JotCast | Dan Szymborski FanGraphs Chat

Dan Szymborski FanGraphs Chat - 1/20/22

Prev Page View Full Chat Next Page

Dan Szymborski

12:29

Correct.

12:30

I'll give him a nod if he can pull of the Triple Lindy.

DSC

12:30

Are you a DJ Lemahieu truther or is last season a more realistic expectation going forward?

Dan Szymborski

12:30

Somewhere in the middle!

Kate

12:30

ETA for getting ZiPS added to player pages after the last post goes up? Loved reading them this year as always, great job!

Dan Szymborski

12:30

Sometime next week! Gotta check for errors, update for players who have moved (fewer than usual!) and see who I missed

Sirras

12:32

At what point should we just run projections on an at-bat by at-bat basis to fill in games given the lockout has no end in sight?

Dan Szymborski

12:32

Follow your heart! lol

Jonathan

12:32

Have you ever considered using overall or even tool-specific grades from public-facing outlets (like FG) to improve your "ZiPS Top 100 Prospects" list? Or does the inevitable turnover of personnel make it impossible to treat data as synonymous from year to year?

Dan Szymborski

12:32

While I have, I don't tend to like to lock on things that others do.

12:33

Especially because then you have to *really* be aggressive at reviewing how THOSE outlets are doing

Tommy Cho

12:33

I feel for ZiPS & you when Ariel does his projection comparison, obv the playing time estimation is huge for the purpose of his comparison. This year he included ZiPS DC & it shined but still left ZiPS in there which I find annoying. It unfairly paints ZiPS in a bad light & I think it is malpractice for Ariel to include it knowing it is apples to oranges. Seemed intentional this year after a *mistake" last year

Dan Szymborski

12:33

Ariel didn't make a mistake, he has my complete blessing.

12:34

He would never do anything to make ZiPS look bad. He was simply showing how ZiPS, without playing time adjustments, doesn't work for fantasy.

Showing people what to use and not use is an important use for a tool.

If I do a video showing that you don't use an electric drill to make a cheesecake, I'm not bashing the electric drill company!

12:35

ZIPS, when not filtered by a person's knowledge of playing time, is not a direct fantasy tool!

Moose_Bolton

12:35

Have you tried Cinnamon Rolls and chili together? My girlfriend says it's good and I've heard it is from others but I haven't had a chance to try it myself.

Dan Szymborski

12:36

I have not.

and I do not think I am going to

Do you need something THAT sweet to contrast with chili?

Matt

12:36

How often do you check the historical player recommendations that come out? Learn anything fun from looking someone up?

Dan Szymborski

12:36

Quite frequently!

Especially when a name comes up

I forget who had the comp last year, but I hijacked like half the accompanying blurb to talk about Scipio Spinks

Snapshot

12:36

What percentage chance do you give to a) Spring Training and b) Opening Day starting on time this year?

Dan Szymborski

12:36

50%, 80%

Scotty

12:36

One more team for ZIPs left, should we expect the rest of the site to be updated with ZIPs numbers sometime next week? Or is that too optimistic?

Dan Szymborski

12:36

Oh, it'll be next week.

12:37

There's no baseball news that's demanding my time be divided!

that dude who posted on reddit about Dan s

12:37

Do you like Mexican food?

Dan Szymborski

12:37

Yes

Jonathan

12:37

I know OOTP licenses the ZiPS projections. Do you happen to know which parts of it they use and which parts they don't use?

Dan Szymborski

12:37

I give 'em the ZiPS and they use it how they want.

Greg

12:37

Re: ZiPS playing time. I understand ZiPS wasn't built for fantasy and the reason behind why it's not good for fantasy is the PT projections. I'll ask again: How do you calculate PT projections for ZiPS? Re: DC zero PA/IP - just make the PT projection 1 PA/IP so that the rate stats still are generated. I'm not sure projecting 250 PA for someone unlikely to get any PA is more useful (which is what ZiPS does today).

Dan Szymborski

12:37

Near seasons weighted, with adjustments for known, specific i injuries, age, position

x2R

12:38

When ZiPS shows a player's projection, what's the most probable percentile, 50th, other? What determines that?

Dan Szymborski

12:38

The weighted mean is slightly more accurate, but I've taken to just reporting the 50th percentile for ease.

Phil

12:38

So, I understand that Zips is, as you put it, agnostic about playing time. But do you have advice about how to use the projections if we are not agnostic? For example, if I am--for my own purposes, which are mostly purely mental--pretty sure that Tanner Houck will get twice as many innings as Zips says, can I just, you know, double everything? Or will more innings mean more times facing the third time through the order and thus eating it more times, and so some regression should be applied?

Dan Szymborski

12:38

It depends on the question at hand!

Connecting…