Proportional MVP by Win Share:

by Matt Rivers on Nov 19, 2005 12:59 AM EST in Events

I recently put up an entry as a review essentially of the MVP voting and David Ortiz vs. Alex Rodriguez.

One interesting way to apply win shares is look at their proportionate distribution amongst a team.

For example.

David Ortiz earns 31 win shares. The Red Sox as a team, collect 285. Hence, Ortiz was worth 11% of the wins, roughly a 10 game swing. That's pretty hefty.

Here's how the percentages break down (Player and Win Share Percentage) amongst the most productive players on each team in both leagues:

BAL (Mora) 12.6%
BOS (Ramirez) 11.9%
CHA (Konerko) 8.0%
CLE (Hafner) 9.7%
DET (Inge) 8.0%
KC (Brown) 11.9%
LAA (Guererro) 9.5%
MIN (Santana) 9.2%
NYA (Rodriguez) 13.0%
OAK (Ellis) 8.0%
SEA (Sexson) 13.0%
TB (Lugo) 11.9%
TEX (Teixeira) 13.5%
TOR (Wells) 8.6%

Based on this list, Mark Teixeira should be the AL MVP. The voters thought he belonged 7th. Sexson, who is tied for 2nd in this list, was 15th overall in the voting. Obviously in this format, it helps when you're a good player on a bad team. Then again, that makes you valuable. The opposite must be true in real life, seeing how Mariano Rivera beat out Sexson in the voting with a WSP of less than 6%. In 2003, A-rod won the MVP with an WSP of 15%. Last year, Guerrero took it home with 10.5%.

Now, the National League.

ARI (Glaus) 10.0%
ATL (Furcal) 10.0%
CHN (Lee) 15.6%
CIN (Dunn) 12.8%
COL (Helton) 12.9%
FLA (Delgado) 12.4%
HOU (Ensberg) 10.9%
LAD (Kent) 14.1%
MIL (Clark) 9.9%
NYM (Wright) 11.2%
PHI (Abreu) 10.6%
PIT (Bay) 16.9%
SD (Giles) 14.2%
SF (Alou) 8.9%
STL (Pujols) 12.7%
WAS (Wilkerson) 9.5%

What's interesting is the higher overall percentages in the national league. In a league where pitching is supposedly a more important, the hitters in the NL are generally doing more to win games for their clubs than the hitters in the AL. Go figure.

Based on our NL list here, Jason Bay is your MVP. The voters decided he should be 12th. Even Jimmy Rollins (8.7%) placed higher. Pujols, this years MVP, finishes 7th.

Another anomoly in the stats is the appearance of only one pitcher, Minnesota's Santana. I suspect this is mostly because no one else on the team did enough to earn the most win shares rather than his pure performance. Pitchers in general are probably undervalued by win shares, but that's a problem to tackle later.

Then again, its a pitcher that holds the mark for most win shares in a season; Charles "Old Hoss" Radbourn in 1884. Radbourn earned 89 win shares, a 35.3%. Pretty safe to say that in those days, the pitcher was ALWAYS the MVP. His numbers for the 1884 Grays are just ridiculous. I'd say he would've been voted in just for his 73 COMPLETE GAMES that SEASON.

The greatest season by a position player is a 59 by Honus Wagner in 1908 (20.06%). Babe Ruth in 1923 is second at 55 (18.7%); Bonds' 2001 season is third (20.0%). Both Bonds and Ruth won the MVP in their respective seasons.

Once again, Not that I buy into this theory of ranking wholesale, I simply desire to learn more about the voting habits of the writers. We can thereby conclude that they did NOT use this method of analysis.

0 recs | Comment 10 comments

Story-email Email | Print |

Comments

You may want to use

Win Shares Above Bench from THT. A "bench" pitcher will earn far fewer Win Shares then a "bench" position player. Adjusting the baseline will help account for that. That's probably why you didn't find any pitchers in the MVP discussion.

by salb918 on Nov 19, 2005 11:32 AM EST 0 recs

Not really relevant

Is being the best player on a really bad team more valuable than being a slightly less dominant player on a good team. Does anyone really think that Adrian Brown is one of the most valuable players in the league, comparable to Manny? I hate to make the argument, but A-Rod's 13.0 is more impressive to me than Teixeira's 13.5. (It would have been nice if Boston had managed to signTeixeira when they drafted him.)

Can the 8.0 for Konerko really be right? That is a truly abysmal offense without him, easily the worst in the majors. I guess all the value is in the pitchers and the D.

by cdamon on Nov 19, 2005 9:34 PM EST 0 recs

Relevant Relevance

"Does anyone really think that Adrian Brown is one of the most valuable players in the league, comparable to Manny?"

Brown was as valuable to the Royals as Manny was to the Red Sox. That's the argument. These percentages level the playing field in terms of a player's contributions to HIS team. The ratio is the same, but raw numbers are obviously different.

"I hate to make the argument, but A-Rod's 13.0 is more impressive to me than Teixeira's 13.5."

You're not making the argument, you're only stating which is more impressive to you.

"Can the 8.0 for Konerko really be right?"

You're more than welcome to double check the math.

by Matt Rivers on Nov 19, 2005 10:03 PM EST 0 recs

Sorry for the slight

My question about Konerko was more rhetorical than meant to actually question your calculations. (I did double check them and you were of course correct.)

I was just suprised that he was that low. I was used to seeing Lee Sinins above average numbers, and Win Shares are more or less absolute zero based (above not playing anyone). The White Sox, with a consistently mediocre lineup, outside of Konerko, show the extreme of this behavior.

To me, the Konerko 8.0 was one of the more enlightening numbers in the piece. Most of Konerko's WS is batting, whereas the team as a whole has almost half it WS in pitching alone. The WSox almost had the only team leader who was a pitcher -- Buehrle was only 1 WS behind Konerko.

So I found it interesting that Konerko was one of the lowest leaders in WS%. In RCAA (Runs Created Above Average), only Bay had a higher % of the positive contributors for his team, and there not by that much. I expected them to be similarly high in WS%. But while Bay and Konerko both almost signle handedly carried bad offenses, Bay played on a team with mediocre pitching and fielding, while Konerko played on a team with great pitching and fielding. The difference between Bay in Konerko in your report really shows the difference in relative value of offense to the two teams.

by cdamon on Nov 21, 2005 11:21 AM EST up 0 recs

Relevant Relevance

I'll try to make the argument.

There are two teams

Team A has an infield including Pujols, A-Rod, Jeter and an outfield of Manny Ramirez, Barry Bonds and Ichiro, with Ortiz DH'ing for the sake of argument, and Arod therefore only gets 10% of that team's win shares.

Team B had players like this year's Edgar R, Mark Bellhorn, Neifi Perez, Kevin Millar and Lugo, and Lugo gets 15% of the team's win shares.

Team A has a winning record and Team B has a losing record.

Your logic says that Lugo should be the MVP.

I don't think that is necessarily the case. He is his team's MVP, and A-Rod is his team's MVP.

Overall, it comes down to what you think the MVP award is designed to recognise. If it is awarded to the player who adds most value to his team, then you're right. If (as I thin it is) it is designed to be awarded to the player who is the best player in the league then you are not necessarily right, though you may be sometimes.

I would say the MVP award SHOULD go to the player who most people would choose as their first pick in a draft of all MLB players based solely on their performance in the year in question.

Do you think most people would / should choose A-Rod or Texeira? Personally, I would choose the former, but I would be casting an envious eye at the guy who gets to pick Ortiz second!!

by UKSoxfan on Nov 21, 2005 6:41 AM EST 0 recs

By your logic

Derrek Lee was far and away the MVP of the NL. I don't necessarily disagree, but I can think of lots of people who will want to crucify you for saying so.

by salb918 on Nov 21, 2005 9:52 AM EST up 0 recs

whose logic?

The system says Bay was the MVP - which probably tells you all you need to know about the weaknesses of the system (which I know matt is not proposing, I don't think I was clear about this before!)

OTOH - if I were being asked to select the number one draft pick in the NL based on 2005 performance only, I would choose Pujols, then Lee. Given the same choice (and same criteria) for the NL Cy Young, I would choose Roger.

by UKSoxfan on Nov 21, 2005 10:14 AM EST up 0 recs

Clarification

I'm saying that your (UKSoxfan) statement that the MVP ought to go to the number one performer in the league, then Derrek Lee is a clear choice over Albert Pujols.

Lee hit more homeruns (46-41), had a higher average (335-330) and slugging percentage (662-609). Pujols beat him in OBP (430-418).

If you like advanced metrics, Lee had huge edge in WARP1 (12.3-10.7) and VORP (106.0-98.8), and Rate2 (109-102).

So, on the basis of performance, Lee over Pujols.

I'm not arguing that the MVP should be awarded in such a manner. It should be given to the player who can beat the other major contenders in a 2.5 km bicycle race, but the BBWAA will never let that happen, now will it?

by salb918 on Nov 21, 2005 10:56 AM EST up 0 recs

Bay is a better choice than many writers made

By any objective measure, Bay was a better player, and certainly more valuable to his team, than was A Jones. While the MVP clearly should have gone to one of Lee or Pujols, voting Bay as one of the top 3 is wholey justifiable. One of the nice points of looking at WS% is identifying players who get too little press because they play for bad teams.

by cdamon on Nov 21, 2005 11:28 AM EST up 0 recs

Bingo

"If it is awarded to the player who adds most value to his team, then you're right. If (as I thin it is) it is designed to be awarded to the player who is the best player in the league then you are not necessarily right, though you may be sometimes."

This is exactly right.

I'm not trying to make a case for MVP using this matrix. I'm only exploring different ways to define value, or "most valuable". The point is, that this is not the way things are looked at, despite fitting the "definition". I don't think anyone would disagree to say that Rodriguez is a more valuable player than someone like Brown, overall speaking. Then again, that's where PPWS can kill a team like the Royals, or the 2001-2003 Rangers.

I don't think I need another disclaimer. It's all just another angle, depending on which definition you want to use.

by Matt Rivers on Nov 21, 2005 10:39 AM EST up 0 recs