Home / Wait A Sec, Why Can’t A Pitcher Win The MVP?

Wait A Sec, Why Can’t A Pitcher Win The MVP?

Please Share...Print this pageTweet about this on TwitterShare on Facebook0Share on Google+0Pin on Pinterest0Share on Tumblr0Share on StumbleUpon0Share on Reddit0Email this to someone

I had never heavily considered why pitchers don't win MVP awards until I tried to spin together some MVP buzz for Detroit Tigers ace Kenny Rogers, just 'cause. Then prolific BC commenter zingzing casually mentioned a counterargument, which was that pitchers do not win MVP awards.

Well, not lately.

It's been 13 years since a pitcher won an MVP award in baseball. Dennis Eckersley racked up a 7-1 record, 51 saves and a 1.91 ERA for the A's that year, and 15 of the 28 voters acknowledged his season with a first place vote.

Since then, here are the highest AL MVP vote receivers for pitchers in each year since '92:

1993: Jack McDowell, White Sox (22-10, 3.37 ERA), 9th place, 0 first place votes
1994: Jimmy Key, Yankees (17-4, 3.27 ERA), 6th place, 0 FPV
1995: Jose Mesa, Indians (3-0, 1.12 ERA, 46 saves), 4th place, 1 FPV
1996: Mariano Rivera, Yankees (8-3, 2.09 ERA, 5 saves), 12th place, 0 FPV
1997: Randy Myers, Orioles (2-3, 1.51 ERA, 45 saves), 4th place, 0 FPV
1998: Roger Clemens, Blue Jays (20-6, 2.65 ERA), 11th place, 0 FPV
1999: Pedro Martinez, Red Sox (23-4, 2.07 ERA), 2nd place, 8 FPV
2000: Pedro Martinez, Red Sox (18-6, 1.74 ERA), 5th place, 0 FPV
2001: Roger Clemens, Yankees (20-3, 3.51 ERA), 8th place, 0 FPV
2002: Barry Zito, A's (23-5, 2.75 ERA), 13th place, 0 FPV
2003: Keith Foulke, A's (9-1, 2.08 ERA, 43 saves), 15th place, 0 FPV
2004: Johan Santana, Twins (20-6, 2.61 ERA), 6th place, 0 FPV

2005: Mariano Rivera, Yankees (7-4, 1.38 ERA, 43 saves), 9th place, 0 FPV

There is a clear history of pitchers in the American League being considered for the MVP. In 1999, Martinez actually had one more first place vote than Ivan Rodriguez, who won it. However, not enough overall voters even put him among their list of ten, despite the fact that Martinez struck out 313 batters, which was not only the highest such mark in the AL since Nolan Ryan, but it was more K's than Rodriguez's combined hits and RBI in that year.

We are currently witnessing an extremely trippy season where, among five or six great candidates, not one man's stats and swagger rise above the rest. As I previously stated, the five most likely batting candidates are Derek Jeter, David Ortiz, Justin Morneau, Jermaine Dye, and Frank Thomas. Out of these five, Ortiz and Dye are statistically wiping the floor with the other standouts, but their teams are fading, if not invisible, from the playoff picture. Last month when Adam Hoff handicapped each league's MVP race, his top five (which is essentially the above five, just swapping Thomas with Vladmir Guerrero) weren't separated by much.

A year such as this could enact a shift on voters' perceptions of the importance of starting pitching. And once the engravers tap the names onto the sides of the hardware, we may find that Justin Verlander will become the first starting pitcher to claim the AL Rookie of the Year since Dave Righetti in '81.

Everyone always preaches that no team can win a championship without a great pitching staff. But two counterpoints consistently hamper arguments for hurlers as MVPs:

  1. Pitchers have their own award, the Cy Young.
  2. Pitchers only throw every fifth day, versus batters who play every day

Regarding the "they have their own award, so they should play in their own sandbox" argument, this is a frame of mind and not a rule. In the last 50 seasons, nine MVPs have been pitchers. But the term Most Valuable Player has been open to interpretation for years. Some say the MVP goes to the best overall player. Some say it's the best player on the best team. And others contend the MVP is the person who, if removed from the team, would have the most disastrous impact on the team's win-loss record.

When it comes to postseason MVPs, the rate of handing the shiny trophy to a pitcher is much higher than the season rate of 9 percent. Taking just the last 12 years, four of the last NLCS MVPs have been pitchers, as have five of the last ALCS MVPs and six of the last World Series MVPs — not counting the 2001 year when two pitchers shared the honor.

In a seven-or-fewer-game series, a pitcher can stand out with a dominant 0.61 ERA and a couple wins, but so does a batter who hits .400 with a few homers. When you turn a seven-game series into a 162-game season, why can't the same standard of dominance apply?

The second argument — the fact that pitchers playing 20 percent of the games, therefore implying less work compared to the team's best hitter — is delightfully curious. Batters play a little over 150 games. An ace starting pitchers takes the mound in just over 30 games. So let's compare Johan Santana to — oh, let's just pick one of the above five out of a hat — Jermaine Dye.

Dye, an outfielder, has played in 140 games. In 571 plate appearances, Dye accumulated 165 hits, 43 of them home runs, for a .318 batting average. He has driven in 118 runs and scored 97 of his own. Awesome year.

Santana, a starting pitcher, has pitched 33 times and won 18 games. However, he has faced 867 batters (almost 300 more PAs than Dye), allowed 179 hits — computing the opposing batting average to .221 — given up 24 home runs and struck out 240.

Johan Santana certainly has fewer game appearances than Jermaine Dye. But who really has more effect on the game?

Suppose that in Santana's starts — which have averaged slightly over 6 2/3 inning — the bullpen does not need to come into the game as early. In 23 of his starts, the 'pen didn't need to start facing the opposition until the 8th inning or later. That certainly does impact the effectiveness of relief pitching, and the ripple effect does permeate through his next start. But I've already drooled over Santana's second-half numbers once this month, so I'll stop with the individual stat talk.

(Note: Santana's numbers do not include the Thursday loss to Boston, so his ERA and opposing BA took a little spike, and his second-half record is a little worse. Just a little, though.)

On some betting websites like Doc's Sports, Santana is not listed as a favorite, although he is part of the field (odds: 24-to-1!) and is specifically mentioned as "entirely possible" to be voted MVP. On this particular list (which is dated September 8), Jeter has the best odds at 3-to-2, Ortiz gets 2-to-1 odds, Dye earns 15-to-2 odds, and Morneau and Guerrero weigh in at 12-to-1. If I were a betting man — which implies I have that kind of extra income on which to place bets — I'd risk a sawbuck on "the field." (Thomas isn't listed here either, which makes a field bet that much more tempting).

I say this while fervently rooting for the Detroit Tigers to win the division over the Twins. With no clear batting favorite for the award, Johan Santana should be this year's American League MVP. Wait — should? No. He clearly is the MVP.

Oh, and an honorable mention for Kenny Rogers, mayhaps.

Powered by

About Suss

  • sal m

    you make the best argument for why pitchers shouldn’t win the mvp when you noted that they only contribute every 5th (or so) day. you reinforce this point when you mention that santana averages less than 7 innings per start.

    so not only does he play every 5th day but he participates in just a hair over 75% of the games in which he plays. this might be great for a pitcher in the current era of baseball, but it still pales in comparison to both the contributions to the team and the burdens assumed by the every day player.

    and while the twins have a great record in the games santana starts, this record (which i think is 27-6) is not even in the top ten winning percentages posted in this category in at least the past 10 years.

    santana for cy young, definitely. for mvp, not.

  • Mark Saleski

    when a pitcher is having a great year, his efforts have ripple effects that extend far beyond the “every 5th day” contribution…including indirect support of the bullpen.

    it’s not that simple.

  • Sal,

    Every fifth day is only a good argument if on that day the pitcher and batter have equal workloads. Which is why I pointed out the comparative plate appearance stat.

    Also keep in mind: Santana is 1/5 of the pitching staff, just as Dye is 1/9 of the lineup. Factor in bullpen and pinch hitters, and Santana still does more than Dye or Ortiz or Jeter.

  • sal m

    those stats don’t have any relevance with regard to the fact that position players effect more games than pitchers. what part of the order an everyday player is of no importance in your above example and does nothing to diminish their accomplishment.

    compare pitchers to pitchers and everyday players to everyday players.

    especially when you consider that an everyday player’s contribution goes well beyond at-bats, as fielding is a huge factor.

    despite the stats provided, there’s no evidence whatsoever that a pitcher like Santana “does more” than Jeter, Ortiz etc.

    and the ripple effect of a good starter is overrated and minimum at best…ask robin roberts and bert blyleven.