Which All-Star Roster Is More Talented?

Every year, some combination of fans, players, and members of Major League Baseball come together to select All-Star teams to represent the leagues in an exhibition. Every year, that group mostly gets it right. And every year, either players miss the game due to injury or pitchers pull out after having started on Sunday, and we still end up with a game featuring baseball’s very best players.

This season, both sides are missing arguably their best player: Mike Trout is aiming to return from injury as the second half begins, while Clayton Kershaw is resting after having started a couple days ago against Kansas City. With each side suffering a fairly major blow, let’s see which team is the most talented.

One indicator of talent is the actual quality of play on the field in the first half. One way to measure that play is using WAR. The graph below shows all the position players in both leagues by WAR accumulated so far this season.

As you can see, the American League has the edge at the top, even without Mike Trout. Aaron Judge, Jose Altuve, and Mookie Betts lead the way, and the AL has five of the top seven slots, with only Paul Goldschmidt and Justin Turner from the NL in that mix. Neither Goldschmidt nor Turner are in the starting lineup, however. After the top seven, the National League has 13 of the next 17 spots and overall, which is enough to narrowly edge the AL in first-half WAR by a count of 53.6 to 52.8. A healthy Mike Trout would make things even closer to equal, but it’s already basically a wash.

Often times, the All-Star game will reward players with somewhat surprising or out-of-character first halves. This isn’t a bad thing. Fans are exposed to potentially new and exciting players and players are rewarded for getting off to a good start. That doesn’t necessarily represent the talent level of the All-Star Game, of course: we know some players’ are bound to experience some regression in the second half. With that in mind, here are the same players, but with their second-half projections, incorporating both a player’s career before this season as well as his first half.

The AL still takes five of the first seven spots by this method, but Bryce Harper possesses the top spot overall. Again, we see, more NL players in the middle and (again) we see a slight edge for the NL overall.

As noted, some players are missing from these calculations. Let’s start with injuries. Only two position players are sitting out the All-Star game with injuries, the aforementioned Trout as well as Starlin Castro of the New York Yankees.

Losing Mike Trout is obviously a blow, forcing the AL to utilize Justin Upton instead. That said, replacing Starlin Castro with Robinson Cano actually makes the team better. So that’s roughly a wash, too.

In addition to injuries, there are also many talented players who simply didn’t make the team. Some due to injuries, some due to a position crunch, and some due to a slow start that might have affected voting what happened in advance of the All-Star Game. Here are the most talented position players who didn’t make the All-Star Game based on second-half projections.

Looks like the major snubs occurred at third base and first base. While Nolan Arenado and Justin Turner are strong representative for the National League at third base, there are certainly good arguments for Bryant and Rendon. The NL’s real problem was taking Daniel Murphy, Josh Harrison, and DJ LeMahieu at second base — particularly LeMahieu, as Harrison is the only Pirates representative and has had a good season. Even taking one of Bryant or Rendon still leaves someone snubbed. Freeman was injured and Rizzo started off slowly, allowing Zimmerman to get a spot, while nobody should argue with Goldschmidt or Joey Votto. In the AL, Donaldson and Beltre have been injured while Machado isn’t having his best season, so no real issues there.

As for the pitching, we have a very close matchup despite the NL’s loss of Kershaw.

Because each club might feature a different quantity of starters and relievers, gaps in WAR to appear. This year, though, both teams feature six starters and six relievers, so that really isn’t at issue here. Chris Sale is beating the world, and his difference over Max Scherzer is essentially the difference overall.

It would appear as though the AL has the bullpen edge. Here’s how the projections look.

By the projections, the NL has the starting-pitching edge, while the AL has the reliever edge. Much like hitting, we are talking about two evenly matched rosters.

As for the players missing, we have a few more on the pitching side than the hitting side due to injuries and the Sunday starter rule. Here are the players missing from both sides due to those issues.

Losing Kershaw is big, but it looks like the AL was hit pretty hard here, too. These injuries are actually what helped give the AL an edge in the bullpen, as Chris Devenski and Roberto Osuna were among the additions. The team did lose Kluber, but Chris Archer is a very good replacement. Replacing Alex Wood with Kershaw hurts the NL’s projections, but doesn’t actually hurt in 2017 WAR, as he is having a ridiculous season thus far.

As for snubs based on talent level, there isn’t the same level of talent missing on the pitching side when compared to the hitters. Here are the pitchers with the highest projected WAR going forward.

Jacob deGrom has a decent case for inclusion, though his ERA might be hurting him a little. Carlos Carrasco is a deserving All-Star and probably should have been named instead of one of the relievers the AL added. Jose Quintana has come on of late and looks — despite all the trade rumors– to be quietly on his way to another four-win season. Verlander and Lester are two veterans who are having solid seasons, but not standing out this year, and Lester wouldn’t have been eligible anyway with his disastrous start on Sunday.

Tonight’s exhibition should be a fun one, and as the table below shows, it is incredibly evenly matched, both in performance so far and expected performance the rest of the season.