JotCast | Dan Szymborski FanGraphs Chat

Dan Szymborski FanGraphs Chat - 8/24/23

Dan Szymborski

12:00

And we are LIVE

Dan's Mom

12:01

Daniel, I am still waiting for my DeadLinesmas present. Should I just come over and pick it up?

Dan Szymborski

12:01

No, they're supposed to be presents for people who know what the trade deadline is.

BlueJayMatt

12:01

Which13 Jays players? I hit return and the question got sent early, sorry Dan

In your excellent article this morning you talk about Toronto saying "Blue Jays have an absurd 13"..."offensive players under contract who project at 1.5 WAR or more in 2024".

Dan Szymborski

12:02

You don't expect me to give *all* the secrets away do ya?

12:03

But let's just say that Schneider and Orelvis and Roden are there

and Barger hasn't fallen below the 1.5 line

and Lantigua's there

Bob

12:05

ZiPS was unusually optimistic about Bogaerts deal this past winter. You re-ran his projections for your recent article, but I didn’t get a sense of how much they have changed or how likely the deal is to be under water now that we’re mostly through a soft first season. Could you comment on that a bit?

Dan Szymborski

12:05

It's definitely come down a bit

12:06

especially the out years

12:07

It's become more underwater

Stevie

12:07

I feel somewhat responsible for jinxing Ohtani now. Stevie: Is there any upside to Ohtani pitching again this year?
2:55
Jay Jaffe: Yes. Because he’s f’ing good at it, in the middle of perhaps the greatest season ever, and we want to be entertained!

Steve

12:08

Dan, does ZiPS take into account any x-stats? Is there any evidence that something like xwOBA has useful predictive value?

Dan Szymborski

12:08

ZiPS uses its own thing which I've talked about a lot lately!

and it *is* desigend to be predictive

xwOBA has predictive value, though it frequently comes under other things

in predicting the future

12:09

However, people assume that's the end of it, but that's not really how it works. You don't just choose One True Input to Rule Them All

xStats and zStats both have knowledge that the other stuff doesn't have

12:10

so even in the case of xstats where it doesn't do that much better than FIP

the question is whether using xstats *improves* FIP or any of the like measures

12:11

Just as another example, there's no point in the season, historically, in which preseason ZiPS team winning percentage is a worse predictor than actual team winning percentage.

But we don't just choose one and ignore the other. Actual team winning percentage does have some information that a preseason projection doesn't

So if that's all we have, while ZiPS preseason beats in-season, ZiPS preseason *and* in-season beats both.

Kip

12:12

Given the widespread adoption of data I’m curious where the elite teams set themselves apart. What’s something the elite teams are doing with analytics most teams aren’t?

Dan Szymborski

12:13

I'm light on specifics, even though I do talk a lot with teams and also sell them a lot of data. Basically, there's a pretty strict two-way wall in order to preserve my independence.

But the *impression* I get is that the teams that do the best with analytics aren't necessarily the ones with the fanciest systems

But the ones who do the best job at taking data and turning the data into *actions* and *plans*

12:14

For example, while the Astros have always had good analytics, what they've really excelled at is communicating that kind of information to players and to actually utilizing it in decision making

12:15

The worst organizations at this frequently have some really good information but there's a chokepoint somewhere in the decision making process

12:16

Surrey can have all the troops in the world, but if they can't get over the Stirling bridge, it doesn't matter how many they have

12:16

How much of a disadvantage do you think franchises would be at if they relied on publicly-available analytics & rankings (i.e. your projections, BA prospect rankings...) instead of internal tools? A semi-recent Boddy tweet suggests the gap isn't that big

Dan Szymborski

12:16

I agree with Kyle. They have very specific information about their own players, but they'll never have the same detail about players on the rest of the team. Where teams excel is having the manpower and the ability to *implement* the data

12:17

And on some level, if the gap was huge, I wouldn't sell shit on the side.

(Of course, I have some non public stuff too)

Totally clear-eyed realist

12:17

Bobby Witt Jr. is going to look good in pinstripes, isn't he?

Dan Szymborski

12:18

THe thing is, that may be the place the Yankees need the least bit of help

leprekhan

12:18

What is one organization that is currently a dumpster fire, but you think there is actually hope that things could turnaround for them relatively quickly?

Load More Messages

Connecting…

Message History

On Off

Your current subscriptions

None

E-mail Address	Chat Owner	Status	Remove

Enter your email address to be notified when Dan Szymborski hosts a chat