Bryce Harper is hitting .338/.467/.667. You know who never had a season like that? Ken Griffey Jr., who never reached any of those figures in a single season, let alone at age 22. Alex Rodriguez hit .358 one year but has never come close to a .467 OBP and has never slugged .667. Miguel Cabrera, in his best season, had a .442 OBP and .636 slugging mark.
Harper's wRC+ -- that's Weighed Runs Created Plus, which controls for park effects and run environment -- is 203, the 13th-highest single-season mark since World War II ended. And yet, Harper doesn't appear to be a lock for the MVP award.
The biggest argument against Harper is that his Washington Nationals aren't going to make the playoffs. That has led to some of the Yoenis Cespedes-for-MVP silliness. Cespedes has had a terrific season, no doubt; if you include his time with the Detroit Tigers, he's fifth among position players in FanGraphs WAR, behind Harper, Josh Donaldson, Mike Trout and Joey Votto. He's ninth in Baseball-Reference WAR. If you want to make the argument his American League stats should count as part of his case for National League MVP ... well, I don't agree with you, but maybe it's not fair to penalize him simply because he was traded. Even with that, he's well behind Harper in WAR, 9.6 to 6.4 on Baseball-Reference.
A bit down both lists is Anthony Rizzo of the Chicago Cubs. He's 14th on FanGraphs (eighth among NL players) and 12th on Baseball-Reference (sixth among NL players). That would appear to make him a fringe MVP candidate, albeit one boosted by the fact the Cubs will make the playoffs, while the only playoff guy ahead of him on either list is Andrew McCutchen.
Dave Studeman of The Hardball Times asks, "What is a most valuable player?" He examines the five leading NL MVP candidates -- Harper, Paul Goldschmidt, Votto, McCutchen and Rizzo. After presenting some of the more basic advanced metrics like weighted Runs Created, he looks at something called RE24, which evaluates the run expectancy generated from each plate appearance.
The other four guys listed above gain value here while Harper (through Monday, at least) loses value. Why? With runners in scoring position, for example, Harper has hit .305/.479/.562. Still excellent, but not as good as his overall numbers. To dig deeper, with runners on first and second he's hit .167 in 42 at-bats.
Rizzo, meanwhile, has hit .276/.386/.518 overall. But with runners in scoring position he's hitting .308/.429/.600. With runners on first and second he's hitting .333 and slugging .667. With a runner on third and two outs he's 8-for-19 (.421) with seven walks. Those numbers increase his run expectancy. As Studeman writes:
This is called situational hitting. People tend to ignore it because it’s not very likely Rizzo will repeat this breakout again. But MVP awards aren't given for repeatable performances. They're given based on what actually happened. Rizzo’s RISP performance actually happened.
Studeman didn't mention what Baseball-Reference codes as high-leverage situations, when games are closest. Rizzo is hitting .393/.508/.609 in high-leverage situations (118 PAs) compared to Harper's .244/.411/.442 (112 PAs).
Studeman's study then goes to a next level and looks at the final margin of the game. I'm not sure how fair this is: If a player hit a home run in the top of the first inning, he doesn't know if the final score is going to be 1-0 or 10-0. Studeman's formula thus gives more credit for runs created in a close game. This is similar to -- but not the same as -- Win Probability Added, which looks at the value of each plate appearance based on the inning and score. So a home run in the first inning of a 0-0 game has the same value regardless of the final score.
Anyway, in Win Probability Added, Rizzo leads NL hitters:
1. Rizzo, 6.4
2. Votto, 6.1
3. Harper, 5.4
4. Goldschmidt, 5.4
5. Kris Bryant, 5.3
In Studeman's final formula, he arrives at this game-adjusted runs leaderboard:
1. Votto, 60.6 runs
2. Harper, 56.4
3. Goldschmidt, 55.5
4. Rizzo, 54.0
5. McCutchen, 43.4
It makes it much more even across the board, at least among the top four candidates, although that's not factoring in position and defense; Harper has the advantage there as a right fielder among three first basemen.
And all of that ignores which team each player did damage against. Harper has played 43 games against the pathetic pitching staffs of the Miami Marlins, Atlanta Braves and Philadelphia Phillies, hitting .357 with 19 home runs and 41 RBIs. Rizzo and Goldschmidt have each played just 19 games against those three teams.
One more consideration, if you want to go another layer deeper: Harper has hit .250/.357/.450 against the New York Mets with three home runs and six RBIs. Against the St. Louis Cardinals, Rizzo has hit .297/.391/.582; against the Pittsburgh Pirates, he's hit .404/.484/.596. Against both teams, Rizzo has six home runs, 21 RBIs, 23 runs and 15 walks in 30 games. He's played his best against the two teams the Cubs had to contend with for the division title. I would argue those numbers certainly shouldn't be ignored.
Harper has certainly had a season for the ages -- statistically the best offensive year we've seen since Barry Bonds. Is he the most valuable player?
I'm still inclined to vote yes, but there's a pretty good case to be made for Rizzo.
Especially if you demand your MVP come from a playoff team.