Regarding and Re-evaluating Mauer Yet Again

[Brock Beauchamp]

 

If any other player was 40 points higher than his career average BABIP, many posters here would be posting about how he was going to drop. But with Mauer, when he a career high .383 BABIP in 2013 and falls back to his career average in 2014, the only explanation can be a concussion. That's what I find so baffling.

If you facture in the higher BABIP in 2012 & 2013, the decline is not rapid.

You're acting as if Joe Mauer is equivalent to Danny Santana. Yes, Mauer had a "career high BABIP" in 2013. It was all of .010 higher than his previous career high. His 2012 BABIP was tied for second-highest in his career to that point, not a career high. When a guy regularly has a .340+ BABIP, jumping up to .380 on occasion is neither out of the norm nor is it going to have a huge impact on his overall OPS.

 

Here are Mauer's BABIP numbers by year, starting in 2006:

 

.364

.319

.342

.373

.348

.319 (bilateralness year)

.364

.383

concussion

.342

.309

.301

.344

 

Again, we're talking 150 points of OPS here. Twenty points of BABIP doesn't come close to covering even half that gap, much less all of it. Removing that injury-ridden 2011, Mauer had one season with a BABIP under .340, all the way back in 2007, another injury season where he only played 109 games. If Mauer was healthy through 2013, he was a virtual lock to have a BABIP higher than .340 because that's who he was as a player.

 

On the other hand, let's look at Joe's hard hit %, same time period of 2006 forward:

 

30.6

36.9

32.9

37.6

41.7

20.8 (again, bilateral year)

37.1

37.4

concussion

28.0

29.1

31.3

35.2

 

There's a clear correlation between a high hard hit % leading to a high BABIP but not the inverse. I think you're looking at this backwards. Joe stopped hitting the ball hard which led to a drop in BABIP (and maybe shifting had some influence here as well). But there's absolutely no way to make a case that Joe stopped hitting the ball hard because defenses shifted him or "his luck ran out".

 

Again, Occam's Razor here. We can either blame the majority of Mauer's decline on the concussion - which aligns with pretty much every Mauer stat we can find - or we have to use mental gymnastics to explain how Joe didn't suffer concussion effects but managed to stop hitting the ball hard, stop making contact as frequently, and had less discipline overall because he got old (which usually doesn't affect power/discipline nor does it usually happen in seven months) or because defenses shifted him (which makes no sense whatsoever).

[KGB]

You're acting as if Joe Mauer is equivalent to Danny Santana. Yes, Mauer had a "career high BABIP" in 2013. It was all of .010 higher than his previous career high. His 2012 BABIP was tied for second-highest in his career to that point, not a career high. When a guy regularly has a .340+ BABIP, jumping up to .380 on occasion is neither out of the norm nor is it going to have a huge impact on his overall OPS.

 

Here are Mauer's BABIP numbers by year, starting in 2006:

 

.364

.319

.342

.373

.348

.319 (bilateralness year)

.364

.383

concussion

.342

.309

.301

.344

 

Again, we're talking 150 points of OPS here. Twenty points of BABIP doesn't come close to covering even half that gap, much less all of it. Removing that injury-ridden 2011, Mauer had one season with a BABIP under .340, all the way back in 2007, another injury season where he only played 109 games. If Mauer was healthy through 2013, he was a virtual lock to have a BABIP higher than .340 because that's who he was as a player.

 

On the other hand, let's look at Joe's hard hit %, same time period of 2006 forward:

 

30.6

36.9

32.9

37.6

41.7

20.8 (again, bilateral year)

37.1

37.4

concussion

28.0

29.1

31.3

35.2

 

There's a clear correlation between a high hard hit % leading to a high BABIP but not the inverse. I think you're looking at this backwards. Joe stopped hitting the ball hard which led to a drop in BABIP (and maybe shifting had some influence here as well). But there's absolutely no way to make a case that Joe stopped hitting the ball hard because defenses shifted him or "his luck ran out".

 

Again, Occam's Razor here. We can either blame the majority of Mauer's decline on the concussion - which aligns with pretty much every Mauer stat we can find - or we have to use mental gymnastics to explain how Joe didn't suffer concussion effects but managed to stop hitting the ball hard, stop making contact as frequently, and had less discipline overall because he got old (which usually doesn't affect power/discipline nor does it usually happen in seven months) or because defenses shifted him (which makes no sense whatsoever).

Let's look at the clear correlation of BABIP to Hard hit % for each year

BABIP Hard Hit %

2006 to 2007 - down up

2008 up down

2009 up up

2010 down up

2011 down down Bilateral season

2012 up up

2013 up up

2014 down down

2015 down up

2016 down up

2017 up up

 

I don't see a flip of the coin as clear correlations. Also the drop in his contract rate occurred in 2013 before his concussion. But if you just want to look at his average for the clear correlations, you are correct.

[Brock Beauchamp]

Let's look at the clear correlation of BABIP to Hard hit % for each year

BABIP Hard Hit %

2006 to 2007 - down up

2008 up down

2009 up up

2010 down up

2011 down downBilateral season

2012 up up

2013 up up

2014 down down

2015 down up

2016 down up

2017 up up

 

I don't see a flip of the coin as clear correlations.Also the drop in his contract rate occurred in 2013 before his concussion.But if you just want to look at his average for the clear correlations, you are correct.

This... doesn't make any sense. Three things:

 

1. I said "correlation", not "rigid adherence to identical trends".

 

2. You're using a binary up/down, which is woefully inadequate to analyze things that are measured in decimal points. There's a season where Mauer's BABIP dropped a tiny .008 but his hard hit went up a small-ish 2%. That's really sloppy analysis and I think you know it. The obvious trend here, the one you seem to be intentionally ignoring, is the one where Mauer spent most of his career hitting balls hard 34-38% of the time and dropped a full nine points immediately after the concussion. Only this year, the year he's showing success once again (shocking!), has that number nudged toward his previous stellar seasons.

 

3. There's a strong correlation to hard hit balls and BABIP. That's rather undeniable at this point, isn't it? There are loads of articles around the internet connecting those two things.

[Mike Sixel]

Fifteenth among first baseman in fWAR. I don't care how he compares to his past, but how he compares to other first baseman, and to other options the twins have. He's clearly better than their other options, but he's also clearly not among the best at the position. Teams win by having better players than other teams. It's great he's better this year, but they need more from him even than this, unfortunately to be a better team. Let's hope that happens somehow, because they don't have a better option right now.

[70charger]

 

Fifteenth among first baseman in fWAR. I don't care how he compares to his past, but how he compares to other first baseman, and to other options the twins have. He's clearly better than their other options, but he's also clearly not among the best at the position. Teams win by having better players than other teams. It's great he's better this year, but they need more from him even than this, unfortunately to be a better team. Let's hope that happens somehow, because they don't have a better option right now.

 

You know a great way to have a better first baseman than other teams? Move Sano over. Of course then you've made third base worse by more than you've made first base better. But hey, then we can compare our first baseman to other teams' first basemen and feel warm and fuzzy while we lose more games.

 

Outside of a world where you can just go get the Mike Trouts and Bryce Harpers of the world tomorrow and with no limitations, comparing players position by position across the league seems a little stupid because teams win as teams. You can maximize a position or you can maximize a team. Sometimes those are in direct opposition because there simply aren't enough superstars to go around.

[KGB]

This... doesn't make any sense. Three things:

 

1. I said "correlation", not "rigid adherence to identical trends".

 

2. You're using a binary up/down, which is woefully inadequate to analyze things that are measured in decimal points. There's a season where Mauer's BABIP dropped a tiny .008 but his hard hit went up a small-ish 2%. That's really sloppy analysis and I think you know it. The obvious trend here, the one you seem to be intentionally ignoring, is the one where Mauer spent most of his career hitting balls hard 34-38% of the time and dropped a full nine points immediately after the concussion. Only this year, the year he's showing success once again (shocking!), has that number nudged toward his previous stellar seasons.

 

3. There's a strong correlation to hard hit balls and BABIP. That's rather undeniable at this point, isn't it? There are loads of articles around the internet connecting those two things.

Excluding the Bilateral leg weakness year, the standard deviation of the comparison ranking of the BABIP and Hard hit rate is 1.88. Based on a sample size of 11 year, you have to admit that isn't a close correlation. Maybe there is an overall correlation within MLB, but not in this specific players case.

 

Back to Occam Razor: My reasoning that the drop is not concussion related is based on Mauer saying he had no concussion symptoms prior to 2014 and a clear and predictable drop in his BABIP. To be fair, he did say 2 year later he was suffering from headaches due to daylight, but he also hit better the year after the concussion in the daylight so it does not appear to have effected him in 2014.

 

Your reasoning is based on a overall MLB correlation that is shaking at best when looking at this individual player and a factual error that the concussion was the reason for his drop in contact.

 

Neither of us will ever know for sure how much, if any, the concussion had on his career. But it's far from conclusive that it's the reason for his drop-off.

[spinowner]

 

Neither of us will ever know for sure how much, if any, the concussion had on his career. But it's far from conclusive that it's the reason for his drop-off.

That depends on your definition of "far". Is it 100% certain that the drop-off in his performance was due to the concussion? Of course not. But is it more likely that it was due to the concussion than to something else? Yes. And I'd even say it's far more likely.

[Mike Sixel]

 

You know a great way to have a better first baseman than other teams? Move Sano over. Of course then you've made third base worse by more than you've made first base better. But hey, then we can compare our first baseman to other teams' first basemen and feel warm and fuzzy while we lose more games.

 

Outside of a world where you can just go get the Mike Trouts and Bryce Harpers of the world tomorrow and with no limitations, comparing players position by position across the league seems a little stupid because teams win as teams. You can maximize a position or you can maximize a team. Sometimes those are in direct opposition because there simply aren't enough superstars to go around.

 

I have no idea what this post is arguing. This is a thread to evaluate Mauer's relative worth, isn't it?

[jharaldson]

 

Again, Occam's Razor here. We can either blame the majority of Mauer's decline on the concussion - which aligns with pretty much every Mauer stat we can find - or we have to use mental gymnastics to explain how Joe didn't suffer concussion effects but managed to stop hitting the ball hard, stop making contact as frequently, and had less discipline overall because he got old (which usually doesn't affect power/discipline nor does it usually happen in seven months) or because defenses shifted him (which makes no sense whatsoever).

 

Regarding making contact as frequently, Joe had the majority of that decline prior to the concussion:

 

K%

2010 - 9.1%

2011 - 11.4%

2012 - 13.7%

2013 - 17.5%

Concussion

2014 - 18.5%

2015 - 16.8%

2016 - 16.1%

2017 - 13.4%

 

This increase in K% was masked by his career high BABIP in 2013 and a 3rd best BABIP in 2012.  I don't think anyone is saying the concussion did not play an issue, I think people are arguing that there is more nuance to this.  If I were to assign blame for his decline I would probably use these percentages:

 

Age/Injury - 45% (Pre-concussion contact rate reduction)

Concussion - 45% (Post-concussion ISO/Hard Hit reduction)

Shifts/Strike Zone - 10% (Contributor to lower BABIP in recent years)

[dbminn]

Watching Mauer during the 2014-16 seasons was miserable at times. Imagine how awful it must have been for him. I've noticed he's smiled more in the last month than he did the two previous years. 

 

Since May 1 - .319/.405/.447

2004-13 -        .323/.405/.468

 

The league is much higher scoring now than it was early in his career. Still, he's had a nice run the past few months. I'm glad we're getting one more run of classic Joe.

[BrianTrottier]

 

Excluding the Bilateral leg weakness year, the standard deviation of the comparison ranking of the BABIP and Hard hit rate is 1.88. Based on a sample size of 11 year, you have to admit that isn't a close correlation. Maybe there is an overall correlation within MLB, but not in this specific players case.

 

Two comments:

 

1) Not sure what you're applying a standard deviation to when you say "the standard deviation of the comparison ranking of the BABIP and Hard Hit Rate." The standard deviation of what?

 

2) A standard deviation is not a measure of correlation. It's a measure of spread around a point of central tendency (i.e. an average). All a SD tells us is how dispersed a set of observations is around a mean, not how correlated two variables are with each other. If you want to measure a correlation, then produce a correlation coefficient. Your 1.88 figure doesn't tell us anything.

[KGB]

 

Two comments:

 

1) Not sure what you're applying a standard deviation to when you say "the standard deviation of the comparison ranking of the BABIP and Hard Hit Rate." The standard deviation of what?

 

2) A standard deviation is not a measure of correlation. It's a measure of spread around a point of central tendency (i.e. an average). All a SD tells us is how dispersed a set of observations is around a mean, not how correlated two variables are with each other. If you want to measure a correlation, then produce a correlation coefficient. Your 1.88 figure doesn't tell us anything

I'll let Brock do that math, it's his statement that we have a clear correlation

[KGB]

Two comments:

 

1) Not sure what you're applying a standard deviation to when you say "the standard deviation of the comparison ranking of the BABIP and Hard Hit Rate." The standard deviation of what?

 

2) A standard deviation is not a measure of correlation. It's a measure of spread around a point of central tendency (i.e. an average). All a SD tells us is how dispersed a set of observations is around a mean, not how correlated two variables are with each other. If you want to measure a correlation, then produce a correlation coefficient. Your 1.88 figure doesn't tell us anything.

 

Okay, I'm at computer now, the Correlation Coefficient is .437. Like I said before,

there is no relation for Mauer between his hard hit rate and his BABIP.

[Brock Beauchamp]

Okay, I'm at computer now, the Correlation Coefficient is .437. Like I said before,

there is no relation for Mauer between his hard hit rate and his BABIP.

I have a newborn in the house so I don't have time to do the math on this but a perfect correlation is one. No correlation is zero. Negative one is a reverse correlation.

 

If you did the math correctly, .437 is a relatively strong correlation.

[jharaldson]

 

I have a newborn in the house so I don't have time to do the math on this but a perfect correlation is one. No correlation is zero. Negative one is a reverse correlation.

 

If you did the math correctly, .437 is a relatively strong correlation.

 

My understanding is that a correlation factor of .437 is considered moderate.  I looked into a statistical paper and they had them ranked this way:

 

.00-.19 “very weak”
.20-.39 “weak”
.40-.59 “moderate”
.60-.79 “strong”
.80-1.0 “very strong”

 

http://www.statstutor.ac.uk/resources/uploaded/pearsons.pdf