JotCast | Dan Szymborski FanGraphs Chat 2/20/25

Dan Szymborski FanGraphs Chat 2/20/25

Prev Page View Full Chat Next Page

Dan Szymborski

12:17

Because catchers have a LOT of blow-up potential

Guest

12:17

Is ZiPS a bit behind on Drake Baldwin? Seems like the consensus is much higher than projections

Dan Szymborski

12:17

It might, it might not be. Only time can tell

Lowetide

12:17

Colt Keith over Spencer Torkelson at 1B for Detroit based on Zips production (for my points league) but how vulnerable is Keith to an in-season upgrade via trade iyo?

Dan Szymborski

12:17

I think he's fairly solvable

Bumper

12:17

Have you ever looked into how BaseballTradeValues does player values? It looks like they take a very similar approach to ZiPS, but they also seem to get more heat than ZiPS does.

Dan Szymborski

12:18

They get more heat becuase they have a one size fits all tool that everyone digs apart!

I only do a trade piece like once a year, and that edefinitely gets people mad at me

Bumper

12:18

Do you plan to use/currently use stuff+ in ZiPS pitcher projections? If not, would you rather build your own or just include an external model?

Dan Szymborski

12:19

I have my own thing I use because it's easier for me to test projection interactions

Basically any model in ZiPS is one I've put together. It feels like cheating to use the input of others

And when I control the model for something, I can also calibrate it myself

12:20

I don't publish my own stuff numbers (I internally call it Ztuff), since I think the confusion of yet another number out there outweighs the possible gain

which is why I only use zWAR for internal and the ZiPS rudnowns

CFH

12:21

Hello - you've mentioned in the past chats that median projections are mostly solved / there's not a huge amount of variance. Reading the zips top 100, I'm wondering does this go for prospect projections as well (and KBO/NPB)? Or is there more variance there?

Dan Szymborski

12:21

That's hard to say simply because I don't think other projectioniers are doing their own top 100!

Rafael

12:21

Why aren't more teams interested in Yoshida? It would solve a lot of Red Sox lineup problems.

Dan Szymborski

12:21

DH only tends to scare away teams unless they're CERTAIN about the offense

War2d2

12:22

Hey Dan! The Cubs just DFA’d Canario, and I’m unreasonably upset about it (mainly because I’m angry all over again that Ross didn’t play him in 2023 to see what he could be). I get that he’s a 40-45fv guy, but I’d rather roster a 24yo power guy than the smoldering corpse of Justin Turner. What do you (and ZiPS) think?

Dan Szymborski

12:22

It's not so much Turner that bothers me

12:24

But do you REALLY need Hollowell, when the alternative is losing Canario?

Reds Enjoyer

12:25

If you were the White Sox, would you trade Luis Robert right now or would you take the risk that he'll be back to form and you might get a better package at the trade deadline?

Dan Szymborski

12:25

It's hard to say since I don't know what the specific offers ARE

DJ Tofu

12:25

What statistics does ZiPS optimize for? e.g. is it more accurate for PA, BA, HR, etc.? And how does it create the 20/80 percentile results?

Dan Szymborski

12:25

I most care about getting the OPS+/wRC+ and the ERA+/FIP right

plus D

Mariners Fan

12:25

Does ZiPS use strikeout park factors, BIP splits (lefty/right/HR/triple/double/single/outs), or something even more fine? Has T-Mobile made you consider adding more park factors?

Dan Szymborski

12:26

ZiPS uses the BIP splits lefty and righty but it also has an algorithm for park factor/pull-spray tendency interactions

Bumper

12:26

Fun projection idea: in your season simulation, halfway through the simulation, re-project players, and re-weight playing time by the new projection. In theory, this would boost teams with more depth (though I imagine it would take a while to run)

Dan Szymborski

12:26

I kinda hope that the methodology I use hits depth pretty well and that's already quite workload intensive!

Johnny5Alive

12:27

With your zips prospect list, with a guy like Caissie for example, when zips has him pretty high in the top 100 but fan graphs has him off their list, what's the big difference? Zips looking at his power production in the minors while Eric is using traditional scouting that has him scared about the swing and miss?

Dan Szymborski

12:28

It would be quite presumptuous for me to speculate on exactly how Eric feels! ZiPS is probably a fan of the hulk smash guys than most scouting guys are

ZiPS was the highest on Judge and Alonso and also Bryce Eldridge

Arenado

12:28

Do you think he gets moved at the deadline? As a yankees fan I can't stop hoping he might want to join Goldschmidt

Dan Szymborski

12:29

Depends on the situation. It's a possibility

Guest

12:29

Does the Zips projection change much for Pivetta with him moving to SD? Last year SD allowed quite a few HRs and he's a flyball guy...

Dan Szymborski

12:29

A bit. His projection in there should be SD!

Tacoby Bellsbury

12:29

If you ever get back to doing ZiPS Time Warps, can you please what-if Chuck Knoblauch never leaving Minnesota or getting the YiPS?

Guest

12:29

How do you balance your writing obligations with life? Like when there are holidays do you "take off" and just decide you're not going to think about baseball or ideas? Or is it much more fluid and you've come to peace with knowing some days you'll have it and some days you won't and just being fine with it?

Dan Szymborski

12:29

Question one? Pretty poorly actually.

Connecting…