't teams try this during the regular season?

Let's assume that the ninth inning is the most important inning - that's why we see one inning specialists - "closers" - that already just pitch one inning at a time. What about the eighth inning? Well, if you have a stud number one pitcher, you could use him there, then your number two pitcher in the seventh inning and on down the line.

Of course, I don't believe that it really works that way. If you get behind early, it could have a negative effect as well. So, let's mix it up a bit and set the daily lineup for pitchers, like we do for hitters.

Inning #1 - #3 SP
Inning #2 - #4 SP
Inning #3 - #5 SP
Inning #4 - #3 MR
Inning #5 - #2 MR
Inning #6 - #2 SP
Inning #7 - #1 MR
Inning #8 - #1 SP
Inning #9 - Closer

So, each day, like with the hitters, you would have a defined position within the baseball game. A pitcher would be used each day to get three outs. Of course, pitchers will need rest, so you would carry two other pitchers on your roster to "spot pitch" innings in order to rest the regular pitchers and serve as extra innings pitchers.

In this case, each pitcher would pitch approximately 162 innings per year (which I believe is around the IP restrictions for young pitchers to avoid injuries). This would be a large increase for some relievers, but a decrease for starters. If overuse is a problem, this would reduce it over the entire season and spread 7 innings over a week instead of on one day (or two, given the week and the order of the standard rotation of today's game, not counting "pitching on the side" or getting up multiple times to warm-up in the bullpen.) Every pitcher would be in the game each day and contribute to the team win - and the "win" statistic would become even less meaningful. 

Of course, I doubt this would ever fly with conventional thinking. Not enough flexibility. Not the way things have been done. I know Tony LaRussa gets heckled a bit when he starts a reliever, and this is going well beyond that. Still, it could be interesting to see if this would tip the favor back toward the pitchers in upcoming years.

Labels: , , ,