In our previous Reward Retrospective article, Erik Manning discussed the last time the NL Cy Young award was as close as it turned out. Today, I wanted to talk about another race in another year that was fairly close, though perhaps not involving the right players.
But first, I wanted to point out something of interest to me. The Baseball-Reference Play Index may be one of the coolest tools on the net if you're a baseball nerd like me. If you're like me and are too lazy/lack the computing power to tinker around merrily with a Retrosheet database, the Play Index is made just for you. One of the cool functionalities of the PI is the ability to make lists of individual player seasons using various criteria of your choice. This allows you to make up your own "clubs," if you will.
For example, everyone knows about the 40-40 club, consisting of the four players who have hit forty home runs and stolen forty bases in one season. But do you know about the other 40-40 club, consisting of players with over 40 doubles and 40 steals in a season. There are 30 such players in the time between 1901 and 2009. I figured this isn't exclusive enough. How about the 50-50 club, comprising of the two players who had over 50 doubles and 50 steals in a season.
The two player seasons were Tris Speaker's 1912 season and Craig Biggio's 1998 season. According to Rally's Historical WAR database, Speaker was worth 11 WAR that season, an impressive season indeed. Biggio's 1998? Worth 6.6 WAR, according to the database. No small feat, but no 11-win one either.
What's the point? Well, even though these seasons are the only ones in the illustrious 50-50 club, the two are not created equal. Which brings us to another illustrious club that consists of only one member, the 2007 season of Jimmy Rollins.
The 2007 NL MVP
First, here is the award voting for the top ten recipients, as it went down that season.
| Voting Results | |||||
|---|---|---|---|---|---|
| Rank | Tm | Vote Pts | 1st Place | Share | |
| 1 | Jimmy Rollins | PHI | 353.0 | 16.0 | 79% |
| 2 | Matt Holliday | COL | 336.0 | 11.0 | 75% |
| 3 | Prince Fielder | MIL | 284.0 | 5.0 | 63% |
| 4 | David Wright | NYM | 182.0 | 0.0 | 41% |
| 5 | Ryan Howard | PHI | 112.0 | 0.0 | 25% |
| 6 | Chipper Jones | ATL | 107.0 | 0.0 | 24% |
| 7 | Jake Peavy | SDP | 97.0 | 0.0 | 22% |
| 8 | Chase Utley | PHI | 89.0 | 0.0 | 20% |
| 9 | Albert Pujols | STL | 50.0 | 0.0 | 11% |
| 10 | Hanley Ramirez | FLA | 49.0 | 0.0 | 11% |
The votes were fairly close that year. Rollins snuck past Matt Holliday by less than 20 points in total. As usual, let's take a look using our typical nerdy value measurement, WAR, to see if this order holds any water. First, we'll take a look at values from Rally's Historical WAR database.
| Rank | Player | Batting | Bsr | Defense | PosAdj | RepAdj | WAR |
|---|---|---|---|---|---|---|---|
| 1 | Jimmy Rollins | 16 | 7 | 11 | 8 | 19 | 6.1 |
| 2 | Matt Holliday | 45 | 2 | 15 | -8 | 18 | 7.2 |
| 3 | Prince Fielder | 44 | -3 | -11 | -10 | 18 | 3.8 |
| 4 | David Wright | 49 | 5 | 3 | 2 | 18 | 7.7 |
| 5 | Ryan Howard | 26 | -3 | -5 | -9 | 17 | 2.6 |
| 6 | Chipper Jones | 56 | 0 | 6 | 1 | 15 | 7.8 |
| 7 | Jake Peavy | -- | -- | -- | -- | -- | 6.2 |
| 8 | Chase Utley | 37 | 3 | 8 | 3 | 15 | 6.6 |
| 9 | Albert Pujols | 51 | -2 | 25 | -10 | 18 | 8.3 |
| 10 | Hanley Ramirez | 41 | 4 | -14 | 7 | 18 | 5.6 |
To be fair, I also included FanGraphs WAR data for your perusal. The baserunning column in the following table only includes non-steal baserunning, from Baseball Prospectus' Equivalent Baserunning Runs (EqBRR).
| Rank | Player | Batting | Bsr | Defense | PosAdj | RepAdj | WAR |
|---|---|---|---|---|---|---|---|
| 1 | Jimmy Rollins | 28.6 | 3.4 | 6.6 | 7.5 | 25.9 | 7.2 |
| 2 | Matt Holliday | 49.9 | 3.4 | 14.2 | -7.4 | 23.8 | 8.4 |
| 3 | Prince Fielder | 49.0 | -1.4 | -8.8 | -12.3 | 22.7 | 5.0 |
| 4 | David Wright | 55.5 | 0.4 | 4.9 | 2.5 | 23.7 | 8.8 |
| 5 | Ryan Howard | 33.5 | -5.0 | 0.2 | -11.1 | 21.6 | 3.9 |
| 6 | Chipper Jones | 52.0 | 0.4 | -1.4 | 1.4 | 20.0 | 7.3 |
| 7 | Jake Peavy | -- | -- | -- | -- | -- | 6.1 |
| 8 | Chase Utley | 44.2 | 1.8 | 15.7 | 1.9 | 20.4 | 8.5 |
| 9 | Albert Pujols | 49.3 | 0.5 | 18.8 | -12.2 | 22.6 | 7.9 |
| 10 | Hanley Ramirez | 49.7 | 0.3 | -19.2 | 7.0 | 23.5 | 6.2 |
Obviously, both these tables contain slight differences between each other, one of which being the difference in replacement adjustment. Each has its own leaders at the top of the list, though the bottom seems to be more or less agreed upon. One thing that we can assuredly tell from both tables is that Rollins made off like a bandit in this MVP race.
In both lists, Rollins ends up 7th in Rally's WAR and 6th in "FanGraphs/EqBRR WAR". ahead of only Hanley Ramirez, Prince Fielder, and Ryan Howard in Rally's estimates and also ahead of Jake Peavy in FanGraph's work. For the sake of ease of comparison, let's eschew Peavy from the argument and compare only position players, since replacement level for pitchers is more widely in debate. Among the nine position players in the list, Rollins ranks last in offensive runs, more than outstripped by everyone other than Howard, who still held a healthy lead in offensive runs.
Rollins was the best baserunner among these players, not surprisingly, and his total defensive contribution when including position was good to lead or be very close to the lead (depending on defensive system) for all nine position players as well, with only Albert Pujols and Chase Utley close to competing. However, these margins were on the order of a combined ten runs above average at most, not nearly enough to compensate the difference between Rollins' 16-28 runs above average and Holliday's 50 runs.
So why did he win?
Unlike many MVP races, the taint of the "RBI leader" is not affecting this one; Rollins only had 94 RBIs that season, less than every competitor on the list other than Ramirez and Peavy. And sure, while Rollins' reputation with the glove was always good, it's doubtful that it could have been so valuable to BBWAA voters; it's more likely that they pretty much nabbed him as the best or among the best defenders in the game that year.
So what made the difference? It clearly could not be the high quality of his bat. As you saw, Rollins had the lowest offensive runs above average total among all the players, and even using a more traditional batting line would agree with that assessment. That season, Rollins batted .296/.344/.531, good for a .875 OPS that was significantly lower than the OPS of each other position player considered. The next lowest OPS was Ramirez's .943.
Ah, but perhaps rather than a rate, a certain set of counting stats was the reason for Rollins' victory. And that's where the illustrious 30-20-30-30 club comes in. Rollins became the only player in the history of the game to hit at least 30 doubles, 20 triples, 30 home runs, and steal 30 bags. He also became the fourth player in the 20-20-20-20 club of 20 of each of those counting stats. The other four seasons were Willie Mays' 1957, Frank Schulte's 1911, and Curtis Granderson's 2007. Of course, the fact that all of these players made this list says nothing about the individual performance of their seasons. Mays' 1957 was worth 57 runs above average according to Rally's database, far and away blowing out Rollins' 16 runs.
I will not get into who deserved to win the award, though certainly Chase Utley, David Wright, Albert Pujols, and Matt Holliday were in the running. However, I can say that Rollins' 2007 season, while great and certainly one of the best of his career, should not get any more credit than it deserves because it placed him in a certain exclusive "club." Ultimately, the job of position players at the plate is not to hit a certain number of doubles, triples, or home runs, and it is not to steal a certain number of bases, but rather to contribute runs. No matter how pretty a 30-20-30-30 club sounds (and even saying it doesn't sound that good), if it does not contribute more runs than a standard 40 HR, 1.000+ OPS season, then it is not more valuable despite its novelty.