Peraza

[chop2chip]

You guys don't think teams read fangraphs? The Braves hired Kiley McDaniel for goodness sake. I would venture to guess it is a consideration among a lot of other things.

The Braves didn't hire Kiley because they read Fangraphs. Coppy was Kiley's boss back with the Yankees. They have a history together.

To your point though, Coppy said in an interview that he used Fangraphs a lot back when he was an assistant in the Wren era.

 

[chop2chip]

Of course. The notion that teams have proprietary statistical data to analyze players is a given one. What I'm saying is that it isn't Fangraphs.

This is undoubtedly true, which is funny because we then analyze their moves by multiplying their Fangraphs WAR by $8 million and subtracting out their contract to determine if they made a good move.

My least favorite part of the statistical evolution is how lazy and myopic the casual stat head has become. We used to dig hard to argue our case on player's value, but now we just look at the last column on the right of a player's fangraph player profile page.

 

[smootness]

Of course. The notion that teams have proprietary statistical data to analyze players is a given one. What I'm saying is that it isn't Fangraphs.

I think teams absolutely read that stuff with more than a passing interest. Or they should. A FO is made of people, same as FG or any other site. Your scouts can only do so much. If you're not taking in as much info as you can, you're not doing your job.

 

[gilesfan]

This is undoubtedly true, which is funny because we then analyze their movies by multiplying their Fangraphs WAR by $8 million and subtracting out their contract to determine if they made a good move.

My least favorite part of the statistical evolution is how lazy and myopic the casual stat head has become. We used to dig hard to argue our case on player's value, but now we just look at the last column on the right of a player's fangraph player profile page.

Because that number is going to give us a better indication than any of us Joe Blows can come up with on our own.

 

[chop2chip]

Because that number is going to give us a better indication than any of us Joe Blows can come up with on our own.

I actually don't think that's true. I know you're statistically savvy enough to dig deep into the vast pool of data to make evaluations on player's ability. For example, when you posted about Matt Duffy's launch angles that led me to research the topic a little bit more and I saw that Ender Inciarte also had a launch angle that would predict that he was a better offensive player than what he was showing at the time. Sure enough his offense has improved as the sample size has grown.

WAR is excellent shorthand for guesstimating approximate value, but there's a lot of publically available statistics that do a better job of demonstrating ability which is a better predictor of future value than past WAR does, IMO.

 

[gilesfan]

I actually don't think that's true. I know you're statistically savvy enough to dig deep into the vast pool of data to make evaluations on player's ability. For example, when you posted about Matt Duffy's launch angles that led me to research the topic a little bit more and I saw that Ender Inciarte also had a launch angle that would predict that he was a better offensive player than what he was showing at the time. Sure enough his offense has improved as the sample size has grown.

WAR is excellent shorthand for guesstimating approximate value, but there's a lot of publically available statistics that do a better job of demonstrating ability which is a better predictor of future value than past WAR does, IMO.

I think launch angle/exit velocity is a good tool, but still kinda in the incubator phase. Obviously Ender isn't as bad as he was earlier in the year and was going to progress.

There are some traits in players that can certainly make you project a player either upward or downward, but in general, WAR is going to tell you how good a player is. Its going to tell you how good a team is.

Who did fangraphs tell us the best teams in the league would be this year? Red Sox, Indians, Astros, Nationals, Cubs, Dodgers.

 

[Enscheff]

Of course. The notion that teams have proprietary statistical data to analyze players is a given one. What I'm saying is that it isn't Fangraphs.

So what's your point? That we should all create our own version of WAR to base our arguments around? Or that we shouldn't base our logic around BRef and FG WAR values at all, and instead use our "gut" like the DBacks do?

Do you have a better metric we can refer to? Maybe one you created and keep up to date?

WAR vs cost is what we have to base our logic around. If I had the Braves version of WAR, I would use that as well.

And there is a pic of Coppy with MLBTR on his PC. It is pretty much a given FO guys refer to these fansites.

 

[yeezus]

So what's your point? That we should all create our own version of WAR to base are arguments around? Or that we shouldn't base our logic around BRef and FG WAR values?

Do you have a better metric we can refer to? Maybe one you created and keep up to date?

WAR vs cost is what we have to base our logic around. If I had the Braves version of WAR, I would use that as well.

And there is a pic of Coppy with MLBTR on his PC. It is pretty much a given FO guys refer to these fansites.

I lost count at the number of straw men in this post. Solid job!

 

[Enscheff]

I think teams absolutely read that stuff with more than a passing interest. Or they should. A FO is made of people, same as FG or any other site. Your scouts can only do so much. If you're not taking in as much info as you can, you're not doing your job.

There is a pic of Coppy with MLBTR on his PC. It is a given that FO guys follow these fansites fairly closely.

 

[bravesfanMatt]

There is a pic of Coppy with MLBTR on his PC. It is a given that FO guys follow these fansites fairly closely.

that pic seemed staged though. it was attached to a trade rumor article IIRC.. But agree that our scouting department has to use their data they collect along with other 'public' sites. I am sure they have a back door access that gives them reporting abilities greater than the average fan. SQL/Oracle ties that can produce data at a much greater detail.

 

[chop2chip]

I think launch angle/exit velocity is a good tool, but still kinda in the incubator phase. Obviously Ender isn't as bad as he was earlier in the year and was going to progress.

There are some traits in players that can certainly make you project a player either upward or downward, but in general, WAR is going to tell you how good a player is. Its going to tell you how good a team is.

Who did fangraphs tell us the best teams in the league would be this year? Red Sox, Indians, Astros, Nationals, Cubs, Dodgers.

You're confusing what I'm saying with "WAR sux", which isn't what I'm trying to say.

I'm arguing the point that it's the best tool we have of evaluating players. Maybe that's true because its really the only all in one tool that we have. I'm just arguing that it's best to separate the components out and challenge the valuations.

 

[Enscheff]

that pic seemed staged though. it was attached to a trade rumor article IIRC.. But agree that our scouting department has to use their data they collect along with other 'public' sites. I am sure they have a back door access that gives them reporting abilities greater than the average fan. SQL/Oracle ties that can produce data at a much greater detail.

It was an Ask Coppy pic and had nothing to do with MLBTR when it was taken. It was later posted to MLBTR's twitter, which is how you probably saw it.

And teams have their own data repositories. I highly doubt they are pulling data from the FG databases. They are pulling their data directly from the same sources FG and BRef pulls theirs. Sites like FG and BRef are not generating raw data, they are analyzing it.

Teams have access to tons of data that aren't available to the public, mostly because it is either expensive to buy (like the radar tracking stuff), or the raw data is generated by the team's themselves (like scouting and medical data), or the data is propietary (like budgetary data). Dave Cameron refers to this additional info teams have all the time as a major limitation to analytics in the public space.

 

[Hawk]

So what's your point? That we should all create our own version of WAR to base our arguments around? Or that we shouldn't base our logic around BRef and FG WAR values at all, and instead use our "gut" like the DBacks do?

Do you have a better metric we can refer to? Maybe one you created and keep up to date?

WAR vs cost is what we have to base our logic around. If I had the Braves version of WAR, I would use that as well.

And there is a pic of Coppy with MLBTR on his PC. It is pretty much a given FO guys refer to these fansites.

My point is/was that you shouldn't get so butthurt when your precious FanGraphs is cut down to size or questioned or scrutinized, especially within the scope of a discussion the context of which is people ridiculing the notion that it is something that a Major League organization seriously considers. To believe that is patently absurd.

It's a website created by fans, for fans.

And there is a pic of Coppy with MLBTR on his PC. It is pretty much a given FO guys refer to these fansites.

Lol ... I'm not even going to touch this.

 

[Enscheff]

My point is/was that you shouldn't get so butthurt when your precious FanGraphs is cut down to size or questioned or scrutinized, especially within the scope of a discussion the context of which is people ridiculing the notion that it is something that a Major League organization seriously considers. To believe that is patently absurd.

It's a website created by fans, for fans.

Lol ... I'm not even going to touch this.

Great, then what stats/sites do you use when trying to research baseball related topics? I would like to expand my knowledge base to include some Hawk-approved sources.

What data do you come armed with? Or are you another low-information fan spewing ignorant opinions with zero fact-based data to back them up?

 

[thewupk]

I don't see why some people try to disprove WAR (the one that fangraphs or baseball-ref use) so much. It correlates highly to actual win totals just like Pythagorean win total does (based on runs allowed vs rusn scored). The teams that have higher WAR (aka the ones who score more runs than they allow) are going to have higher win totals. There are exceptions like the Rangers this year but it's easy to see that their 28-8 record in 1 run games is the major reason their record is what it is. Those teams happen (O's from a couple of years ago is another good example) but they are flukey. Probably the best chance the Braves have for being good next season.

 

[chop2chip]

Great, then what stats/sites do you use when trying to research baseball related topics? I would like to expand my knowledge base to include some Hawk-approved sources.

What data do you come armed with? Or are you another low-information fan spewing ignorant opinions with zero fact-based data to back them up?

You're coming about as close to the point Hawk's clearly trying to make as Kemp does with a fly ball in the gap.

 

[chop2chip]

I don't see why some people try to disprove WAR (the one that fangraphs or baseball-ref use) so much. It correlates highly to actual win totals just like Pythagorean win total does (based on runs allowed vs rusn scored). The teams that have higher WAR (aka the ones who score more runs than they allow) are going to have higher win totals. There are exceptions like the Rangers this year but it's easy to see that their 28-8 record in 1 run games is the major reason their record is what it is. Those teams happen (O's from a couple of years ago is another good example) but they are flukey. Probably the best chance the Braves have for being good next season.

The crux of the discussion is "do teams rely on Fangraphs?" not "Does WAR suck?".

On an aside, just because something is extremely accurate in the aggregate doesn't mean it's necessarily as accurate on an individual basis where player evaluation occurs.

 

[Hawk]

Great, then what stats/sites do you use when trying to research baseball related topics? I would like to expand my knowledge base to include some Hawk-approved sources.

What data do you come armed with? Or are you another low-information fan spewing ignorant opinions with zero fact-based data to back them up?

So is this you finally admitting that FanGraphs is your grail, your solitary source for statistics, the mecca of your 'knowledge' of the sport?

Because you are literally making my argument for me.

I realize that statistics are a crutch for you, but - completely excluding the basic idea that baseball is more than a game of numbers, which is not something even being discussed here - surely you realize that there is an absolute plethora of statistical information available (to the general public) that can factor into the analysis of a particular player that goes well beyond the confines of one website.

 

[thewupk]

The crux of the discussion is "do teams rely on Fangraphs?" not "Does WAR suck?".

On an aside, just because something is extremely accurate in the aggregate doesn't mean it's necessarily as accurate on an individual basis where player evaluation occurs.

Teams rely on the same data that Fangraphs uses however. They may calculate it in a different way but I would bet most offensive models are based on linear weights. Now the values they plug in may differ. Same with defensive data. They may have their own methods but I would also bet they are based on actual play by play data.

And you may be right on the individual basis part. But it's the best we have right now and again it all correlates very well. I'm going to give it the benefit of the doubt until something better comes along instead of just basing my opinions on the eyeball test or an ancient thought process on the game.

 

[yeezus]

I don't see why some people try to disprove WAR (the one that fangraphs or baseball-ref use) so much.

oh. my. lord.