clock menu more-arrow no yes

Filed under:

The 2008 Brewers in an alternate universe

New, comments

Will Carroll had a throwaway line in a blog post a couple of weeks ago that piqued my interest.  He suggested the following:

here's a scary exercise ... take the top PECOTA comparable for each starter on your team, then ask yourself, "Would that team win?"

Of course, I immediately wondered what the comparables would look like for the 2008 Brewers.

Before I list the team, a few words about what this all means.  To varying degrees (the details get complicated), PECOTA projections are based on a group of comparable players.  This is different from things like Bill James's similarity scores, which you can see at Baseball-Reference.  A lot of work has been done to determine which characteristics mean the most when projecting a player (power, yes; earned runs, not so much), and PECOTA focuses on those skills (along with age, body type, and a variety of other things) to determine which players are most alike.

Since PECOTA is based on a large number of comparable players, this isn't some kind of super projection I'm doing right now -- to say that Prince Fielder looks a lot like Boog Powell at the same age doesn't mean you should go look up Boog's numbers for that season and mark Prince down for that.  But what it does do is put a human face on projections, which are a little tougher to get a "gut feeling" about.

What I like most about it is that it forces you to think about your team in a more objective sense.  Those of you who contribute on this site aren't as given to this as others, but all of us have a tendency to overrate our guys.

The other cool thing is that it reminds you of some players who were really good for a while, but didn't stay good long enough to really imprint themselves on your memory.  There are a lot of names like that on this team.  So here goes:

C: [Kendall] Joe Girardi
1B: [Fielder] Boog Powell
2B: [Weeks] Joe Foy
3B: [Hall] Dave Henderson
SS: [Hardy] Kevin Elster
LF: [Braun] Tony Conigliaro (honorable mention: Braun's #2 comp is A-Rod)
CF: [Cameron] Ron Gant
RF: [Hart] Sammy Sosa

SP: [Sheets] Alex Fernandez
SP: [Yo] Jeremy Bonderman
SP: [Suppan] Chris Bosio
SP: [Cappy] Kenny Rogers
SP: [Bush] La Marr Hoyt
SP: [Villy] Steve Dunning
SP: [Vargas] Jason Johnson

CL: [Gagne] Roberto Hernandez
SU: [Riske] Roy Lee Jackson
RP: [Shouse] Buddy Groom
RP: [Torres] Mike Marshall
RP: [Mota] Turk Farrell
RP: [Turnbow] Charlie Hough (huh?!?)
RP: [Parra] Joe Saunders

I'm not enough of a historian (especially with the 70s/early 80s guys) to have a good idea of how that team would fare.  It's especially hard when you've got guys like Tony C., who were really good for a while, but have gone down in memory as disappointments.

I've included the links to each player's baseball-reference page so you can investigate further.  If someone can explain to me how any algorithm would make D-Bow's top comp Charlie Hough, I'd really appreciate it.