February 14, 2013
In A Pickle
Not All Stars Are All-Stars
It's February, so perhaps it's not timely to write about the All-Star Game, but blame Sam Miller, who raised on Effectively Wild a few weeks ago the Sandy Alomar Conundrum: specifically, how did such a mediocre player manage to appear in six All-Star Games?
You know well that the selection of All-Star rosters is weird and fouled up in all sorts of ways. The popularity-contest aspect has been around as long as fan voting has, and even before that was the method du jour, you might still expect that fame played an important role because the Game had to sell tickets. Managers leaning toward their own players for reserve roles is an entirely understandable decision ("Coach, why'd you take Smith over me?") and yet one that doesn't necessarily result in the best possible roster. Starting pitchers are now routinely passed over if their spot in their real team's rotation interferes with their ability to pitch in the Game itself. Relief pitchers not named Mariano Rivera populate the rosters in a weird homage to the idea of relievers as uniquely special and valuable players rather than simply failed starters. Good measures of total player value tend to be ignored come All-Star time in favor of raw offensive stats and good players sometimes get shafted because some random hamster had a great first half that he'll never ever repeat in his life.
On and on we could go, I'm sure, but that's not the point. The point is merely that it might be fun to think about which players were over- and under-appreciated in terms of All-Star selection due to the distortions of these processes. One way to examine this is by the ratio of a player's career WARP to his career number of All-Star Game selections. Luckily, we have here at BP a highly trained team of database gurus who pulled that very thing for me. (Thanks specifically to the splendid Tim Collins and Ryan Lind for their help here.)
First, some notes on the data:
Second, some context: the mean WARP/ASG for the sample (697 players, 2882 All-Star-seasons) is 7.6, so a player with two All-Star appearances and 15 WARP represents the mean, or three All-Stars and 22.5, or four All-Stars and 30 WARP. Mark Loretta, for instance, or Joe Rudi, or Jimmy Key. These are your average multiple-time All-Star players."
The median WARP/ASG is almost the same: about 7.5, represented by Bill Monbouquette and Chuck Finley. (I'll admit that this is the first time I've ever heard of the gloriously named Bill Monbouquette, who was mostly a Red Sox starter. There's black ink six times on his Baseball-Reference player page: losses once, hits allowed twice, earned runs once, walks per nine once, and K:BB ratio once. The latter two are black ink in the positive way, but the former four are not. But hey, he made three All-Star Games, won 20 games once, and threw a no-hitter in 1962 against Early Wynn, Luis Aparicio, and the rest of the White Sox, so what more do you want? He's in the Red Sox Hall of Fame.)
Third, here are the five highest ratios of WARP to All-Star appearances, which you can read as the most underappreciated players, the ones who racked up value year after year without recognition from the fans, the managers, their peers, or pretty much anyone on earth except a lonely computer to which Colin Wyers is prone to whisper sweet nothings.
Darrell Evans, 71.3 WARP, two All-Star Games
Evans is exactly who you would expect to see at this end of the list. He provided his value via walks (1605 in 10,737 career plate appearances, leading the league twice) and power (.183 isolated power, 414 homers) and fielding (121.3 fielding runs above average), not batting average (.248 career) or RBI (he topped 100 just once). His best years came in the service of bad teams: the '78 Giants won 89 games, finished third in the NL West, and represented the high mark for team records in his seven seasons of five WARP or more. (He did play well for the '87 Tigers, but he was 40 years old by then and marooned in the same league as Wade Boggs and George Brett.)
In Evans's five best non-All-Star seasons (determined by his season-end WARP), he was beaten out by Joe Torre, Ron Cey (twice), Pete Rose, and George Brett for the starting spots. That's a list Evans shouldn't feel too ashamed about losing to (Rose aside), but the crew of backups who made the roster over him is just as illustrious: Ron Santo, Mike Schmidt, Bill Madlock, Cey, Wade Boggs, and Paul Molitor. The worst player on that list, Madlock, had 38.2 career WARP himself and in any case was the sole Cub chosen for the '75 All-Star team.
Evans, in short, saw it all stack up against him: his personal value came in underappreciated ways, his teams didn't do him any favors, and he had to compete against some all-time great third-sackers for recognition.
Really, it's a shame.
Willie Davis, 66.2 WARP, two All-Star Games
Davis was an excellent defensive center fielder (just shy of 100 FRAA for his career) and good basestealer (398 at a 75 percent clip) who didn't put up big offensive numbers that drew the eye. It didn't help that he frequently played in brutal offensive environments—in 1968, his batter park factor was a shocking 86, and his parks for his career came out to 6 percent below average in run-scoring.
The NL outfield was also loaded during Davis' prime. From 1961 to 1970, the first 10 full years of his career, he never had a WARP below 2.3, he batted at or near .300 a handful of times, and he stole bases, as mentioned before, but there was this guy Willie Mays playing center at the time. You may have heard of him. By 1961, Mays had already been an All-Star seven straight years and no Southern California upstart was going to unseat him.
Even as a backup, the All-Star lineup was tough to crack. Here were the 1961 outfield All-Stars (for the first of two games that year): Orlando Cepeda, Willie Mays, Roberto Clemente, Hank Aaron, Stan Musial, and Frank Robinson. Yow. By 1970, things weren't much better: Mays, Aaron, and Clemente were still around and joined by Pete Rose, Cito Gaston, Rico Carty, and Rusty Staub. It's not as illustrious a list, but Staub and Gaston were the only representatives from their teams, and Carty, a starter, was batting .400 as late as June 17th. Pete Rose, sadly, was Pete Rose.
Such a shame.
Bert Blyleven, 63.7 WARP, 2 All-Star Games
I don't have much to say about this. He was a pitcher. Do you know what kind of random-ass pitchers made All-Star Games while Blyleven, from 1974 to 1984, was putting up an ERA 24 percent better than league average, completing over a third of his starts, striking out 1824 batters, and so forth? Here's one pitcher from each of those years: John Hiller, Steve Busby, Bill Travers, Jim Kern, Ben Lindbergh, Ross Grimsley, Mike LaCoss, Jim Bibby, Britt Burns, Mark Clear, Matt Young, and Bill Caudill. (I put in one ringer to see if you'd notice. I bet you can't pick him out.) I don't want to be mean to these players. Most of them had good years and are/were surely magnificent human beings. But gracious me, we couldn't find room for Bert Blyleven on at least a handful of those teams?
It's a damn shame.
Bobby Abreu, 59.7 WARP, 2 All-Star Games
Abreu is interesting because by FRAA's lights, he was not a good defender, posting just four above-average seasons out of 17 and winding up over five wins below average with the glove. He also played a corner spot and didn't rack up anything worth talking about on the bases despite being a pretty good basestealer. All Bobby Abreu did was hit. (Maybe, by the way, Abreu won't like me speaking about him in the past tense, and he might actually have something left at 39, but it's not clear that any team cares to find out.)
Abreu, of course, built his value on singles, doubles, and walks on walks on walks (.292 career average, .396 OBP). On the one hand, that's the profile that doesn't get recognized because there aren't sexy home-run numbers, there aren't .340 batting averages, and so forth. On the other hand, Abreu scored 100 or more runs eight times (and thrice more was over 95) and had eight 100-RBI years as well. (Five of those eight overlapped.) He batted over .300 six times. He stole 30 or more bases six times. Abreu played in a high-offense era, so those numbers may look better to us now than they did at the time, but if you go by rank instead of by round numbers, Abreu was top 10 in runs six times, doubles five times, RBI twice, and steals seven times. It was, I think it's fair to say, extremely obvious how good a player Bobby Abreu was. Hell, he got MVP votes in seven different seasons!
Let's play the Blyleven game. Here are some outfielders who made the All-Star team in the five years in which Abreu got MVP votes but was not on the All-Star roster: Brian Jordan, Cliff Floyd, Preston Wilson, Alex Rios, and Nelson Cruz. We're very close to Abreu's career, so saying "Nelson Cruz!" perhaps doesn't strike anyone as weird right now, but I have a feeling that in 20 years, we'll look back at him as no different from Mike LaCoss. Just a guy who, if we look deeply, probably wasn't undeserving of being on the All-Star roster, but who was not an all-time memorable talent like Abreu or Blyleven, either.
Gosh, what a shame.
Jose Cruz, 57.9 WARP, two All-Star Games
This is Jose Cruz the dad, not Jose Cruz the son. The Jose Cruz who piled up 81.9 FRAA as an outfielder and who managed a .292 career True Average despite a .420 slugging because the Astrodome attempted to bring total obliteration of the spirits and souls of Houston hitters.
It's hard to get as upset about Cruz as it was some of the other players above, though, in part because he was less excellent (6.9 career-high WARP, four seasons over 5.0) than he was relentlessly good (14 years of 1.6 WARP or more). Cruz did finish in the top 10 in the MVP voting in 1983 and 1984, though, winning the Silver Slugger at his position each season, without making the All-Star team. It's worth noting, on the other hand, that in 1984, Cruz hit .288/.350/.421 in the first half and .339/.416/.509 in the second.
For sha— well, no, not really.
Fourth, a summary. I doubt the above list has revealed anything to you, the thinking fan, but it's another lens on the questions of proper ratings and whence underappreciated value arises. Or at least used to arise. On-base ability separate from batting average, hitting in tough parks, pitching for bad teams, and defense at key positions have all in the past (and likely still do today, though hopefully to a lesser extent) been skills exhibited by top-notch players who the public and perhaps even those inside baseball to a degree have treated as merely above average.
Fifth, in near-bullet fashion, here are some other notable notes from the data:
The rest of the top 10 in WARP per All-Star Game: Roy White, John Olerud, Brian Giles, Robin Yount, and Vada Pinson.
Best player to make only three All-Star teams: Yount
Best player to make only four All-Star teams: Jeff Bagwell
Best player to make only five All-Star teams: Frank Thomas
Finally, as a tease for a later article and one more responsive to the original kernel of an idea for this piece (remember Sandy Alomar), the worst player to make 10 or more All-Star teams (non-relief-pitcher category): Steve Garvey.