Baseball By The Numbers By Michael Salfino

May 19, 2006
Why Save The Best For Last?

Saves is a statistic that’s seemingly been created with fantasy baseball in mind. Closers are viewed as the team’s best reliever, yet they’re often employed in non-crucial situations only because of how the save rule is written.

Closers are 3.5 times more likely to be brought into the ninth inning with a three-run lead than with the score tied. Managers continue to do this despite the fact that there’s no significant decrease in the chance of winning when up three runs even if the final frame is entrusted to the low man in the bullpen hierarchy.

According to a recent tome titled, “The Book: Playing The Percentages in Baseball” (Tango, Lichtman, Dolphin), teams won 97 percent of the time when up three runs in the final inning with no one on base. The rate climes to only 97.5 percent with the most trusted relievers, defined as pitchers who faced the highest percentage of batters with the game tied or leading by three runs or less. The least trusted relievers converted these “saves” (in name only) 95.5 percent of the time over the three seasons studied.

As recently as 20 years ago, bullpen usage and the number of saves by league leaders were dramatically different. Even with starting pitchers now working less innings (and thus leaving more to the bullpen in general), usage by closers is down almost 30 percent. That means that lesser quality arms are being called upon to fill an equally greater percentage of innings that are demonstrably more important than most save situations.

The one-inning closer was born in 1988 when Tony LaRussa determined that was the best way to deploy Dennis Eckersley. By 1991, the practice was common in the NL, too, as Cardinals began using Lee Smith the same way. 

With managers now being forced to use non-closers in so many more important innings, other relievers have to be better than in prior generations. The narrowing of the gap between the number one and number two man in the bullpen has created greater volatility than ever in who managers chose for easier ninth-inning role.

Let’s look at some current bullpen situations and try to anticipate changes.

Buy

Matt Capps (P, Pirates): Mike Gonzalez has a nice ERA in his current role, but continues to walk five or six guys per nine innings. As a 30-year-old lefty, Gonzalez also is a prime candidate to be traded to a contender this summer. The 22-year-old Capps has the stuff and the stats to close right now. The righty also sports impeccable control.

Eric Gagne (P, Dodgers): He threw all his pitches in a bullpen session last week and will face hitters for the first time since his April elbow surgery. Expect an early June return.

Hold

Akinori Otsuka (P, Rangers): Does not have lights-out stuff, but is so steady in limiting walks and homers that he should continue to thrive working mostly with multi-run leads.

J.J. Putz (P, Mariners): No one saw this kind of dominance coming, as he’s striking out about 12 batters per nine innings. He’ll have to continue blowing away batters to hold off the dynamic Rafael Soriano (11 Ks per nine innings).

Brad Lidge (P, Astros): He was demoted last week because of control problems, but indications are the move is temporary. Arm trouble is the big concern when command mysteriously disappears, so don’t be a buyer.

Sell

Chris Reitsma (P, Braves): Just seven Ks in his first 16-plus innings sets off all kinds of alarms. The only bright spot is that there’s no one on the current roster pitching well enough to replace him.

Joe Borowksi (P, Marlins): His control is too spotty. The good news is that he hasn’t given up a homer and is striking out guys at an acceptable rate. Travis Bowyer is still the “Future Closer,” but is sidelined at AAA with a weak shoulder.