Sox to Make Hard Push for Granderson

[Kyyle23]

After reading all the comments on De Aza's base running I'm convinced that a good comparison is Lexi's D. Lexi is a great SS, one of the best in the bigs, but some people only see his mistakes and determine he's garbage there, this seems to be the same scenario for De Aza's base running.

 

If Lexi is truly one of the best in the bigs, then the comparison does not hold up at all. De Aza is not one of the best baserunners in the bigs if you just forget the mistakes he made.

[Dick Allen]

I agree with everything you said, but the point is that our ideas of valuation are off-base. ADA made 10ish dumb baserunning mistakes, which is a handful more than he would be expected to make on average. This hurt his value, but his overall contribution remained positive. I understand that those mistakes may have FELT like disasters that outweigh the positives of what he did, but our emotional reaction doesn't affect the actual value. That's why we want to use statistics to quantify this value, so that our biases can't affect it and make us make a mistake.

 

So if you (the collective you, not specifically you) think that a statistic is off, then be all means argue that the statistic is off. But make that argument based on your understanding of the stat and why it won't add up. It's sole purpose is to be there to inform you better than your human brain, which is notoriously flawed for this type of exercise, so that fact that it may have felt like ADA's errors were more costly than his contributions is nothing like evidence against the numbers.

But this is another thing I don't understand. Why are baserunning and offensive metrics never "off", but defensive metrics always can be wrong? What makes offensive numbers foolproof, weighted perfectly, but defense you need a sample size of 3 years.

[Balta1701]

But this is another thing I don't understand. Why are baserunning and offensive metrics never "off", but defensive metrics always can be wrong? What makes offensive numbers foolproof, weighted perfectly, but defense you need a sample size of 3 years.

From a statistical point of view only, the answer would be "scatter" or "variability". What that probably tells me is that guys are very inconsistent about which balls at the edge of their range they're actually able to get to, to the point where you need lots and lots of data to come up with some sort of meaningful statistical statement.

[southsider2k5]

If you're going to call an objective measurement of something garbage, you should tell us why it's garbage and how your intuition and selective human memory are more accurate.

 

Go back and read the rest of the posts.

[2nd_city_saint787]

If Lexi is truly one of the best in the bigs, then the comparison does not hold up at all. De Aza is not one of the best baserunners in the bigs if you just forget the mistakes he made.

 

I was just speaking in "generalities"(quotes for lack of a better word) you can replace the whole "best in the bigs" comment with "good" or at the very least "not garbage"

 

Or let me change the phrasing here.

 

"After reading all the comments on De Aza's base running I'm convinced that a good comparison is Lexi's D. Lexi is a good SS, but some people only see his mistakes and determine he's garbage there, this seems to be the same scenario for De Aza's base running."

 

Better?

 

I'm in the camp of wanting De Aza to stay, or get traded for a good major league piece. I'd love to see the Sox add a better lead off hitter and move ADA down into the 6-7 spot. He's been one of the Sox best hitters with RISP (over and .850 OPS the last 2 years.)

Edited October 14, 2013 by scs787

[iamshack]

I think it's pretty clear that the hangup is about the mental errors with ADA. He's value lies a lot in his speed, which people are hesitant to give him credit for since it's an inherent skill which people view him as having little or nothing to do with. The mental errors are conscious decisions he makes, and so people are going to hold those against him very harshly.

 

But as some have said, you can live with the stupid mistakes because he brings value elsewhere. Whether that value comes from him being an outstanding athlete or the smartest, most intuitive player ever, it doesn't matter. Would you like him to be smarter? Yes, of course. But regardless of how frustrating he may be, you can't base your decision on that

[Eminor3rd]

But this is another thing I don't understand. Why are baserunning and offensive metrics never "off", but defensive metrics always can be wrong? What makes offensive numbers foolproof, weighted perfectly, but defense you need a sample size of 3 years.

 

Because of the nature of the way defensive plays are recorded, which is on a grid and based on where the player begins and where the player ends up. There are just a lot of contextual exceptions (bad hop, screened by runner, covering second for a steal, etc.) that muddy the data up.

 

That said, the knock on UZR-like stats isnt necessarily that they aren't accurate, it's that UZR scores tend not to be predictive short of three seasons of data. Part of this probably has to do with the factors above, but part of it probably also has to do with the fact that players' defensive performances fluctuate more than we are inclined to expect.

[Eminor3rd]

Go back and read the rest of the posts.

 

I don't see anything where you are telling me why the two statistics below provide inaccurate results:

 

http://www.fangraphs.com/library/offense/ubr/

http://www.fangraphs.com/library/offense/wsb/

 

Are you referring to where you talked about the demoralizing effect of a killed rally? I thought wite pointed out that all kinds of outs kill rallies all the time, and that other types of outs occur drastically more frequently. What if Garcia pops one up because he swung at a bad pitch, that's still a mental error, right?

 

And besides, how can you possibly make a claim as to quantifying an effect like that? It's one thing to suggest that it could be a factor, but entirely another to just decide that its effect is immense enough to make ADA the worst runner in White Sox history despite other evidence to the contrary.

[Eminor3rd]

I think it's pretty clear that the hangup is about the mental errors with ADA. He's value lies a lot in his speed, which people are hesitant to give him credit for since it's an inherent skill which people view him as having little or nothing to do with. The mental errors are conscious decisions he makes, and so people are going to hold those against him very harshly.

 

But as some have said, you can live with the stupid mistakes because he brings value elsewhere. Whether that value comes from him being an outstanding athlete or the smartest, most intuitive player ever, it doesn't matter. Would you like him to be smarter? Yes, of course. But regardless of how frustrating he may be, you can't base your decision on that

 

Right -- this is the point. It seems like ADA messed up a bunch, and he did, but when you add everything up, he was still among the top few players on our team. He's nothing like a star, but he's one of the last places we need an upgrade just because there are like 6 other worse players that we start.

[southsider2k5]

I don't see anything where you are telling me why the two statistics below provide inaccurate results:

 

http://www.fangraphs.com/library/offense/ubr/

http://www.fangraphs.com/library/offense/wsb/

 

Are you referring to where you talked about the demoralizing effect of a killed rally? I thought wite pointed out that all kinds of outs kill rallies all the time, and that other types of outs occur drastically more frequently. What if Garcia pops one up because he swung at a bad pitch, that's still a mental error, right?

 

And besides, how can you possibly make a claim as to quantifying an effect like that? It's one thing to suggest that it could be a factor, but entirely another to just decide that its effect is immense enough to make ADA the worst runner in White Sox history despite other evidence to the contrary.

 

Taking human emotion out of the game is just silly. All outs are not created equally. That is a faulty assumption I don't agree with.

[Eminor3rd]

Taking human emotion out of the game is just silly. All outs are not created equally. That is a faulty assumption I don't agree with.

 

No one is is claiming that emotion doesn't matter, but (1) you can't measure it, so it's silly to assign some huge, arbitrary value to it based only on intuition. In this case, it's better to leave it out of analysis entirely and accept something as being an incomplete picture rather than throw in a valuation completely out of nowhere for the sake of completeness at the cost of making the entire thing wrong, and (2) I think you are overstating the difference between a big out on the bases and a big out of any kind. Further, I think you are underestimating the ability for professional players to stay focused despite disappointment. Part of the reason they are in the Majors is their ability to perform at a consistently high despite physical and emotional fatigue.

Edited October 14, 2013 by Eminor3rd

[southsider2k5]

No one is is claiming that emotion doesn't matter, but (1) you can't measure it, so it's silly to assign some huge, arbitrary value to it based only on intuition. In this case, it's better to leave it out of analysis entirely and accept something as being an incomplete picture rather than throw in a valuation completely out of nowhere for the sake of completeness at the cost of making the entire thing wrong, and (2) I think you are overstating the difference between a big out on the bases and a big out of any kind. Further, I think you are underestimating the ability for professional players to stay focused despite disappointment. Part of the reason they are in the Majors is their ability to perform at a consistently high despite physical and emotional fatigue.

 

You can't measure it, so throw it out and pretend it doesn't exist? I can't agree with that either.

[Eminor3rd]

You can't measure it, so throw it out and pretend it doesn't exist? I can't agree with that either.

 

What's the alternative? If we can measure 80% of something and assume that the rest is unpredictable, we have a much more useful picture than "10 mistakes is a lot! No it's not a lot! Yeah but they were big mistakes! No they were medium mistakes! They cost us the whole season because everyone got depressed!" You're essentially choosing to measure nothing at all as an alternative.

 

EDIT: I'm saying nothing at all about pretending things don't exist.

Edited October 14, 2013 by Eminor3rd

[southsider2k5]

What's the alternative? If we can measure 80% of something and assume that the rest is unpredictable, we have a much more useful picture than "10 mistakes is a lot! No it's not a lot! Yeah but they were big mistakes! No they were medium mistakes! They cost us the whole season because everyone got depressed!" You're essentially choosing to measure nothing at all as an alternative.

 

All outs are not created equally. Any stat that pretends they are is flawed. Pretending that base running mistakes are somehow equal to an infield pop up is just silly. Saying it can't be quantified so therefore it can't be factored isn't good enough either. There is obviously an effect, otherwise it wouldn't be talked about as much as it does when compared to other outs. Common sense tells you that.

[witesoxfan]

All outs are not created equally. Any stat that pretends they are is flawed. Pretending that base running mistakes are somehow equal to an infield pop up is just silly. Saying it can't be quantified so therefore it can't be factored isn't good enough either. There is obviously an effect, otherwise it wouldn't be talked about as much as it does when compared to other outs. Common sense tells you that.

 

But outright saying Alejandro De Aza is a bad base runner, as in he hurt the team, because he made a few mistakes is just as egregious. I have no problem if you say he is a dumb base runner or that he could improve and eliminate those silly mistakes. The fact is that he provided excess value on the base paths, even if that was just because of his speed.

[Balta1701]

All outs are not created equally. Any stat that pretends they are is flawed. Pretending that base running mistakes are somehow equal to an infield pop up is just silly. Saying it can't be quantified so therefore it can't be factored isn't good enough either. There is obviously an effect, otherwise it wouldn't be talked about as much as it does when compared to other outs. Common sense tells you that.

I can't imagine that the baserunning stats ignore the situation when they're calculating how valuable an out/stolen base/move from first to third is. That should be quite readily quantified by the change in the relative probability of scoring between having the runner on first and the runner retired/extra out.

[Eminor3rd]

All outs are not created equally. Any stat that pretends they are is flawed. Pretending that base running mistakes are somehow equal to an infield pop up is just silly. Saying it can't be quantified so therefore it can't be factored isn't good enough either. There is obviously an effect, otherwise it wouldn't be talked about as much as it does when compared to other outs. Common sense tells you that.

 

Why are you assuming that people think all outs have the same value?

 

That's why these stats are based on linear weights. Run values are produced based on the weighted differences in base/out states. For example, if De Aza is caught stealing with 2 outs and a runner on third, it's far more harmful than if he's caught stealing (or picked off) with the bases otherwise empty and no outs. All that stuff is baked in, league adjusted and updated every season based on changes in the run environment. The only thing that isn't here is the "emotional impact' on the rest of the team.

 

It seems like the only people that are against these stats are those who haven't taken the time to understand them.

[southsider2k5]

Why are you assuming that people think all outs have the same value?

 

That's why these stats are based on linear weights. Run values are produced based on the weighted differences in base/out states. For example, if De Aza is caught stealing with 2 outs and a runner on third, it's far more harmful than if he's caught stealing (or picked off) with the bases otherwise empty and no outs. All that stuff is baked in, league adjusted and updated every season based on changes in the run environment. The only thing that isn't here is the "emotional impact' on the rest of the team.

 

It seems like the only people that are against these stats are those who haven't taken the time to understand them.

 

I'm against ones that make no sense in reality.

 

[Eminor3rd]

I'm against ones that make no sense in reality.

 

Yet you clearly have no idea how it works, because you keep accusing it of not taking factors into consideration that it DOES take into consideration. How can you be against something you don't know about? How do you know it makes no sense in reality?

Edited October 14, 2013 by Eminor3rd

[southsider2k5]

Yet you clearly have no idea how it works, because you keep accusing it of not taking factors into consideration that it DOES take into consideration. How can you be against something you don't know about? How do you know it makes no sense in reality?

 

If the stat states that De Aza was a good baserunner, it makes no sense in reality. Period.

[Eminor3rd]

If the stat states that De Aza was a good baserunner, it makes no sense in reality. Period.

 

According to what? Make an argument for it. It isn't good enough for you to just say "because I think so." Show me why the widely accepted state-of-the-art measurement is wrong.

 

By the way, I'm not being rhetorical. People are always questioning these things with reasoned arguments, and it drives them to constantly improve the stats.

[southsider2k5]

According to what? Make an argument for it. It isn't good enough for you to just say "because I think so." Show me why the widely accepted state-of-the-art measurement is wrong.

 

By the way, I'm not being rhetorical. People are always questioning these things with reasoned arguments, and it drives them to constantly improve the stats.

 

Show me where this is "widely accepted" and "state of the art". How many franchises use this number for anything of substance, such as game planning and utilization of their base running strategies.

[knightni]

I'm confused about this entire argument. :|

[witesoxfan]

Let's make an analogy. Anybody suggesting Dunn is a good hitter based on his average is going to be chastised. Anybody suggesting Dunn is a good hitter because of his body of work is likely to be closer to the truth, but still somewhat wrong. People suggesting that Adam Dunn can provide positive value to a team based on his body of work but can still acknowledge that he's a flawed player are going to be right. I mean, if Dunn wasn't strong and couldn't hit the ball a long ways, he wouldn't be valuable, would he?

 

If we can't get past this concept, then there won't be agreement.

 

 

 

[Eminor3rd]

Show me where this is "widely accepted" and "state of the art". How many franchises use this number for anything of substance, such as game planning and utilization of their base running strategies.

 

Who knows? Franchises don't tell anyone what they use. But linear weights metrics are the basis of the research done by tons of analysts that have been hired by teams over the years, from FG guys like Matt Swartz (Orioles) to BP scouts like Kevin Goldstein (Astros). There are no less than three conferences per year led by organizations like the Society for American Baseball Research and Baseball America where these statistics are unveiled, discussed, and scrutinized. Dave Cameron and several FanGraphs staff are at the forefront of much of this. Their particular set of constants and formulae are among the most popular in existence, and are particularly useful in relation to Baseball References formulae because FG prioritizes accuracy over completeness whereas BR does the opposite, giving us a wide palette of different angles from which to reference in a field that is difficult to predict.

 

I'm not challenging your opinion, I'm challenging you to take a look at this stuff before you form your opinion. It's fine if you ultimately don't agree with it, but at least try to make sense of it before you decide it's BS.

Edited October 14, 2013 by Eminor3rd