1/4 Mark NL and AL MVPs

[RedEye]

Outstanding!

That tells me RBI's are one stat that needs to be considered with all others, not a stat that shouldn't be trusted.

How so? What he posted also had information on how many RBI's the players got per out. To me, that says that I shouldn't trust RBI's unless I know more information how those RBI's were earned. Just saying "Brandon Phillips leads the NL in RBI's" does tell me that he is doing something right, sure, but it doesn't give enough contextual information to say whether he should be a legitimate award contender.

[LaFlamaBlanca]

It's a big deal when you're talking about comparing individual players' stats. Not from the team perspective and from the won games perspective.

Okay, but this is after all a Most Valuable Player discussion. Not sure how you can differentiate a player's individual production, which contributes to the team's production.

[_Sir_Charles_]

Okay, but this is after all a Most Valuable Player discussion. Not sure how you can differentiate a player's individual production, which contributes to the team's production.

I was only commenting on that last sentence of Homer's post. Nothing more, nothing less. Which is why I was trying to explain myself and that I was talking in terms of the team and not the individuals. But as usual, I did a piss-poor job of that. :O) Big surprise, right? *grin*

[757690]

How so? What he posted also had information on how many RBI's the players got per out. To me, that says that I shouldn't trust RBI's unless I know more information how those RBI's were earned. Just saying "Brandon Phillips leads the NL in RBI's" does tell me that he is doing something right, sure, but it doesn't give enough contextual information to say whether he should be a legitimate award contender.

But that's true of every stat. No stat tells us the whole story. That doesn't mean no stat should be trusted. It just means no stat alone should be used to evaluate a player.

[757690]

They absolutely tell us who the more productive player is. A run created in Colorado is not equal to a run created in Seattle.

I agree that a run created in Colorado is not equal to a run created in Seattle, but I still don't think that park adjusted stats tells us who was more productive.

As I said earlier, they are estimates based on broad generalities. They are hypotheticals, so they really don't tell us who actually was the most productive. They do not represent what actually would happen if Trout played for the Tigers and Cabrera played for the Angels. Players adjust their approach all time, based on a myriad of factors, the ball park being a big one. So we really don't know how productive each player would be if they played on each other's team, or if they both played on the same team.

And why are just adjusting for ballpark effects? A run in a game where Cliff Lee is the opposing pitcher is not worth the same as run in a game where Mike Leake is the opposing pitcher. And a run in a game with the wind blowing in is not the same as a run with the wind blowing out.

Players adjust to who the opposing pitcher is, who the defense is, the weather, the time of day, shadows on the field, the umpires strike zone, etc. Why don't we adjust players stats based on those factors?

The point is that luck plays a part in production all the time. It just seems silly to me to single out park effects as the one luck factor to adjust for.

[dougdirt]

The point is that luck plays a part in production all the time. It just seems silly to me to single out park effects as the one luck factor to adjust for.

Because over the course of a full season, most guys are going to face a similar number of "Cliff Lee" types and "Mike Leake" types. But over the course of a full season, not everyone is going to play in a place like the AL West with three huge pitcher friendly parks compared to a place like the NL Central with several very hitter friendly parks.

[Raisor]

You stole it from FC Lane though

Yeah, but I'm ok with that.

[LaFlamaBlanca]

I was only commenting on that last sentence of Homer's post. Nothing more, nothing less. Which is why I was trying to explain myself and that I was talking in terms of the team and not the individuals. But as usual, I did a piss-poor job of that. :O) Big surprise, right? *grin*

Fair enough brother. All of here are obviously smarter than the average bear just by rooting for the right baseball team.

[Raisor]

Reason why I trust RBI's.

The Runs Created stat represent theoretical runs that we are making happy guesses about probably scoring in general over the course of a hypothetical season.

.

The thing is, we can test runs created against actual runs and see how accurate the stat is.

I use the ESPN version of the RC forumla

Runs created
[(H + BB + HBP - CS - GIDP) times (Total bases + .26[BB - IBB + HBP] + .52[SH + SF + SB])] divided by (AB + BB + HBP + SH+ SF)


Going into today's game, the Reds have "created" 188.1 runs. The have scored 195 real runs.

That's 96.4% accuracy.

So, to date, we can tell with 96.4% accuracy how many runs each player has been responsible for this season:

Choo 41.9 RC in 191 TPA. Adjusted to 650 TPA = 142.59
Votto 34.8 RC in 193 TPA. Adjusted to 650 TPA = 117.2
Phillips 23.0 RC in 176 TPA. Adjusted to 650 TPA = 84.94
Bruce 21.2 RC in 179 TPA. Adjusted to 650 TPA = 76.98

I adjusted to 650 TPA to represent a full season of Plate Appearances so we can look at apples to apples.

Choo is on pace to create 57.65 more runs than Phillips.

That's HUGE.

I've been a fan of RC for years, and really don't think it's used enough.

[757690]

The thing is, we can test runs created against actual runs and see how accurate the stat is.

I use the ESPN version of the RC forumla

Runs created
[(H + BB + HBP - CS - GIDP) times (Total bases + .26[BB - IBB + HBP] + .52[SH + SF + SB])] divided by (AB + BB + HBP + SH+ SF)


Going into today's game, the Reds have "created" 188.1 runs. The have scored 195 real runs.

That's 96.4% accuracy.

So, to date, we can tell with 96.4% accuracy how many runs each player has been responsible for this season:

Choo 41.9 RC in 191 TPA. Adjusted to 650 TPA = 142.59
Votto 34.8 RC in 193 TPA. Adjusted to 650 TPA = 117.2
Phillips 23.0 RC in 176 TPA. Adjusted to 650 TPA = 84.94
Bruce 21.2 RC in 179 TPA. Adjusted to 650 TPA = 76.98

I adjusted to 650 TPA to represent a full season of Plate Appearances so we can look at apples to apples.

Choo is on pace to create 57.65 more runs than Phillips.

That's HUGE.

I've been a fan of RC for years, and really don't think it's used enough.

I love Runs Created. I use it all the time.

But we really don't know if it's 96% accurate when applied to individual players. It's works great for teams, because it includes all the data of every game. For an individual player, not so much. It might be more accurate for individual players, we simply don't know.

For the record, I think Choo has been more valuable that Phillips, and I don't really need RC to tell me.

Hey, guess what's 100% accurate at telling us how many runs a player helped his team score? I'll give you three guesses and the first two don't count

[Raisor]

I love Runs Created. I use it all the time.

But we really don't know if it's 96% accurate when applied to individual players. It's works great for teams, because it includes all the data of every game. For an individual player, not so much. It might be more accurate for individual players, we simply don't know.

I've never understood this line of thinking. All a team's RC total is are the player numbers added together.

If a double is worth x for the team it's worth x for the player

[757690]

I've never understood this line of thinking. All a team's RC total is are the player numbers added together.

If a double is worth x for the team it's worth x for the player

And yet, when we calculate each individual's RC for a team, then add them together, we never get the same the number as when we figure out a team's RC.

[Raisor]

And yet, when we calculate each individual's RC for a team, then add them together, we never get the same the number as when we figure out a team's RC.

It's because of rounding. They only show the first digit past the decimal point. If you use the raw numbers they do match up

[757690]

It's because of rounding. They only show the first digit past the decimal point. If you use the raw numbers they do match up

Nope. Try it. Calculate each players RC on your own, go to whatever decimal you want. Add them up, that number won't be the same as the one you get when you calculate a team's RC. I've done it numerous time. Yes, my life is that pathetic.

[Superdude]

Nope. Try it. Calculate each players RC on your own, go to whatever decimal you want. Add them up, that number won't be the same as the one you get when you calculate a team's RC. I've done it numerous time. Yes, my life is that pathetic.

That's one of the big knocks against RC if I remember correct. I haven't dug into SABR stuff in years, but wasn't there some thought that RC overrated great hitters because the formula essentially places them in a context of themselves instead of an average lineup? I'm guessing that's why the two numbers don't align.