|
|
Post by noetsi on Feb 5, 2020 17:43:31 GMT -5
I thought about this in the discussion of the HOF. It would be a new measure to determine who was the better player, controlling for change over time (aka standardizing).
1) You chose the 10 most important measure of offense and defense (aka slugging, fielding percentage etc). You assign a weight so that all the weights add up to 1.0 2) You take the rank of each player on the given measure (so if you are first in that year in a given measure you are a 1, if you are second best a 2 etc). Then you average this over their career and multiply by the weight. Then you sum the ten measures.
I decided to name it after the one person who might be interested in it on the board. Probably someone already invented this  |
|
|
|
Post by kaybli on Feb 5, 2020 17:51:13 GMT -5
I thought about this in the discussion of the HOF. It would be a new measure to determine who was the better player, controlling for change over time (aka standardizing).
1) You chose the 10 most important measure of offense and defense (aka slugging, fielding percentage etc). You assign a weight so that all the weights add up to 1.0 2) You take the rank of each player on the given measure (so if you are first in that year in a given measure you are a 1, if you are second best a 2 etc). Then you average this over their career and multiply by the weight. Then you sum the ten measures.
I decided to name it after the one person who might be interested in it on the board. Probably someone already invented this There are many Hall of Fame metrics. Check it out:
|
|
|
|
Post by noetsi on Feb 5, 2020 18:39:46 GMT -5
I was really hoping someone could comment on mine |
|
|
|
Post by rizzuto on Feb 5, 2020 19:06:45 GMT -5
Batting Average
Total Bases
Base on Balls
Other On Base (Intentional Walk, Hit By Pitch, Base on Error)
Base Advance (Steal, Balk, Wild Pitch, Error, Passed Ball)
Move Runners (Sacrifice Fly, Sacrifice Bunt, Ground Out)
Batting Average Relative To League Average
OPS Relative To League Average
|
|
|
|
Post by chiyankee on Feb 5, 2020 19:57:49 GMT -5
I was really hoping someone could comment on mine Are you going to pick the 10 most important offensive and defensive categories? |
|
|
|
Post by inger on Feb 5, 2020 20:01:20 GMT -5
I was really hoping someone could comment on mine Na-Na-Na-Boo-Boo... |
|
|
|
Post by inger on Feb 5, 2020 20:48:08 GMT -5
The problem is that all metrics and formulas fail one test or another. Most lean toward the offensive numbers, and for good reason. Defensive stats are either non-reliable (today), or non-existent (years ago) and evening border on meaningless (fielding pct).
I don’t know if any way to compute the value of fielding vs. the value of batting vs. the value of base running vs. the immeasurable value of having baseball instincts, or being a good or bad team mate. (The fingers of an Inger?)...
Seems folks always wind up falling back on OPS + in these discussions, and indeed it is a handy measurement, as it keeps everything measured vs. peers. But... it ignores all aspects that are non-offense.
If we are to proceed with some kind of new stat, there would have to be some way of determining the pct. of value assigned to each aspect of the game, and thus creating a new stat, or a set of stats that can be totaled. Hmm. Seems very much like WAR, and more than a bit like JAWS.
The hours of calculation and thought that seem to have been required to create those stats are not something I see myself doing, especially with a system that I can see a flaw in at this time.
I appreciate the idea and am honored to have been fingered for the stat name, but I don’t know where to go with the idea at this point. The only gratification I could feel would come from being able to prove or disprove the qualification of existing or future players in an organization that I don’t really believe in. You might as well ask me if the players belong somewhere else I don’t believe in, like heaven...
|
|
|
|
Post by noetsi on Feb 6, 2020 17:22:56 GMT -5
I was really hoping someone could comment on mine Are you going to pick the 10 most important offensive and defensive categories? I was hoping more learned here would. |
|
|
|
Post by noetsi on Feb 6, 2020 17:24:29 GMT -5
The problem is that all metrics and formulas fail one test or another. Most lean toward the offensive numbers, and for good reason. Defensive stats are either non-reliable (today), or non-existent (years ago) and evening border on meaningless (fielding pct). I don’t know if any way to compute the value of fielding vs. the value of batting vs. the value of base running vs. the immeasurable value of having baseball instincts, or being a good or bad team mate. (The fingers of an Inger?)... Seems folks always wind up falling back on OPS + in these discussions, and indeed it is a handy measurement, as it keeps everything measured vs. peers. But... it ignores all aspects that are non-offense. If we are to proceed with some kind of new stat, there would have to be some way of determining the pct. of value assigned to each aspect of the game, and thus creating a new stat, or a set of stats that can be totaled. Hmm. Seems very much like WAR, and more than a bit like JAWS. The hours of calculation and thought that seem to have been required to create those stats are not something I see myself doing, especially with a system that I can see a flaw in at this time. I appreciate the idea and am honored to have been fingered for the stat name, but I don’t know where to go with the idea at this point. The only gratification I could feel would come from being able to prove or disprove the qualification of existing or future players in an organization that I don’t really believe in. You might as well ask me if the players belong somewhere else I don’t believe in, like heaven... The point of using ranks, which most stats don't apparently is that you can control for say hitting being lower in one area than another. The Roy White conversation the other day, which I think Inger started, rasied this point. |
|
|
|
Post by inger on Feb 6, 2020 18:37:30 GMT -5
The problem is that all metrics and formulas fail one test or another. Most lean toward the offensive numbers, and for good reason. Defensive stats are either non-reliable (today), or non-existent (years ago) and evening border on meaningless (fielding pct). I don’t know if any way to compute the value of fielding vs. the value of batting vs. the value of base running vs. the immeasurable value of having baseball instincts, or being a good or bad team mate. (The fingers of an Inger?)... Seems folks always wind up falling back on OPS + in these discussions, and indeed it is a handy measurement, as it keeps everything measured vs. peers. But... it ignores all aspects that are non-offense. If we are to proceed with some kind of new stat, there would have to be some way of determining the pct. of value assigned to each aspect of the game, and thus creating a new stat, or a set of stats that can be totaled. Hmm. Seems very much like WAR, and more than a bit like JAWS. The hours of calculation and thought that seem to have been required to create those stats are not something I see myself doing, especially with a system that I can see a flaw in at this time. I appreciate the idea and am honored to have been fingered for the stat name, but I don’t know where to go with the idea at this point. The only gratification I could feel would come from being able to prove or disprove the qualification of existing or future players in an organization that I don’t really believe in. You might as well ask me if the players belong somewhere else I don’t believe in, like heaven... The point of using ranks, which most stats don't apparently is that you can control for say hitting being lower in one area than another. The Roy White conversation the other day, which I think Inger started, rasied this point. I used “ranks” when I did my Yankee catcher rankings. I felt it to be an appropriate shortcut in that situation because there was a small body of players. If there had been thousands of catchers, the rankings could have been less appropriate. For example, what if a particular player had very few triples, and was ranked something like 567th in triples? His rankings would take a substantial hit. Now, what if that player played for the Yankees in a triples-suppressing ball park?...All stats are not alike nor are all circumstances... |
|
|
|
Post by inger on Feb 6, 2020 21:41:02 GMT -5
I’ve always been of the mind set that all team sports are decided by an equal application of offense and defense. You must score to win, and your opponent cannot win if they cannot score.
To me, the old axiom that pitching is 75% of the game never rang true. I’d have to say that offense is 50% and defense is 50%. So what percentage of defense is fielding? And how much has that changed over the years? It’s interesting to at least ponder...
|
|
|
|
Post by rizzuto on Feb 7, 2020 0:10:20 GMT -5
The point of using ranks, which most stats don't apparently is that you can control for say hitting being lower in one area than another. The Roy White conversation the other day, which I think Inger started, rasied this point. I used “ranks” when I did my Yankee catcher rankings. I felt it to be an appropriate shortcut in that situation because there was a small body of players. If there had been thousands of catchers, the rankings could have been less appropriate. For example, what if a particular player had very few triples, and was ranked something like 567th in triples? His rankings would take a substantial hit. Now, what if that player played for the Yankees in a triples-suppressing ball park?...All stats are not alike nor are all circumstances... That’s why I prefer total bases and league adjustments. |
|
|
|
Post by inger on Feb 7, 2020 1:10:28 GMT -5
I used “ranks” when I did my Yankee catcher rankings. I felt it to be an appropriate shortcut in that situation because there was a small body of players. If there had been thousands of catchers, the rankings could have been less appropriate. For example, what if a particular player had very few triples, and was ranked something like 567th in triples? His rankings would take a substantial hit. Now, what if that player played for the Yankees in a triples-suppressing ball park?...All stats are not alike nor are all circumstances... That’s why I prefer total bases and league adjustments. Total bases per game played, or per at bat (which is essentially SLG) would be more appropriate, as the number is more achievable the more games one is healthy enough to play in. That leads me also that subject. Players who tend to miss a lot of games expose their team to the use of the dreaded “replacement player” in many circumstances. So should they not be penalized in some way? League adjustments are at the least a good start... I suppose that I may be good enough at mathematics to study how to create the equations, but the number of calculations would require a mammoth spread sheet with lots of data. For some reason I’ve never been good at creating spread sheets. It’s almost like I see them in a mechanical sense and that loses me. I am far from mechanically inclined... |
|
|
|
Post by sierchio on Feb 7, 2020 9:38:10 GMT -5
I don't think there will ever be one perfect stat. To judge a ball player, you need to look at a variety of stats.. How you weigh those stats is personal preference and what makes baseball discussions so great. Person A might value slugging percentage while Person B would rather a high OBP... and I think a successful team has a mix of all different skill sets...
|
|
