Closers: Part I

Pete has made several comments about how closers on bad teams deal with a sort of "price ceiling" whereas closers on good teams enjoy a sort of "price floor". It is worth going into each of these terms briefly for a better explanation. A closer with a price ceiling has a minimum ADP that he theoretically should not rise above a certain number, regardless of peripherals. These would be pitchers such as Joakim Soria or Huston Street. On the other hand, closers with a price floor should be drafted no later than a particular ADP, even if they have an irrationally awful ERA and WHIP. We are, of course, talking about the Joe Borowski's and Todd Jones's of the world.

Lets look at saves vs. team wins for the 2007 season. Because most lists include save opportunities for setup men, I'm going to look only at save opportunities for closers. This assumes that the number of days closers are given off is either roughly constant across the league and also unpredictable, which I think is a fair assumption because one cannot predict "consecutive days requiring a closer".

• Boston - 96 wins                     Papelbon - T 10th with 37 saves
  
• Cleveland - 96 wins                Borowski - 2nd with 45 saves
  
• Anaheim - 94 wins                  Rodriguez - T 5th with 40 saves
  
• New York (AL) - 94 wins       Rivera - 16th with 30 saves
  
• Colorado - 90 wins             Corpas/Fuentes - total of 39 saves
  
• Arizona - 90 wins                  Valverde - 1st with 47 saves
  
• Philadelphia - 89 wins          Myers/Gordon - total of 27 saves
  
• San Diego - 89 wins             Hoffman - 4th with 42 saves
  
• Detroit - 88 wins                 Jones - 9th with 38 saves
  
• Seattle - 88 wins                  Putz - T 5th with 40 saves
  
• New York (NL) - 88 wins        Wagner - 13th with 34 saves
  
• Chicago (NL) - 85 wins          Howry/Demptser - total of 36 saves
  
• Atlanta - 84 wins                 Wickman/Soriano/Gonzalez - 31 SVs
  
• Milwaukee - 83 wins               Cordero - 3rd with 44 saves
  
• Toronto - 83 wins                   Accardo/Ryan/Frasor - 36 saves
  
• Los Angeles - 82 wins          Saito - 8th with 39 saves
  
• Minnesota - 79 wins            Nathan - T 10th with 37 saves
  
• St. Louis - 78 wins                Isringhausen - T 15th with 32 saves
  
• Oakland - 76 wins                 Street/Embree - total of 33 saves
  
• Texas - 75 wins                     Gagne/Wilson/Benoit - 34 saves
  
• Houston - 73 wins                  Lidge/Wheeler - total of 30 saves
  
• Washington - 73 wins             Cordero - T 10th with 37 saves
  
• Chicago (AL) - 72 wins             Jenks - T for 5th with 40 saves
  
• Cincinnati - 72 wins                Weathers - 14th with 33 saves
  
• San Francisco - 71 wins            Hennessey/Wilson/Benitez - 34 SVs
  
• Florida - 71 wins                      Gregg - T 15th with 32 saves
  
• Baltimore - 69 wins                  Ray/Walker/Baez - 26 saves
  
• Kansas City - 69 wins                 Dotel/Soria - 28 saves
  
• Pittsburgh - 68 wins                  Torres/Capps - 30 saves
  
• Tampa Bay - 66 wins                  Reyes - 26 saves
  

I won't claim infallibility here, for I may have forgot someone who served as a closer due to demotion or injury-though I think I got most of them. Also, these are saves as opposed to save opportunities, but we already know that a good closer on a good team is the best option, and bad closers on bad teams are worth little. Hence, bad closers who blew saves on good teams will benefit from excluding their blown saves.

The correlation between team wins and saves is .554 (r=.554, r squared = .307), which is essentially a crap shoot. Also, the standard deviation of saves is only five and a half. Reyes, who was not a good closer, only had 11 less saves than Papelbon, who pitched for the best team in the league.

All this points to a draft strategy that doesn't weigh "team strength" heavily when choosing a closer. The average saves for closers last year (this time omitted the outlier) was 35. Only Rivera and Wagner, plus the Braves trio, failed to pass this among teams with winning records. Among teams with losing records, Nathan, Cordero, and Jenks surpassed the mark. So, it looks as if having a winning record does influence a sort of barrier around the average, though variation is rather low. In my Yahoo! league last year, 5.5 saves would have meant a maximum of 1 point gain in the league (and at what expense with guys like Jones and Borowski?). For those curious, the deviation in the league was 32...

Next week, look for an analysis of projected winning teams this year, and closers to target after your draft.