Last night, in reference to Jacoby Ellsbury’s base stealing prowess, Joe Morgan remarked that Ellsbury’s success rate of 85% (his season and career figures) was good, but not particularly remarkable or laudable. Morgan claimed that a success rate of 75% was passable, and to be considered a good base stealer, it had to be at least 80%. As always, Morgan’s “logic” and “math” is laughable.
Morgan is generally right with his levels, in that, sabermetrically, a 78% success rate is required for a base stealer to improve his team’s chance of winning. A 75% stealer will be given the green light in situations where he is less likely to hurt his team with an out (e.g. 2-0 count, 0 out), so that is typically a passable percentage, yes. But as we all know, percentage points in baseball can make a world of difference. And if, as Morgan claims, 5% is enough to separate “good” from “passable”, how is it not enough to separate “great” from “good”? Analogy: .265 is a passable average for most positions and .295 is a reasonable threshold to be a considered a good hitter - about 20% of everyday players are hitting .295 or above this year. Joe Morgan is essentially saying that .325 isn’t good enough to be considered a great hitter. Only 8 guys are hitting .325 or above right now - to me, that seems like a fair threshold. (Keep in mind that even frequent base stealers generally have 4 times as many plate appearances as base stealing opportunities, so a smaller percentage in batting average can have a bigger impact on seasonal performance.)
Let’s take a look at numbers for some of history’s great base-stealers, or at least the guys that get talked about:
Rickey Henderson: 80.8%
Lou Brock: 75.3%
Tim Raines (Sr.): 84.9%
Jackie Robinson: 75.1%*
Vince Coleman: 80.9%
Willie Wilson: 83.3%
*Jackie Robinson’s Caught Stealing numbers are incomplete in two seasons, and his percentage is based only on seasons where full data is available. Ty Cobb and Honus Wagner were left off this list as very limited Caught Stealing numbers are available for their era.
Some of these numbers were somewhat surprising to me. In general, most of these players’ percentages actually improved later in their careers, as they slowed down and were forced to become more selective. Raines, for instance, was actually always selective, but Henderson was not - he lead the majors in times Caught Stealing five times in his career, all in his first eight years, and his success rate was only 75.6% in his famed 130 SB year. Similarly, Coleman lead the NL in CS in three of his first four years before significantly improving his percentages by the time he got to the Mets and Royals. Not surprisingly, Coleman was considered a bust at this point in his career, even though in four of his six years with the Cardinals, Coleman’s base stealing actually cost the team runs.
In the meantime, Ellsbury’s career success rate of 85.0% (including his pickoff in last night’s game) is better than any of the great base stealers above. And his percentage may actually improve as his career goes on as he learns to be more selective and is forced to compensate for slowing speed years down the road. It wouldn’t be surprising to see him retire well ahead of the others on the list.
And what about Morgan himself? After all, he was no slouch on the basepath, and is actually 11th all-time in career Stolen Bases. Jon Miller asked him this very question last night, to which Morgan responded (paraphrasing), “I’m not sure of the exact number, but it was definitely 80%, definitely over 80%,” set up of course to make sure that he would be in the “good” category. Morgan turned out to be correct - his lifetime percentage was 81.0%. Of course, had only one call at second been flipped the other way in every other full season in the bigs, Morgan’s lifetime percentage would only be 79.9%. Not so “good” anymore.
