How accurately does the number of All-Star teams a player makes in his career measure that player’s talent? When people use All-Star teams as any kind of meaningful metric, my built-in, shockproof BS detector goes wild, not least because people (and even some non-people, such as forces-of-nature like Bill) will sometimes cite the number of All-Star appearances in assessing players’ greatness (or lack thereof) though it’s really the crudest sort of approximation, subject to so many variables that have little to do with talent: which era a player plays in, who his teammates are, what position he plays, who his contemporaries are at his position, who’s managing the team, how beloved (or detested) that player is by fans, how long his career lasts…the list goes on and on.
Some of these factors even out, I suppose, but others may multiply, and I’m sure there are deserving players who never made an All-Star team as well as players on multiple All-Star teams who did not deserve their fates. I’ll leave nominations for The Best Player Never to Make an All-Star Game and The Worst Player Who Made the All-Star Game Repeatedly open for your comments, but Tony Phillips and Bobby Richardson leap to the forefront of my mind. Phillips had so much better a career than Richardson, almost ridiculously so, having had a full career’s worth of seasons that each rank higher than Richardson’s best year. I’m quite sure there are better examples in each category—Phillips and Richardson just pop into my head. My point is that using career All-Star squads as a meaningful statistic is akin to using a brick as a fly swatter—you may often hit what you’re aiming for, but you’re going to make a god-awful mess in the process.
An anecdote that I find more funny than instructive does teach a little bit, too. I just perused some All-Star rosters from the 1960s, a period when I was only slightly less focused on Major League Baseball than a German Shepherd is on a piece of steak dangled three feet above his nose. One year, according to Baseball-Reference, the AL featured a player on their All-Star team named Don Leppert, a name that in 2015 meant absolutely nothing to me. No recognition at all. I couldn’t even guess what position Leppert played, so I looked up his stats. (I'd thought they sent Lepperts into special colonies so no one else would catch the disease.) Turned out his stats weren’t all that bad: he scored 46 runs and hit 15 HRs, so okay, maybe that does warrant an All-Star selection. Problem was, though, that those aren’t his stats for half a season, and they’re not even his stats for the entire 1963 season, which was his All-Star year. They’re Leppert’s career stats. The guy played parts of 4 seasons, and I mean "parts": he never once played half of his team’s games. That year, there were 16 AL catchers (he played catcher, as it happens—everyone who knew that, raise your hand) who caught more innings than Don Leppert did, some of them with pretty good careers, most of them having zero All-Star games to their credit. Now, I’m not arguing for a second that these guys deserved that All-Star spot while Leppert didn’t, though it would be hard to argue that Leppert had a career half as good as Gus Triandos, Buck Rodgers or a few other of these guys, but I am arguing that being selected for an All-Star Game or two isn’t always a good measure of players’ quality.
Leppert made the 1963 AL All-Star team, of course, not because he was among the hundred best players in the league that year, or ever, but because he played for the Senators, who weren’t exactly awash with talent. According to the rules, however, Washington got to claim one spot on the All-Star team, so Leppert got that spot. Maybe it also worked in his favor that there were only two catchers in the AL, the Earl of Battey and the Elston of Howard, who were having a distinctly better first half of the season than Leppert--they were the only two to catch more than 1000 innings that year, and they cleared that figure easily. Battey was selected as the starting catcher, and Howard, his backup, was the league’s Gold Glover and eventual MVP, but no other AL catcher had a decent year. According to Baseball-Reference, Leppert’s season ranked 9th in WAR on the Senators’ 1963 roster, with a whopping 0.2. (Sorry, Don, if you’re a BJOL subscriber, or if you’re Googling your name and come across this passage—I don’t mean anything personal in denigrating your career like this. I’m sure you’re a nice guy, just about to turn 84 years old next week, and deserve this abuse even less than you deserved your All-Star team selection in 1963.) There are all sorts of other problems in using All-Star games as a crude metric of quality, of course, such as the undeserving All-Stars chosen by their own managers out of gratitude or loyalty or some other virtue. All admirable human traits, yes, but not exactly what we’re looking for in using All-Star squads as a measure of quality.
But those other problems are not an injustice I need to whine about here. (Feel free, of course, to whine away in the comments section below—I’ll be interested in reading your nominations for most unjust All-Star selections and omissions.) Rather, I’m going to propose a few new wrinkles in the selection process that should make the process a little more rational, if not actually more fair, and a lot more fun.
I would have different groups of people vote at different times during the season. At the very start of the season, the first vote could actually take place, a vote to elect captains of the All-Star squads. Nearly every All-Star team has a player or three who, by merit of his play that season (or half-season) might not deserve selection, but who, whether he’s played in numerous prior All-Star games or not, is well-thought of, popular and, well, just flat-out belongs on the All-Star squad, however well or poorly he’s playing just now. Team would nominate one veteran player each, loosely defined, whom they think deserving of this honor, and the players on each opening day roster would vote for this year’s captain. (I like the notion of players being ineligible to vote for their teammates, and would institute this as a rule in players’ voting.) This selection would take place in the final days of Spring Training.
A function of spreading the vote out over time and between different bodies of voters is to ensure fairness and to generate publicity. There would be a little buzz, don’t you think, about who this year’s captain is going to be. Newspaper columns could promote and denigrate certain candidates, argue why or why not Candidate X should be chosen over Candidate Y, and we’d get the result just as the season is beginning. It's a feel-good story.
Then we’d start the fans voting for a starting lineup. To prevent ballot-stuffing abuses, we can hold a variety of different sorts of fans’ votes. (Bill originally proposed a "precincts" model a few decades back. This is by way of updating that idea.) One, an online precinct, where fans can stuff the ballot to their hearts’ content, using as many different email addresses and multiple ballots as they like. Two, (or in practice, over two dozen precincts, one in each ballpark), ballots distributed at the ballpark, valid only on the day of the ballgame, to reward those fans who actually come out to the park, especially in cold spring weather. Again, multiple voting is allowed, but only one ballot will be distributed with each ticket: buy thirty tickets to April and May games, you get thirty votes. (I’d also print ballots limited to the league the particular ballpark is in, on the assumption that these are the fans most familiar with the players they’re voting for.) Three, I’d have a ballot sent (ONE ballot) to each household subscribing to MLB.tv. (In the case of bars, barbershops, restaurants etc., I’d encourage them to hold their own internal votes among their customers to vote on how the establishment’s ballot be filled out.) On the on-line ballots, I’d update each player’s stats to reflect that player’s stats, and WAR, over the last 162 games, just so fans relying on last season’s stats for convenience, will be able to avoid over-dependence on out-of-date data. And add a few more such sorts of precincts of different groups of fans (ballots printed in local newspapers, maybe some fantasy-baseball sites, like that), the governing idea being that if you’ve got enough precincts, of different sizes, each one’s preference would count only once, so there's little sense in rigging any one precinct in particular.
With several dozen precincts, each would select nine All-Stars, but it would be damnably double difficult for any one precinct to stuff all the ballot-boxes. If Job Lowe wins the Cincinnati or the Kansas City ballpark vote (and why are Midwesterners so prone to ballot-stuffing anyway? I thought they were decent, honest salt-of-the-earth types), but Moe Joe wins a majority of the other precincts, what have Job Lowe’s devoted fans accomplished? They’ve wasted their ballots, true, and with far more effort expended (and a lot less fun) than if they’d simply made paper airplanes out of the ballots and threw them at each other. The idea is to get different groups of fans involved in the voting. Again, I’ll be interested in seeing your suggestions for more possible fan-precincts that I haven’t thought of.
We could conclude this fan voting fairly early on in the process. Fans would be choosing, as they do, based on reputation, which means they would be choosing players largely on last year’s stats (and this year’s presence on the roster), both of which would be known in April and May. Since some ballots would need to be printed in advance, it would be difficult to ensure that late roster changes, injuries, etc. would be reflected in the voting. The next round of balloting in early June would reflect recent developments—surprise starters, rookies, part-timers emerging into full-time players—rather than asking fans to keep up with events as they occur. This next round (closing around mid-to-late June) would be a players’ vote (again, on players in their league, but not on their teams). By mid-June, when they’ve seen every team play against them, if there’s a rookie or a player filling in for an injured teammate or someone having a breakout year, the players will be best equipped to account for that—it’s not as if some very impressive player would go unnoticed by the people whose ears he’s been beating in for the past three months. Each active player would get a ballot consisting of every player in the league, minus those already chosen as captains and by the fans (and those on their own team, of course) and they would elect four pitchers and four non-pitchers.
With 18 players selected (one captain, nine starters, and eight more players chosen by the players), I’d turn the next round over to a group of respected savants: One BBWAA rep, one broadcasting rep, one freelance stats guru (like Rob Neyer or Tom Tango), etc. This committee of five to ten savants would serve on a one-time-only basis (to ensure a minimum of tradeoff-type deals—"Vote for my guy this year and I’ll vote for yours next year" kind of thing) and would actually meet, in person, to discuss their candidates, argue their cases, and arrive at a consensus of the five best players in the each league not already chosen. You could even schedule the discussion as a two-hour special live show on ESPN. I’d certainly tune in to hear this discussion: it would be like "PTI" or "Highly Questionable?" only this hour or two of TV-blather would actually have consequences. This committee of savants meets the final week of June. That gets us up to 23 All-Stars.
The first week of July, we ask every living former All-Star player (I suppose at any point in time, there are probably close to a hundred of these around) to select five additional players. All of these different bodies, by the way, would be correcting the previous groups’ mistakes and oversights, so no one could complain that the All-Star team’s voters had a blind spot—if several groups each get a shot to compensate for previous groups’ outrageous omissions, well, I guess they’re not that outrageous, after all. And then, a week or so before the game, the manager and his coaching staff (which consists, as it does now, of managers from teams other than his own, of course, preventing the manager from doling out All-Star spots to his personal favorites) can select two more players who can best balance the team out, in the event that the process has shorted the team at any given position. They can also vote on replacements, a few days before the game takes place, to account for any last-minute injuries or players who decline the honor of playing in the game. Personally, I’d do away with the requirement that every team has a representative (I think it’s a silly and condescending rule) but if you really want to keep that rule (can we call it the Don Leppert Rule?), you could also require that the managers and coaches include any overlooked teams in these last-minute selections,
That’s thirty players, a goodly number given the modern number of teams. Before moving on to the Game itself, let’s address the issue of players declining to appear in All-Star games. I wouldn’t allow it. If you’re voted on as an All-Star, you’re going to the game. Not badly injured players, of course, but I would declare a mandatory and immediate 15-day disabled list appearance for anyone who declines to participate in the game. A player who’s hurting but doesn’t want to go on the DL, can opt to ask his A-SG manager to leave him on the bench, but he’s still going to have to show up at the ballpark, take BP, pose for pictures, do interviews, etc.—who’s going to beg off and still go to all that trouble without playing in the game unless he’s really hurt?
Many players are contractually rewarded for being chosen as All-Stars, though their contracts don’t specify that must actually play in the game to collect—I suppose that’s why it’s so common for recent players to prefer the three days off to the honor of playing. Sorry, but that’s no longer allowed, at least not without taking yourself off your own team’s roster for 15 days, a move that I’m sure the player’s teammates will disapprove of heartily.
My next issue with the All-Star Game is the whole "incentive to win" thing. Standing proud at the very top of my list of "Idiotic Innovations to MLB that I Detest," just above "the DH," is making home field advantage for the World Series depend on the outcome of the All-Star Game. It makes no sense at all to think that All-Stars are going to play harder, or better, or anything, really, just because some team in their league (and the odds are very heavy that’s it’s not going to be their own team) is going to benefit from an All-Star Game victory. Are we even sure that players are always rooting for their own league’s team to win the Series? As a NL fan, I often want the other league’s Series entry to kick the everloving tar out of the NL representative, especially if it’s a team that’s played my team rough or dirty or nasty during the regular season. You think Bryce Harper is sitting in front of his TV this week, cap-backwards, with Cheetos and a pitcher of beer, chanting "Let’s Go, Mets"? More likely, he’s rooting for the Mets’ team plane to crash en route to LA.
Here’s a much stronger motivation for the All-Star team to win the Game: self-interest. Instead of giving a one-game home-field advantage in the World Series to some team that’s probably not their own, let’s make the margin of victory in each year’s All-Star game determine the number of players on next year’s All-Star squad. That is, if the AL team wins the All-Star Game by four runs in 2016, then the 2017 AL squad gets four extra players on the team. This would motivate the players because (since most All-Stars are going to play in the same league next season) each run opens up a whole extra slot, thus increasing each player’s chances of making next year’s All-Star team; as already stated, many players have a bonus clause in their contracts if they make the All-Star team. This also ensures that even in a blowout most teams will be playing their best, and that even in the biggest blowout games no one will stop trying to score runs or to prevent the other team from scoring until the final out is made.
This proviso also means no manager is going to run through his bench and bullpen unnecessarily in a close game, because he will need those players if the game goes to extra innings. Each manager will keep a few starters available if that’s the case, and keep some pinch-hitters ready as well—IOW, it will be much more like a real game than like some stupid exhibition where the goal is to make sure that every player gets into the game, or whatever the hell is going on in the minds of All-Star managers.
OTOH, we don’t want to abuse the players, or bore the fans, so instead of an occasional All-Star Game ending at 2 AM with the score tied and every pitcher on the team having ruined his arm, and several players at positions they haven’t played since middle school, here is how I’d settle ties: after a certain predetermined point (I’d say the 12th inning) of a tied game, the tie would go to (wait for it) the team that won the HR Derby.
Yes, I’d turn that pre-game spectacle (which I personally haven’t watched in, uh, how long ago was "forever"?) into a tie-breaker. You can bet that I’ll pay attention now to that otherwise unwatchable spectacle. Now the players chosen to participate in HR Derby will be hotly contested instead of "Who? He’s in the HR Derby this time? Why? Oh, well, what does it matter?" And you can structure the HR Derby so that it’s far more competitive. Each batter gets to bring a non-All-Star teammate with him to the Game to pitch to him, and they’d work as a mini-team, the batter describing the pitches he wants and the pitcher serving them up at the speed and in the location the batter wants. The batter must swing at every pitch—if he doesn’t, or if he misses it completely, he is eliminated from the tournament, so the pitcher’s control will obviously be critical here. Whichever league emerges from the HR Derby with the winner also wins any game that is tied after twelve innings. That will pack some fannies in the seats.
In fact, you could re-institute many of the pre-game contests that Bill rightly complains about disappearing from the game. They were fun. They prompted excitement on fans’ part, on players’ part. They were a means to get non-fans, people who know nothing about the game, to get interested. So why have these contests of skill been discontinued? Fear of injury? To be fair, in a few cases I can see where the skills demanded do risk injury, particularly the long-throw.
But, hey, don’t outfielders practice long-throwing? I would assume so—otherwise how would they maintain the long-throwing skills they must employ under game conditions? Let’s stipulate we want to eliminate the risky contests. There are plenty of skill contests that aren’t risky in the least that I’d love to see. Wouldn’t you love to see a contest of control pitching, or maybe "Beat The Radar Gun"? Obviously, after the pitchers are thoroughly warmed-up, on a day, say, or two days before the Game itself, when most of them will do some throwing anyway, you’d set up a small target 60’ 6" from the pitching rubber, and you’d see who can hit it the most times and at the fastest speed. Wouldn’t that interest you? You can probably measure all sorts of fancy stuff with radar guns and Questec and the like that you could hardly imagine twenty years ago--so let's employ some of this cutting-edge technology in this game which is supposed to be an entertainment extravaganza.
And don’t fast players practice running all-out? Of course they do. You could have sprints from home to first base, timed down to three decimal places, or longer sprints around the bases. You could feature a lot of these contests, all involving no contact, all involving skills that players must routinely practice all the time. And you could have any of them serving as meritorious tie-breakers if you don’t want (who does?) to have marathon All-Star games. I’d sure be curious to see which player can run from home to second base the quickest, or which pitcher can knock a milk bottle off a stool with a 92 MPH fastball.
Maybe these tweaks would make All-Star rosters a little more rational than they are now, and more fun, and there would be more merit in citing players’ appearances in All-Star Games as a measure of their quality. Maybe I’ve just gotten blasé or bored with the spectacle after a lifetime of watching it and caring who won, but I remember when it was one of the high points of my season. I can easily imagine it being so again.