Peraza

Hawk said:
I confess, I was really just trying to goad you into bringing up the D1 thing again
Click to expand...

You're still deflecting. I'm not shocked.

What data should I be referencing if not BRef or FG? What data/experience do you draw from?

You are oddly silent on the topic. Almost like you don't want to admit to being a low information fan.
 
yeezus said:
This all started with zito calling the Kemp move moronic and then reading off some FG numbers. I said something to effect of "that's great, but teams don't just look at FG to make decisions." Then a barrage of posts saying "TEAMS MOST DEFINITELY LOOK AT FANGRAPHS YOU'RE NAIVE IF YOU THINK THEY DON'T AT ALL" came up. Then you interpreted something as someone trying to "disprove" WAR or saying the eyeball test is better when that literally didn't happen one single time.

In short, I never said anyone said teams only look at FG. I said analyzing the Kemp move by looking at FG is a lot cruder than the way the FO would have done it. Now we're here, and apparently WAR is being attacked but I don't know where.
Click to expand...

To defend zito I don't see how you can't see that point of view. They Braves added ~$8.5 million in salary per year for 3 years for Kemp. The last 1 5/6's seasons Kemp has provided 0.5 total WAR. If he does that the next two seasons the Braves will have paid close to 17 million for 0.5 WAR. With the going rate of 8 million per WAR that is a horrible use of money. And one could consider that moronic. Now there are other considerations I won't get into that I personally agree with but the reality is the Braves will be overpaying for Kemp. And to some people that is not a good thing.
 
thewupk said:
I'm not missing anything. What is your issue with fangraphs? The data that it provides? I am curious on why you bash it so often? That it disagrees with your viewpoint?
Click to expand...

Again: Where have I bashed fangraphs? I love that site. I'm on it every day. I think it's great. I have zero issue with it. So please show me where I've bashed it.
 
thewupk said:
Yes I agree that teams have their own models that they use. However I don't believe they differ all that much on the on field performance side since the raw data is going to be pretty close to the same.
Click to expand...

The raw data isn't the same though. Dave Cameron (as ensheff mentioned earlier) as said as much. Each individual team pours a hundred fold more resources into the data collection process than Fangraphs does. I'm sure the valuations correlate strongly on the aggregate, but I'm sure there plenty of examples of individual players whose individual valuations differ dramatically.

Fangraphs and Baseball Reference have a version of WAR that on the aggregate probably come close to equaling each other, but in certain cases they differ dramatically. The Braves may have a model that tells them that Matt Kemp is worth 3 wins per season (or maybe not). It's about as close as the difference between what FG and Baseball reference say about Ender Inciarte. What everyone is trying to say is don't take everything FG produces at face value.
 
thewupk said:
To defend zito I don't see how you can't see that point of view. They Braves added ~$8.5 million in salary per year for 3 years for Kemp. The last 1 5/6's seasons Kemp has provided 0.5 total WAR. If he does that the next two seasons the Braves will have paid close to 17 million for 0.5 WAR. With the going rate of 8 million per WAR that is a horrible use of money. And one could consider that moronic. Now there are other considerations I won't get into that I personally agree with but the reality is the Braves will be overpaying for Kemp. And to some people that is not a good thing.
Click to expand...

And again: The Braves did not analyze whether or not to make that move by going to fangraphs and looking at Kemp's WAR. Much more went into it. That's not an attach on fangraphs, or WAR. Perhaps they feel the defense in LF won't matter as much as WAR would suggest, and the added power to the lineup outweighs it more than WAR would suggest. I'm sure the FO is capable of doing the WAR x Salary calculation. It's not like this is information they missed, or were too moronic to comprehend.
 
Enscheff said:
They don't differ all that much, except probably when it comes to evaluating defensive contributions.
Click to expand...

I'm sure that's not true, at least in the sense of predicting future production. The sabermetric community hasn't reached a consensus on BABIP, the value of the strike-out, hard hit balls, etc. We are just now starting to grasp these issues since Statcast data is starting to trickle out to the general public.
 
chop2chip said:
The raw data isn't the same though. Dave Cameron (as ensheff mentioned earlier) as said as much. Each individual team pours a hundred fold more resources into the data collection process than Fangraphs does. I'm sure the valuations correlate strongly on the aggregate, but I'm sure there plenty of examples of individual players whose individual valuations differ dramatically.

Fangraphs and Baseball Reference have a version of WAR that on the aggregate probably come close to equaling each other, but in certain cases they differ dramatically. The Braves may have a model that tells them that Matt Kemp is worth 3 wins per season (or maybe not). It's about as close as the difference between what FG and Baseball reference say about Ender Inciarte. What everyone is trying to say is don't take everything FG produces at face value.
Click to expand...

It is very interesting to note the large differences in WAR values between BRef and FG on players like Cano and Braun (going from memory). For example, FG has Cano with a very good career WAR of 48, but BRef has him a fringe-HoF worthy 61. The differences are almost strictly related to defensive values.

Discrepancies like that are why I don't put much stock into defensive valuations compared to offensive valuations. I think they are useful to give us a general idea of who is a good defender (Inciarte) and who is a bad defender (Kemp), but I think they do a poor job comparing offense value to defense value.
 
If the Braves have a model that says Kemp is worth 3 wins per year, I offer my services to help adjust it.
 
chop2chip said:
I'm sure that's not true, at least in the sense of predicting future production. The sabermetric community hasn't reached a consensus on BABIP, the value of the strike-out, hard hit balls, etc. We are just now starting to grasp these issues since Statcast data is starting to trickle out to the general public.
Click to expand...

There is lots of stuff out there with expected BABIP, and other stats revolving around batted ball profiles.

Statcast data is going to finally help clarify defensive contributions though.
 
gilesfan said:
If the Braves have a model that says Kemp is worth 3 wins per year, I offer my services to help adjust it.
Click to expand...

They probably have a model that says he is worth 1-2 WAR, which is why they agreed to take on that much additional money on his contract. They are clearly a team that doesn't value defensive metrics as much as a team like the Cubs (they let Heyward go, traded away Simmons, signed Markakis, traded for Kemp), so it wouldn't be surprising at all if they think Kemp is an overall average player.
 
thewupk said:
To defend zito I don't see how you can't see that point of view. They Braves added ~$8.5 million in salary per year for 3 years for Kemp. The last 1 5/6's seasons Kemp has provided 0.5 total WAR. If he does that the next two seasons the Braves will have paid close to 17 million for 0.5 WAR. With the going rate of 8 million per WAR that is a horrible use of money. And one could consider that moronic. Now there are other considerations I won't get into that I personally agree with but the reality is the Braves will be overpaying for Kemp. And to some people that is not a good thing.
Click to expand...

What if three years from now the Braves model says that Kemp was worth 2 wins per season and Fangraphs says he was worth 0.5 wins per season. That's the difference in saying Kemp was worth 48 million dollars rather than 12 million dollars. That's not to say we can't have our own opinions about players and that we can't evaluate the front office, but it should make you think twice.
 
chop2chip said:
What if three years from now the Braves model says that Kemp was worth 2 wins per season and Fangraphs says he was worth 0.5 wins per season. That's the difference in saying Kemp was worth 48 million dollars rather than 12 million dollars. That's not to say we can't have our own opinions about players and that we can't evaluate the front office, but it should make you think twice.
Click to expand...

The differences in player valuation models is precisely why trades happen. Teams trade for players they value more highly than the other team, and trade away players they value less.

It then all comes down to who was right. Whichever team had the better valuation model wins more games.
 
chop2chip said:
The raw data isn't the same though. Dave Cameron (as ensheff mentioned earlier) as said as much. Each individual team pours a hundred fold more resources into the data collection process than Fangraphs does. I'm sure the valuations correlate strongly on the aggregate, but I'm sure there plenty of examples of individual players whose individual valuations differ dramatically.

Fangraphs and Baseball Reference have a version of WAR that on the aggregate probably come close to equaling each other, but in certain cases they differ dramatically. The Braves may have a model that tells them that Matt Kemp is worth 3 wins per season (or maybe not). It's about as close as the difference between what FG and Baseball reference say about Ender Inciarte. What everyone is trying to say is don't take everything FG produces at face value.
Click to expand...

Yeah. What I'm saying is the raw data is the same. It's all from the same game(s). How they calculate it and what they use is what can be different.
 
Enscheff said:
There is lots of stuff out there with expected BABIP, and other stats revolving around batted ball profiles.

Statcast data is going to finally help clarify defensive contributions though.
Click to expand...
Agreed. But WAR doesn't adjust for expected BABIP and I'm not even saying it should, but there should be some version that is more predictive.
 
chop2chip said:
What if three years from now the Braves model says that Kemp was worth 2 wins per season and Fangraphs says he was worth 0.5 wins per season. That's the difference in saying Kemp was worth 48 million dollars rather than 12 million dollars. That's not to say we can't have our own opinions about players and that we can't evaluate the front office, but it should make you think twice.
Click to expand...

I mean if they have a model that suggests Kemp is an average player then great but I would be very skeptical of such a model. He's barely an average hitter right now.
 
You must log in or register to reply here.
Back
Top