Baseball By The Numbers By Michael Salfino

May 26, 2006
Steals More Prevalent in '06

The soldiers in baseball’s statistical revolution have sought to slay the stolen base by placing it directly in their crosshairs.

Proving that no one is immune to faulty analysis, these “sabermetricians” generally fail to account for game situations in their calculations. But they have proven convincingly that managers generally should not attempt a steal unless convinced the runner has a greater than 70 percent chance for success.

Why is the break-even point for a steal so high? Let’s look at a hypothetical situation where there’s a runner on first and no one out. On average, teams score 0.9 runs after gaining that situation. If you send the runner and he gets caught (leaving one out and no one on), the expected runs goes down to about 0.28. If the steal is successful, the gain in expected runs rises to only 1.2. With the risk in runs a little more than twice the reward, the gamble is foolish unless the chance for success compensates for this (which gives us the 70:30 break-even percentage).

This relatively new “steal math” has played a role in the dramatic category decline since 1999, when major league teams accounted for 3,421 thefts. Last year, the total slipped to 2,565, the fifth-straight year of decline. This year, major league teams are on pace for about 2,800 steals, with almost all the gain coming in the National League.

The teams that run the most are the Mets, Angels, Reds, Dodgers and Marlins. The Mets, Reds and Dodgers are also efficient, successful on 80 percent or more of attempts. Detroit  needs to lose its running shoes, leading the majors in outs on steals and lowest success rate (15-for-33, 45 percent).

Heading into the weekend, the toughest teams to run on are the Orioles (49 percent caught), Tigers (45 percent), A’s (43 percent), Reds (43 percent), Twins (41 percent) and Rangers (40 percent). Amazingly, teams have attempted as more steals versus the Orioles (41) than against the most inept team in nailing baserunners, the Dodgers (38 attempts despite just 11 percent caught). Other trailers in catching thieves are the Padres (12 percent), Indians (12 percent), Nationals (16 percent) and Rockies (18 percent).

What role do pitchers play in holding runners? We try to gain insight by looking at the caught stealing percentages of catchers who switch teams. This year, Mike Piazza and Paul Lo Duca are exactly in line with ’05 numbers. But Ramon Hernandez is nailing twice as many baserunners since moving to the Orioles and Bengie Molina half as many since signing with Toronto.

Buy

Willy Taveras (OF, Astros): On pace for just 20 steals despite being projected for about 40. Buyers must hope Phil Garner figures out Taveras has no offensive value unless running wild and removes the chains.

Corey Patterson (OF, Orioles): Stole 13 in a row before getting caught and then bounced back with two more the next night. Is cemented now as starter.

Dave Roberts (OF, Padres): The fact that he plays in a park that limits scoring makes his steals more valuable and thus more likely to continue.

Orlando Cabrera (SS, Angels): Since the beginning of ’05, he’s 29-for-31 in steals. He’s got a little pop, too, which helps make up for the paltry attempts.

Hold

Rafael Furcal (SS, Dodgers): He leads the majors in caughts after going 46-for-56 last year. The important thing is that the attempts are holding to projected levels.

Sell

Felipe Lopez (SS, Reds): While unexpectedly among the NL steal leaders, Lopez has a steal in only four games in May with four one night against the Nationals (who couldn’t throw out your Aunt Betty right now).

Hanley Ramirez (SS, Marlins): White hot now in every way, Ramirez was only 25-for-38 stealing in AAA last year. Plus, he won’t sustain a high average if he continues to strike out about once per game.