20 April 2007

Jeter Closest Thing to Jackie Robinson in Today’s Game

Baseball celebrated the 60th anniversary of Jackie Robinson's first game in the major leagues on Sunday with a bizarre numbers game, in which 2,347,629 different players were all allowed to wear Robinson's retired #42 that day. Interestingly enough, all box scores also were altered, so that every position on the field was considered "second base". Not surprisingly, official scorers recorded a record number of "4-unassisted" double plays.

OK, not really.

Another way to celebrate the greatness and unique nature of someone like Jackie Robinson is to try to compare him to some of the modern game's great players. More accurately, you can discuss how difficult it is to find a comparable player in today's game, and instead describe an amalgam of some of the best skills from several of today's players, as Rob Neyer has. He suggests that Robinson would hit with Miguel Cabrera's batting avearge and patience, if not quite so much power, but would play defense at the Keystone with the acumen of Orlando Hudson or Pokey Reese, and would steal bases as well as Chone Figgins. Diamond Mind Baseball simulated his 1951 season in today's game, and in 2006, they suggested Jackie would have hit roughly .354/.439/.565, with 53 doubles, 23 homers, 138 runs and 120 RBI's, and would steal 46 bases as well.

Real Jackie 153 548 113 33 19 88 79 27 25 8 9 .338 .425 .527 .953
DM Jackie 159 607 138 53 23 120 82 86 46 16 17 .354 .439 .565 1.004

Holy crap. Jackie wouldn't just be great, he'd be the best player in the league. Of course, how he manages to get 70 more plate appearances while playing only six more games, I haven't figured out yet, but those numbers sure look cool, don't they?

Anyway, for the sake of context, last year, Ryan Howard won the NL MVP award, and with his stellar campaign, Baseball Prospectus says he garnered 9.5 Wins Above Replacement Position (WARP), a very good number. Albert Pujols was almost half again as good, with 13.1, mostly because he made fewer outs than Howard and played much better defense. And Jackie? Well, numbers like the ones Diamond Mind generated would give him about 15 WARP3 (adjusted for all time). Nobody else in baseball was particularly close to that number in 2006. In fact, only some of the most stellar seasons of all-time have ever approached that number. Some of the best seasons of Barry Bonds, Rogers Hornsby and Ted Williams have surpassed that, but the best efforts of Ty Cobb, Willie Mays, Hank Aaron and Mickey Mantle never did. Was Jackie Robinson really better than those guys?

Obviously, we're arguing something we can never really know, not without a Way-Back Machine or a Delorean with a flux-capacitor in it, but is it reasonable to expect that Jackie Robinson would be able to not just compete with today's players, but to dominate them? At the risk of being branded a racist or something worse, I'm going to suggest that Jackie Robinson would not be so great today.

For me, at times like this, I always go back to the well: Baseball Prospectus. Their Davenport Translations for Robinson's 1951 season aren't quite as generous (14.1 WARP). It should be noted that BP's adjustments are for all-time, though, not just to the 2006 NL, so that should have some effect as well, though I can't say what.

Real Jackie 548 113 33 19 88 79 27 25 8 .338 .425 .527 .953
BP.com Jackie 543 104 41 27 89 74 41 40 10 .335 .426 .575 1.001

He gets a few more homers, but not as many walks, steals or doubles as Diamond Mind was ready to give him. One of the major differences between Diamond Mind and BP is the strikeout totals, 41 for BP, compared to 86 for Diamond Mind. Given that Jackie's actual K total in 1951 was 27, and that strikeouts are issued about twice as frequently now (6.7/game in 2006) as they were in 1951 (3.8/game), I see no reason to believe that Jackie would have only whiffed 41 times. Eighty-six may be a little high, but not much. In fact, adjusting for the differences in the league rates for other stats, as well as the fact that there were eight more games played per team in 2006, we can get a rough idea of how Jackie's stats from the summer of '51 would translate to the 2006 NL:

Real Jackie 548 113 33 19 88 79 27 25 8 .338 .425 .527 .953
TMN-adjusted 580 120 46 27 101 79 50 43 9 .344 .434 .578 1.012

On a cursory basis, it seems to me that the homer total is probably a little high, and the strikeouts are probably a bit low. very few players in today's game can slug .575 or better without more than 50 strikeouts. Pitchers just throw too damn hard these days. Not like back in the old days, when everybody sucked.

Of course, this is just my rough means of adjusting from the 1951 NL to the 2006 NL, and does not take into account the effect of the home park or a myriad of other factors. Baseball-reference.com, however, can do this. In fact, if you're a subscriber, they can take anybody's stats for thier career and adjust them for any year, any league and any park in that league. When I did this for Robinson, I found that his stats for 1951 translate very well to the 2006 NL, but not as well as Diamond Mind or Rob Neyer would have suggested. Here's what they came up with:

Real Jackie 548 113 33 19 88 79 27 25 8 .338 .425 .527 .953
Bball-ref.com 570 116 35 20 96 85 28 27 8 .347 .441 .539 .980

Baseball-Reference explains the algorithm they use here, but if you don't want to read it all, I can summarize by saying that they take the change in league run scoring and use that to back-calculate everything else. I used the changes in rates of the individual stats (2B, 3B, HR, SB, RBI, etc) and then got my percentage numbers form those changes, which explains why Jackie gets more homers, doubles, strikeouts and steals in my adjustment (because the league HR, 2B, and K rates have grown more than the overall run-scoring rate) than in that of www.Baseball-reference.com. Today's all-or nothing, swing-for the fences kind of game lends itself to those things, and to ignore that seems silly to me.

In addition, players and teams steal bases a heck of a lot more nowadays than they did in Jackie's day. His 25 steals in 1951 ranked 3rd in the NL, 4th in MLB, and there were guys in the top ten in each league that stole only 10 or 11 bags. Between 1929 and 1960, nobody in the National League stole more than 40 bases in a season, with the league leader usually in the 25-35 range. Last year alone, half a dozen players in the NL stole 41 or more, and five more players in the Junior Circuit stole at least 40. There's simply no way that Jackie Robinson, in today's game, would steal only 27 bases.

The one place where BR does have me is on park adjustment, because I did not make one, but since Dodger Stadium was essentially neutral last year (park factor of .997 according to Baseball Prospectus, 102 according to Baseball Reference), that wouldn't have much effect anyway. In any case, as you might expect, I feel most comfortable with the numbers I generated myself, however flawed they may be, but that's only half of the story. The other half is to ask who compares well with Jackie in today's game. While there is admittedly nobody with Robinson's combination of bat control, speed, defensive prowess and moderate power, we have a fairly close comparison playing in the major leagues today, and as it happens, he too is an ethnic, middle infielder playing for a team in New York. You guessed it:

Miguel Cairo.

No, not really.

Actually, I'm talking about Derek Jeter. Let me show you:
               G    AB   R    H   2B  3B  HR  RBI  BB   SO  SB  CS   AVG   OBP   SLG   OPS 
TMN's Jackie 162 580 120 199 46 5 27 101 79 50 40 8 .344 .434 .578 1.012
Real Jeter 154 623 118 214 39 3 14 97 69 102 34 5 .343 .417 .483 .900

No question, it would seem that Jackie has a considerable edge in power, with seven more doubles and 13 more homers despite getting 43 fewer at-bats. Jackie's strikeouts are also dwarfed by Jeter's, with more than double the translated amount, but studies have been shown to essentially indicate that an out is an out, so that matters a lot less than you would think. Besides, as I mentioned earlier, I think my translated numbers wind up with the homers too high and the strikeouts too low, but that's more of a "hunch" than anything else. Jackie also steals a handful more bases, but gets caught a few more times, so that's a net wash. Their percentage numbers, other than slugging average, are eerily similar, as both players hit for very high batting averages and walk a decent amount, but not excessively. Runs scored and driven in are within a few ticks either way as well, despite the difference in plate appearances. Miguel Cabrera's batting numbers would have been even closer in some cases, especially in the power numbers, (50 doubles and 26 homers), but he struck out even more than Jeter, and hardly steals any bases at all. Besides, who watches the Marlins? Are we even certain that this so called "Miguel Cabrera" exists? I didn't think so.

Which leaves us with Jeter. He's an excellent hitter for average, with decent patience, great baserunning ability, and moderate power, very much like Jackie. Jeter's also won three Gold Gloves as a Shortstop, though his having earned them is a very debatable premise. Baseball Prospectus inidcates that Jeter's defense at short last year was +7 FRAA (Fielding Runs Above Average), while Jackie was +26 as a secondbaseman in 1951. That's a big disparity too, even bigger if you, like me, don't happen to be a believe in Jeter as a good shortstop, but however it is that BP measures this stuff, they at least got a number comfortably above average for Jeter last year. He may not be excellent, like Robinson was, but "good" may not be too much of a stretch. Jackie even appears on Jeter's list of comporable players (7th) according to Baseball Prospectus, though even that is a modest comparison at best (only a 23% similarity score).

Baseball-reference.com uses Bill James' formula for similarity scores, a very different one, but they have some odd comparables for Jackie:

1. George Grantham

2. Denny Lyons

3. Edgardo Alfonzo

4. Freddie Lindstrom

5. Jeff Cirillo

6. Mike Greenwell

7. Irish Meusel

8. Joe Randa

9. Gregg Jefferies

10. Bruce Campbell

George Grantham? The 'Fonz? Jeff Cirillo? If this guy was one of the all-time greats, why is his list of supposedly "comparable" players riddled with flame-outs like Gregg Jeffries and mediocrities like Joe Randa? Bizarre, isn't it? The toruble here is that Bill James' formula uses career stats, and because Jackie's official major league career didn't get started until he was 28, and because he preferred to hang up his spikes at 38 than to play for the cross-town rival Giants, Robinson's only got about half a career worth of stats. If not for segregation, Jackie could have been in the majors at least two years earlier, though probably not much more than that, because of World War II. And if not for his pride, he might have played another two or three years, into his early 40's. Even as a spot-starter and bench player, Jackie could have padded his stats a bit, at least enough to knock Joe Randa off the list, don't you think?

But we can't do much about that right now. Robinson's legacy, such as it is, will have to be enough. But we can thank him for the privilege of watching Derek Jeter (and other non-whites) play today. He's a worthy successor.

Stumble Upon Toolbar

No comments: