Splits and Stolen Bases

Ideas spring from odd places at times, and this particular idea I owe to a rather spirited Millennial with whom I shared an adult beverage in the not so distant past. It just so happens we watched Austin Jackson steal a base and I overheard him mutter something to the effect of “don’t get used to that” — which, considering the sprained ankle Jackson suffered just days after, perhaps Jackson’s spiritual adviser might be interested in a chat. But I digress.

After some discussion, his point was less about Austin Jackson than about stolen bases in general. Stolen bases, he posited, fall off in the second half of the season. I told him I was unaware of such a phenomenon and asked if he had seen research to that end. Indeed, just look at 2014 said my young budding analyst. So I did. If stolen bases were less common in the second half of the season, this could have implications for standard rotisserie league strategy, inasmuch as your approach to your place in the standings.

If you take a peek at 2014 speedsters, it does indeed appear that he is kinda-sorta-maybe on to something. If you look at the actual stolen bases the number is almost always going to be lower, because duh, the second half is shorter (thus not really a half at all. It’s just the second part. Sigh, digression). Therefore, with the statistical wizardry help of one Jeff Zimmerman, below are your 25 stolen base leaders from 2014, sorted by the difference in stolen bases per game in the second half.

name	Season	Split	SB	G	SB/G	Split	SB	G	SB/G	total	Difference
Jarrod Dyson	2014	1st Half	18	59	0.20	2nd Half	18	48	0.38	36	0.18
Lorenzo Cain	2014	1st Half	12	69	0.13	2nd Half	16	61	0.26	28	0.13
James Jones	2014	1st Half	17	65	0.19	2nd Half	10	33	0.30	27	0.12
Carl Crawford	2014	1st Half	10	46	0.11	2nd Half	13	58	0.22	23	0.11
Ian Desmond	2014	1st Half	10	91	0.11	2nd Half	14	63	0.22	24	0.11
Carlos Gomez	2014	1st Half	17	87	0.19	2nd Half	17	60	0.28	34	0.10
Denard Span	2014	1st Half	15	84	0.16	2nd Half	16	63	0.25	31	0.09
Michael Brantley	2014	1st Half	10	90	0.11	2nd Half	13	66	0.20	23	0.09
Ben Revere	2014	1st Half	26	85	0.29	2nd Half	23	66	0.35	49	0.06
Jose Reyes	2014	1st Half	17	79	0.19	2nd Half	13	64	0.20	30	0.02
Leonys Martin	2014	1st Half	18	90	0.20	2nd Half	13	64	0.20	31	0.01
Jason Kipnis	2014	1st Half	13	68	0.14	2nd Half	9	61	0.15	22	0.00
Jacoby Ellsbury	2014	1st Half	24	90	0.26	2nd Half	15	58	0.26	39	-0.01
Jimmy Rollins	2014	1st Half	19	90	0.21	2nd Half	9	47	0.19	28	-0.02
Charlie Blackmon	2014	1st Half	18	93	0.20	2nd Half	10	60	0.17	28	-0.03
Rajai Davis	2014	1st Half	24	73	0.26	2nd Half	12	56	0.21	36	-0.05
Starling Marte	2014	1st Half	21	82	0.23	2nd Half	9	52	0.17	30	-0.06
Brett Gardner	2014	1st Half	15	88	0.16	2nd Half	6	57	0.11	21	-0.06
Alexei Ramirez	2014	1st Half	15	95	0.16	2nd Half	6	63	0.10	21	-0.07
Dee Gordon	2014	1st Half	43	91	0.47	2nd Half	21	57	0.37	64	-0.10
Elvis Andrus	2014	1st Half	20	94	0.22	2nd Half	7	63	0.11	27	-0.11
Alcides Escobar	2014	1st Half	22	94	0.24	2nd Half	9	68	0.13	31	-0.11
Billy Hamilton	2014	1st Half	38	86	0.42	2nd Half	18	60	0.30	56	-0.12
Eric Young	2014	1st Half	25	62	0.27	2nd Half	5	36	0.14	30	-0.14
Jose Altuve	2014	1st Half	41	93	0.45	2nd Half	15	65	0.23	56	-0.22

So 12 players had higher stolen base rates in the second half and 13 had lower stolen base rates. But what jumps out to me is that four of the top five total base stealers had lower rates in the second half, and considering Hamilton and Altuve, significantly so. This really doesn’t strike me as a huge surprise, considering it’s a long season and maybe there’s some fatigue in them legs. However, no real smoking gun here. But then again, this is just 2014. So I decided to look across a few more seasons, drawing on data back to the year 2006 season, looking at every player to have ever stolen a base in the first half and the second half. And the results were a little unexpected.

Since the 2006 season, there have been 14,829 stolen bases in the first half over 219,456 games played for a ratio of .067 stolen bases per game. In that same time span, there have been 11,549 stolen bases in the 2nd half of the season over 174,168 games played for a ratio of (drum roll) .066 stolen bases per game. Basically, no difference at all. Sorry little buddy, next round is on me.

What does this mean? Probably nothing — but piqued enough is my curiosity that it might deserve a deeper look. Because the grand totals include everyone, pitchers and non-pitchers alike and in that bubbly milieu the real story might just get lost in the noise. If it turns out that burner-types are less likely to keep pace in the second half, then it means stolen bases are coming from new, or different, sources. In the short term if this stability should hold up in 2015, it means you might not be completely safe if you’re killing it in the stolen base category — or alternatively, completely screwed if you’re in the cellar. In part two, I’ll take a look at top base stealers from, say, 2010-2013 to see if it tells a similar story to 2014. Stay tuned.