JotCast | Dan Szymborski FanGraphs Chat

Dan Szymborski FanGraphs Chat - 4/8/21

Prev Page View Full Chat Next Page

Dan Szymborski

12:34

heh

12:34

The composition of the recent-vintage baseball has increased exit velocities. Has it done the same for pitch velocities? Don't think I've read anything on this...

Dan Szymborski

12:34

There's still some quirkiness with this data

Pwn Shop

12:34

Be honest Dan, did you start your projection system because you were betting on baseball? It's why I started mine and then basically found out that people like you were so far ahead of me that I stopped doing it on my own.

Dan Szymborski

12:34

Actually, I did it because I was blogging about transactions and it seemed like something useful

Earned

12:35

I've always thought a pitcher's error that caused runs etc should not be used to protect his ERA. I.E. he earned those runs. What do you think.

Dan Szymborski

12:35

It's weird in a way simply because while it's his fault, it's an "accounting" thing

It's probably best to keep it consistent and have it show up in his pitcher defense as a negative

If we're using a philosophical earned runs that means more "deserved" runs, we have way better ways of doing this than how earned runs are done anyway

Dan

12:36

How does ZiPS calculate percentiles? Is it simply the bound of a confidence interval? Or are you using quantile regression?

Dan Szymborski

12:36

Essentially, ZiPS takes the xth percentile offensive WAR projection. And then calculates the most likely shape of the player to reach that xth percentile projection

Shape is difficult and I've struggled in good methods to do this logically

Aaron

12:36

How long can Hiura continue to swing and miss like he has been before he is sent to AAA?

Dan Szymborski

12:37

If he's still hitting like this, there's no way he doens't go down when AAA starts

Adam

12:37

Roberts had Smith catching and Barnes DHing yesterday, just to remind us all that it's April and the Dodgers can do whatever they want.

Guest

12:37

Re: Porcello, what are your opinions on FIP, especially as a major component of fWAR? The article about Porcello needing a job from a couple of weeks ago used an fWAR chart showing the Porcello was more valuable than Cole, Fried, Gallen, and several others who had pretty good seasons last year. Do you just chalk this up to Porcello being in outlier in that FIP shows him to be much better than he was? At any rate, I don't find the model useful in this case!

Dan Szymborski

12:37

The thing is, FIP is *most* useful in single seasons and less useful in long careers

12:38

In this case, while Porcello was unlucky in one way with BABIP

He was kinda lucky with the HRs

12:39

One of the problem is that the stats *affect* perception of performance, even when no stats are being used

I'd go back to Schilling 2005

12:40

He had like a .370 BABIP and a FIP below 4 but an ERA near 6

And the tendency there was YEAH BABIP, BUT...

Because he looked so bad. But observations aren't going to be truly independent from data

Chuck

12:41

On a scale of 1-10 how concerned are you with Judge already sitting a game with left side soreness?

Dan Szymborski

12:41

Adam

12:41

More likely: Orioles finish over .500 or Padres finish under?

Dan Szymborski

12:41

Latter

Sharp

12:43

How many innings do you need to see from Drew Smyly at this level before his projected ERA moves below 4.00?

Dan Szymborski

12:43

ZiPS is a computer projection!

Kate

12:43

Is lost velocity a bigger deal, not as big a deal or an equally big deal for a guy like Yarbrough who throws so soft vs. a guy with big velo?

Dan Szymborski

12:43

*lost* velocity tends to be a very bad sign

PC Guy

12:44

I'm thinking of water cooling the back of my 3090 but I think I might have to run it to an intake fan/radiator to get it to cool on its own. You think that's bad form? It will probably be the least hot component.

Dan Szymborski

12:44

Whether you use radiator intake or exhaust doesn't actually mean as much as people think. There's enough airflow going through the temperature deltas between air before and after goign through the radiator is not *that* much different

Tater

12:46

how long do M's give Trammell?

Dan Szymborski

12:46

Like a lot of players of this type, I think AAA not being a thing yet buys him a little more time

Earned

12:46

Agreed it's like ERA knew it was trying to be defense-neutral, without really doing it all the way.

Guest

12:46

Dan, thank you for hearing me out. Do you know where I can find team batting average standings with runners in scoring position?