August 26, 2016

Prospectus Feature

Coleman/Hamilton, Pt. 1: What We're Missing Out On

by Rob Mains

At this year’s Saberseminar, I was out for dinner and one of the people in our group asked what contemporary player we think we could tell our grandchildren, “I saw him play.” Mike Trout was a gimme, but we had a hard time coming up with someone else. My thought was that Billy Hamilton would be such a player, but he was born about 30 years too late.

I know, Billy Hamilton is not a great player. He’s having his best season, and his career on-base percentage is still below .300. And it’s not like he compensates with power: Of the 172 players with 1,200 or more plate appearances from 2014-2016, his .088 ISO ranks 164^th. He does play a good center field, but he’s got a .236 career TAv. That’s a lot of bad bat to carry with a glove.

But he can run. Man, can he run. Through games of August 16, when he was sidelined by a knee contusion, he has 29 stolen bases. Since the All-Star break. The only players, other than Hamilton, with that more swipes all year to that point were Jonathan Villar, Starling Marte, and Rajai Davis. And that doesn’t include plays like this:

Yet, there could be more, couldn’t there? When I think of Hamilton, I think of Vince Coleman, who played 13 seasons, mostly for the Cardinals, from 1985 to 1997. Here’s a comparison of Coleman and Hamilton during their first three seasons in the majors (1985-1987 for Coleman, 2014-present for Hamilton):

	G	PA	R	BB	SO	BA	OBP	SLG	OPS+	ISO	TAv
Coleman	456	2,064	322	180	339	.263	.329	.325	79	.062	.246
Hamilton	376	1,485	192	94	276	.246	.294	.334	72	.088	.234

Coleman was a better hitter (though without any power—his .062 ISO was last among 99 players with at least 1500 plate appearances during those years, and he was a corner outfielder), but he played in a better run environment. The National League averaged 4.25 runs per game in Coleman’s first three years; it’s averaging 4.15 so far in Hamilton’s. So while Coleman has the better slash line, their normalized batting stats are closer, and when you add Hamilton’s defensive play, he’s been, by WARP, a more valuable player in less playing time.

But there’s a big difference between the two:

	Starts	Batting 1	Batting 2	Batting 7	Batting 8	Batting 9
Coleman	444	444	0	0	0	0
Hamilton	346	219	30	20	4	73

The Cardinals had Coleman on their lineup card in all but 41 of their games during his first three seasons. And every single time, he led off. With his .329 on base percentage, in a league where non-pitchers had an OBP of .333. And he ran wild when he got on base. Coleman stole 110, 107, and 109 bases in his first three seasons.

If you don’t remember him, Coleman really was a sensation. He was a unanimous choice for NL Rookie of the Year in 1985. (You could have made a case for Reds pitcher Tom Browning or Giants third baseman Chris Brown, but Coleman wasn’t a bad pick.) Considerably more amazingly, he finished 11^th in the MVP voting in 1985, a year in which his .267/.320/.335 slash line, 110 stolen bases, and 2.4 WARP garnered more support than Tim Raines (.320/.405/.475, 70 stolen bases, 6.9 WARP), Ryne Sandberg (.305/.364/.504, 26 home runs, 55 stolen bases, 5.7 WARP) and Orel Hershiser (19-3 in 239 2/3 innings, 2.03 ERA, 2.67 FIP, 2.84 DRA, 6.8 WARP). Coleman finished 12^th for MVP in 1987 as well, with .289/.363/.358, 109 stolen bases, and 3.5 WARP, outpolling Mike Schmidt (.293/.388/.548, 35 home runs, 7.0 WARP) and Pedro Guerrero (.338/.416/.539, 27 home runs, 6.3 WARP). The 1985 World Series will be remembered for Don Denkinger’s call in Game Six, but the big story going into the Series was Coleman’s absence due to being run over by the tarp at Busch Stadium. People with a straight face argued whether Coleman or Rickey Henderson was the game’s top leadoff hitter. (I should mention here that Coleman was fifth in the National League in runs scored in 1985 and 1986 and second in 1987, his three top finishes. Henderson led the American League in runs six seasons, including both of Coleman’s first two years.)

Hamilton, by contrast, has started only 344 of the Reds’ 450 games since the start of 2014. Of them, he’s led off only 218 times, 63 percent of the time, which is 37 percent less frequently than Coleman. Yeah, he’s got that .294 on-base percentage, well below the .325 non-pitcher league average, but Coleman was below average, if less so, as well. What if Hamilton had been used like Coleman?

Let’s start with his plate appearances. As noted, Coleman led off in 444 of the Cardinals’ 485 games in his first three seasons, or 91.55 percent. Through August 24, the Reds had played 450 games since the start of 2014. In that time, Reds leadoff hitters have amassed 2,063 plate appearances. Let’s grant Hamilton better health, for fun, and lets let Whitey Herzog fill out the lineup each day, and give Hamilton 91.55 percent of his team's leadoff spots. That would give him 1,889 plate appearances.

Using the same proportions as above, Hamilton would have 323 singles, 120 walks, 4 hit by pitch, 13 reached on error, and 66 doubles. In 1985-87, Coleman had 410 singles, 180 walks, 5 hit by pitch, 24 reached on error, and 47 doubles. I’m going to use the Baseball-Reference definition of a stolen base opportunity as every time a player’s on first or second with the next base open. Coleman, in his first three years, had 652 stolen base opportunities out of the 666 times he got himself to first or second, plus the 239 times he got from first to second by way of a stolen base. So he had a stolen base opportunity 72 percent of the time he made it to first or second. (Yes, I know I’m excluding times when, e.g., he was on first, didn’t steal, and was moved over to second, creating two stolen base opportunities. That’s okay; as you’ll see, I’m looking for a rough proportion, not an exact figure.)

Let’s assume Hamilton had the same opportunities. He’d have made it to first or second as a batter 526 times. He also has 117 steals of second. That’s a total of 643 times that he made it to first or second. Multiply that by 72 percent and you get 463 steal opportunities for Hamilton if he’d been used the way Coleman was.

But that’s not all. Coleman attempted 387 steals in his first three seasons, equal to 59.4 percent of his steal opportunities, and was successful in 84.2 percent of his attempts. Hamilton’s had a similar success rate, 81.4 percent (88.0 percent over his past two seasons), but he’s tried to steal in only 48.3 percent of his opportunities. So bump that up to Coleman’s 59.4 percent, and apply Hamilton’s 81.4 percent success rate, and you’ve got Hamilton stealing 224 bases from 2014 to 2016. And if Hamilton had stolen bases at Coleman’s clip, he’d find himself on second base following a steal more often, with 158 steals of second instead of 117. Those additional stolen base opportunities add another 20 steals for a total of 244 stolen bases in two-and-change seasons.[1]

So Billy Hamilton, were he deployed by the 2014-16 Reds the way the 1985-87 Cardinals used Vince Coleman, would have stolen 87 bases in 2014, 89 in 2015, and 68 so far in 2016. Granted, that’s not quite at Coleman’s clip, but those are big numbers. Way bigger than his totals of 56 in 2014, 57 in 2015, and 53 so far in 2016. If he could keep up a pace like that for a few more years…yeah, you’d want to tell your grandkids about seeing him on the basepaths.

So why hasn’t it happened? I’ll delve into that early next week.

Rob Mains is an author of Baseball Prospectus. Follow @Cran_Boy
Click here to see Rob's other articles. You can contact Rob by clicking here