By the numbers: Stats tell story of All-Star selection (VIDEO)
By the numbers: Stats tell the story of All-Star selection
By the numbers: Stats tell the story of All-Star selection
By Russell A. CarletonBaseball Prospectus
It’s time for everyone’s third-favorite part of the All-Star Game, behind the Home Run Derby and the thing where they all line up and the PA guy introduces all the players. Yes, it’s time to complain about the system for selecting the actual players who make the team. You get bonus points if you create a crazy new system that makes things “better.” Or more “fair.” Or at least gets your favorite player onto the team.
If you step back and think about it, it is a weird system. The fans vote for the starters. Then the players vote for each other. Then the managers have their turn, but they have to make sure that they pick someone from each team. And then everyone fakes an injury because they want three days off rather than a trip to Minnesota, and managers have to scramble to find replacements.
And then there’s that free-for-all “final vote” run-off. It seems a little silly, but it at least makes for some good discussion on those horrible days when there’s no baseball.
Then, once they finally play the game, “it counts” to determine home field advantage in the United States and Toronto Series.
Missing from all of the arguments though is the answer to a question: How much do all of these strange rules really matter? Can we put some numbers to it? Are the fans really getting it all wrong? If anyone ever actually did anything about the system as it is now, would it actually make a difference? Let’s take a look at some of the most common objections to the way things are done now, and try to figure out whether it’s really that big a deal.
WARNING! GORY MATHEMATICAL DETAILS AHEAD
Objection No. 1:The fans pick the popular players and the ones who have had success in the past, rather than the ones who are actually good now.
Most Common Retort: It’s not a game for the best players, it’s a showcase specifically for the players that the fans want to see. Also… Derek Jeter.
Here’s a rundown of whom the fans chose compared to who is the best in the league at that position by Baseball Prospectus’s Wins Above Replacement Player measure (WARP, and all numbers in this article are current as of Sunday night). WARP, for those who are unfamiliar, is a stat which tries to incorporate all of a player’s contributions, including his hitting, defense, and baserunning, and compare him to bench and fringe and AAA players who play his same position. The idea is that if a player were to disappear into thin air, how much would a team lose as a result of his absence? It’s not a perfect stat, but it’s the most comprehensive measure of between-the-white-lines value that we have for a player.
Top Vote Getter (AL)
Top by WARP (AL)
Top Vote Getter (NL)
Top by WARP (NL)
Catcher - M. Wieters
First base - M. Cabrera
Second base - R. Cano
Third base - J. Donaldson
Shortstop - D. Jeter
OF - J. Bautista
OF - M. Trout
OF - A. Jones
DH - N. Cruz
Looking through the list, there are several good matches. On the AL side, four of the nine spots were given to the best player by WARP, and on the NL side, the fans went 5-for-8. Of note, at the time I wrote this, Robinson Cano was running neck-and-neck with Jose Altuve and they may easily have flipped by the time you read this. Also, in a fit of cosmic justice, while the fans voted in the out-for-the-season Matt Wieters, Salvador Perez will start anyway.
There aren’t a lot of absolute howlers in there. Derek Jeter is not having an All-Star season, but we all know why he was elected. Maybe the biggest injustice is Aramis Ramirez (near the bottom of the league among starting third baseman in WARP) being voted in over down-ballot MVP candidate Todd Frazier. Of the WARP leaders, only Kyle Seager, who lost out to second place third baseman Josh Donaldson, didn’t make the team at all.
But OK, how big a deal should we make of this? If the fans’ choices had been replaced by the top players by WARP, then the National League would have gathered seven more wins above replacement, and the American League would have five and a half more. Remember that if All-Star teams were real, they would likely have won 65 or so games out of the 85-to-90 that most teams have already played, so taking away five wins isn’t that big a deal. If you sum together the elected starters WARP for the National League and compared it to the top WARP players, the elected starters are about 78 percent of the top team by WARP. In the American League, it’s 81 percent.
Of course that’s just the starting lineups, and those guys play just a couple of innings. One of the joys of the All-Star Game is that the next day, the box score is really long because both teams bring 2-3 changes of costume. And yes, the best players vs. fan-favorites question is one that can be debated around and around, but in reality, the fans got about 80 percent of what they could have had. Maybe that’s not the end of the world.
Objection No. 2: My favorite player got left off the team because the Padres needed a representative.
Most Common Retort: But if they didn’t have one, the folks in San Diego wouldn’t watch the game.
There comes a point in every All-Star game that you realize that the fate of the game (and home field advantage in the World Series, more on that later) will come down to Tyson Ross facing off against Brandon Moss. These are the Ron Coomer (1999 Twins representative) All-Stars. Are these one-off guys ruining all the fun?
I looked for players on rosters who were named as either reserves or pitchers, and who did not have a teammate who made the All-Star roster. For example, Chase Utley is the only Philadelphia Phillie to make the NL squad, but made it because he was voted in. I'm looking for the guys whom we could reasonably say might be there only because of the one-each rule.
In the National League, there were four such players. One of them is lone Marlins representative Giancarlo Stanton, who leads all position players in baseball in WARP. I think we can give him a pass. Then, there's second baseman and new dad Daniel Murphy of the Mets. Murphy currently ranks as the third-best second sacker on the Senior Circuit, and the two men better than he (Utley and Dee Gordon... raise your hand if you saw that coming) are already in the game, so he's not displacing anyone at second. That leaves us with lone National Jordan Zimmerman and lone Padre Tyson Ross. Both are displacing seemingly better options (why not Stephen Strasburg, if you need a Nat?)
In the American League, two position players qualify. Michael Brantley of the Indians ranks third among all AL outfielders, behind Trout and Gordon. Jose Altuve is the lone Astros' representative, but as we discussed previously, is either the best or second best AL second baseman.
On the pitching side, Jon Lester was the only member of the Red Sox to make it, but ranks second among AL pitchers in WARP. David Price is the only member of the Rays (for now...) to make the AL team. Price isn't way down the list, but Chris Sale and Corey Kluber probably deserve better than having to run against each other in the "Final Vote."
You might quibble with some of the other choices that the players and managers made (Charlie Blackmon? Really?), but the one-player-per-team rule isn't a good scapegoat this year. At most, we're talking about three spots that might have otherwise been filled by more deserving players -- out of 66. And yeah, kids from San Diego will tune in to see Tyson Ross be introduced, and hopefully will become lifelong fans of the game. MLB is counting on their future patronage. That’s why Tyson Ross is on the team.
Objection No. 3:It “counts” for home field advantage in the World Series. Why on earth does it count?
Most Common Retort: Because in 2002, they tied. And also, look at the NFL Pro Bowl which doesn’t count for anything at all and eventually ends up looking like a bunch of guys goofing around while playing flag football.
This past weekend, the Chicago Cubs and Oakland A’s pulled off a true blockbuster trade, with the A’s sending three prospects (and a player to be named later) for Cubs starters Jeff Samardzjia and Jason Hammel. Samardzjia was widely considered one of the best starters available on the trade market, and adding him to the rotation makes the A’s, already with the best record in the American League, that much more of a force to contend with. Why bring this up? The Baseball Prospectus playoff odds report is a handy tool for an honest look at whether your team is a contender or a pretender. It simulates the rest of the season thousands of times and tries to come up with an estimate of how likely each team is to make the playoffs. Before the big deal, the A’s were considered to have a 98 percent chance to advance to the playoffs, and a 13.8 percent chance to win the United States and Toronto Series. As of Monday night, the A’s had moved to a 14.5 percent chance of being crowned champions at year’s end. The big deal (and two wins over the Blue Jays) netted them a total of 0.7 percentage points of winning it all.
The team which wins the All-Star Game next Tuesday will give its league home field advantage in the 2014 World Series. What we know from looking at the past is that in baseball, the home team usually wins 53-54 percent of the time. The game isn’t completely tilted against the visitor, but there really is an edge to playing at home. In a seven-game series where one team gets four home games (if needed) and the other gets three, that’s not a trivial matter. In fact, if we assume that the two teams who will contest the World Series are roughly equal to one another (and at this point, we might as well make that assumption until we find out who’s actually in it), then we would expect the team with home field advantage to win the series 52 percent of the time, just based on having that extra home game.
The inline being referenced is no longer available (deleted from the original page). Please select another inline or remove the component
By winning the All-Star Game, we roughly expect that one team will get a two percentage point advantage bump in the World Series. Now, the A’s, for example, can’t say that they own the entirety of that two point jump if the American League wins on Tuesday. In order to claim those two points, they have to make it to the World Series first. But, at the moment, the A’s are projected to have a little more than a 25 percent chance of being the AL representative in the Fall Classic, so in some sense, they would own about a quarter of that, or half a percentage point of advantage if the AL picks up the win.
So, as far as the A’s are concerned, they would derive about as much benefit to their chances of winning the World Series by having the American League win the All-Star Game than they did by making a gigantic deal. That’s right, the outcome of a meaningless exhibition game counts as much as a mega-deal. I know that Bud Selig just wanted to give the game a little kick when he introduced this rule, but I do wonder if he knows how much of an effect that rule is.
For the next week or so, there will be plenty of debate over who got snubbed and who shouldn’t have made the team. There will be a festschrift for Derek Jeter. Someone will hit a home run and immediately be in the lead for the MVP award. And yeah there’s room to argue about which 68 men should be on the field (isn’t baseball great?) but all told, I’d say that those weird rules that baseball has about selecting some of them are pretty good. Not perfect in terms of getting all of the best players, but it gets most of them and it allows some space for having it be a marketing event for MLB. That’s kinda neat. For all of the schemes that people have proposed for reforming the system, I don’t think it really needs a reform.
But the part about the game “counting” may have consequences well beyond what MLB originally intended. I’m not sure if anyone did the math to figure out just how much that prize of home field advantage can really affect a team’s chances. Obviously, someone has to have home field advantage, but I don’t think that the All-Star Game is a good way to award it. It’s one random game, played by players who are trying harder than they are in other sports, but probably not at 100 percent. That’s not a good representation of anything, and it can have real (and really large) effects on the hopes of some teams.
Then again, in ye olde days, the home field advantage was given on a rotating basis, which was arbitrary as well. Maybe this is just another case where the rule doesn’t make statistical sense, but as a marketing gimmick, it works wonders.
Maybe I need to stop fixing the game and just enjoy it. Next Tuesday. On FOX.