Win Shares and "Valuableness"
As glad as I am that Derek Jeter did not win the MVP, there are still things to be said about the MVP award and overall player "valuableness" (if that's even a word). So to begin...
Offensive Valuableness: Although this idea just hit me yesterday, I'm sure that it already exists as some stat that I've read about before but forgotten. So we've been saying here that Jeter really isn't that valuable to the Yankees (surrounded by 8 other All-Stars & 2 former MVP's, etc..). But there hasn't really been a way to quantify it, until now. (Note the following stats are from ESPN's Stats page provided by Stats, Inc.) So everyone knows that Jeter beat Morneau in RC with 128.2 to 119.7 which does indicate that Jeter provided more offensive production than Morneau. However, what if you took Jeter's RC (i.e. offensive contribution to the Yankees) and found it as a % of the total RC by the Yankee team (952.5) and did the same for Morneau with the Twins (809.7). Jeter accounts for 13.5% of the Yanks RC (theoretical) and Morneau for 14.8% of the Twins RC (theoretical). So Morneau created a larger % of runs (i.e. offensive value)for his team. Doesn't that therefore make him more valuable? Now the obvious problem with this method of calculation is that if you took Travis Hafner and put him on a team with 8 Nefi Perez's, his % of RC would probably be somewhere near 75%, but of course the team would never win. So yes as your team diminishes, a player's value would increase, but past a certain point the argument doesn't hold. It probably isn't good for an absolute comparison between 2 players, but rather a relative comparison between 2 players that are very similar in other statistical areas (this year Jeter, Ortiz, Hafner, Frank Thomas, Mauer, etc..) Now one could make the argument that transplanting someone like Hafner onto the Yankees and instantly his % of RC (and therefore valuableness) would drop. However, I'd like to think that if Hafner batted with AROD/Giambi/6 other All-Stars protecting him, his stats (and RC) would improve since he would see more pitches to hit with a much better lineup and thus accumulate hits rather than walks. Granted, this is only a measurement for offensive production and does not take into account defense or baserunning (or "intangibles"--which basically means being lucky enough to win 4 WS rings in your first 5 years). It seems to me that this would be a logical way to judge how valuable a player was to his team on a purely offensive basis during a year.
Win Shares: Win shares is a statistic used by sabermaticians to argue about player's statistical performance during a season. Although the idea is well-intentioned "evaluating the contribution of individual players to their teams' overall performance..", it fails in that player's who play on teams which underperform relative to their RS and RA are punished (in that their WS totals are lower than they should be). The majority of Win Shares accrued by players is based on offense so this discussion is only dealing with those. There are 4 possible scenarios that can happen in any game due to 2 categories of outcomes summarized in the table below:
When a player's team wins he gets the large or small amount of credit that is due to him based on his offensive performance. When the team loses, the player gets 0 credit towards WS. If the player performed poorly, then he was losing very credit he would have gotten if they won (although if you use a system with negative win shares, I suppose it is possible that in winning a player would receive negative win shares for extremely poor performance which would lower his WS as compared to adding 0 from a loss). Now, if the player performed well but the team loses, he gets no credit, whereas he should have received a large amount. This is the fundamental flaw (as I see it) in WS. Players who perform well in games where they team still loses are penalized in that their WS are artificially low. Case in point: Travis Hafner. Hafner played on a team last year which scored 90 more runs than they allowed last year, but finished 4 games under .500. How does this happen? Terrible pitching or more specifically terrible bullpen pitching. A theoretical example follows: suppose Travis Hafner has an excellent day at the plate (2-5, HR, 3 RBI). The Indians have the lead entering the 8th 4-3. However, the White Sox come back and steal the win 5-4. Thus, Hafner gets no (due) credit for his performance. But if they could have held the lead, then Hafner would have gotten something added onto his WS total. Perhaps you realized that his scenario I just described actually happened. Multiple times. The Indians blew 23 save chances last year, which didn't even lead the AL--thank you KC, but they did have the worse save % in all of baseball--51%. If the Indians were league average and saved 66% of their games, they would have had 7 more wins, which would have meant some more credit for Hafner (the level of which is unknown). The whole point of this is that Hafner's WS total is dependent upon 2 things: his individual performance to produce offense (which he has complete control over--for the most part) and his team's ability to win baseball games (which he has somewhat minimal control over--especially on the pitching side). Thus, when relief pitchers or even starting pitchers blow games and the Indians lose those games even though Travis Hafner played tremendously well, he doesn't get any credit towards helping his team win (even though in some cases he clearly did).
P.S. He beats Jeter in OBP, SLG and provided essentially the same value as Jeter offensively in 5 months that Jeter did over 6 months. Plus, he has his own freakin' candy bar. Not some gay cologne. We love Travis Hafner.
Offensive Valuableness: Although this idea just hit me yesterday, I'm sure that it already exists as some stat that I've read about before but forgotten. So we've been saying here that Jeter really isn't that valuable to the Yankees (surrounded by 8 other All-Stars & 2 former MVP's, etc..). But there hasn't really been a way to quantify it, until now. (Note the following stats are from ESPN's Stats page provided by Stats, Inc.) So everyone knows that Jeter beat Morneau in RC with 128.2 to 119.7 which does indicate that Jeter provided more offensive production than Morneau. However, what if you took Jeter's RC (i.e. offensive contribution to the Yankees) and found it as a % of the total RC by the Yankee team (952.5) and did the same for Morneau with the Twins (809.7). Jeter accounts for 13.5% of the Yanks RC (theoretical) and Morneau for 14.8% of the Twins RC (theoretical). So Morneau created a larger % of runs (i.e. offensive value)for his team. Doesn't that therefore make him more valuable? Now the obvious problem with this method of calculation is that if you took Travis Hafner and put him on a team with 8 Nefi Perez's, his % of RC would probably be somewhere near 75%, but of course the team would never win. So yes as your team diminishes, a player's value would increase, but past a certain point the argument doesn't hold. It probably isn't good for an absolute comparison between 2 players, but rather a relative comparison between 2 players that are very similar in other statistical areas (this year Jeter, Ortiz, Hafner, Frank Thomas, Mauer, etc..) Now one could make the argument that transplanting someone like Hafner onto the Yankees and instantly his % of RC (and therefore valuableness) would drop. However, I'd like to think that if Hafner batted with AROD/Giambi/6 other All-Stars protecting him, his stats (and RC) would improve since he would see more pitches to hit with a much better lineup and thus accumulate hits rather than walks. Granted, this is only a measurement for offensive production and does not take into account defense or baserunning (or "intangibles"--which basically means being lucky enough to win 4 WS rings in your first 5 years). It seems to me that this would be a logical way to judge how valuable a player was to his team on a purely offensive basis during a year.
Win Shares: Win shares is a statistic used by sabermaticians to argue about player's statistical performance during a season. Although the idea is well-intentioned "evaluating the contribution of individual players to their teams' overall performance..", it fails in that player's who play on teams which underperform relative to their RS and RA are punished (in that their WS totals are lower than they should be). The majority of Win Shares accrued by players is based on offense so this discussion is only dealing with those. There are 4 possible scenarios that can happen in any game due to 2 categories of outcomes summarized in the table below:
| Results on Offensive Win Shares | Team Wins | Team Loses |
| Player performs well | Players gets large credit towards WS | Players gets no credit (deserved) towards WS |
| Players performs poorly | Players gets little credit towards WS | Players gets no credit (undeserved) towards WS |
When a player's team wins he gets the large or small amount of credit that is due to him based on his offensive performance. When the team loses, the player gets 0 credit towards WS. If the player performed poorly, then he was losing very credit he would have gotten if they won (although if you use a system with negative win shares, I suppose it is possible that in winning a player would receive negative win shares for extremely poor performance which would lower his WS as compared to adding 0 from a loss). Now, if the player performed well but the team loses, he gets no credit, whereas he should have received a large amount. This is the fundamental flaw (as I see it) in WS. Players who perform well in games where they team still loses are penalized in that their WS are artificially low. Case in point: Travis Hafner. Hafner played on a team last year which scored 90 more runs than they allowed last year, but finished 4 games under .500. How does this happen? Terrible pitching or more specifically terrible bullpen pitching. A theoretical example follows: suppose Travis Hafner has an excellent day at the plate (2-5, HR, 3 RBI). The Indians have the lead entering the 8th 4-3. However, the White Sox come back and steal the win 5-4. Thus, Hafner gets no (due) credit for his performance. But if they could have held the lead, then Hafner would have gotten something added onto his WS total. Perhaps you realized that his scenario I just described actually happened. Multiple times. The Indians blew 23 save chances last year, which didn't even lead the AL--thank you KC, but they did have the worse save % in all of baseball--51%. If the Indians were league average and saved 66% of their games, they would have had 7 more wins, which would have meant some more credit for Hafner (the level of which is unknown). The whole point of this is that Hafner's WS total is dependent upon 2 things: his individual performance to produce offense (which he has complete control over--for the most part) and his team's ability to win baseball games (which he has somewhat minimal control over--especially on the pitching side). Thus, when relief pitchers or even starting pitchers blow games and the Indians lose those games even though Travis Hafner played tremendously well, he doesn't get any credit towards helping his team win (even though in some cases he clearly did).
P.S. He beats Jeter in OBP, SLG and provided essentially the same value as Jeter offensively in 5 months that Jeter did over 6 months. Plus, he has his own freakin' candy bar. Not some gay cologne. We love Travis Hafner.
posted by DWTHTB at 7:18 PM
0 Comments:
Post a Comment
<< Home