JotCast | Ben Clemens FanGraphs Chat - 3/30/26

Ben Clemens FanGraphs Chat - 3/30/26

powered by

Prev Page View Full Chat Next Page

Ben Clemens

2:28

that's not how our odds work, I think it is probably reasonable to say that FanGraphs odds underestimate the Brewers

but we are not doing ad hoc 'oh look at their records, figure out hte managers, do some bayesian weighting...'

like, we're taking a formula that generaly speaking does a good job of approximating results

2:29

I think it's fair to say that it hasn't done a great job of getting the Brewres right. It definitely misses certain teams for certain eras

it's not perfect

we haven't found a good way to reduce bias in an algorithmic sense

Alan

2:29

Obviously drawing conclusions from a single series is a fools errand, but Jordan Walker's O- and Z- swing% have been markedly improved upon, as has his GB rate. Even his already elite bat speed is up. How much data do we need before we can start wonder if he's actually turning the corner?

Ben Clemens

2:29

let's say 200 pitches

bosoxforlife

2:30

Actually down one because the previous runner was already across the plate.

Ben Clemens

2:30

ah yeah sorry, I was thinking 'need two runs to score to win'

2:31

but like... the point is that was a send where it's not in a vacuum, you really do need to consider that not sending him puts you in a good spot, sending him doesn't materially improve your outcomes, and getting thrown out is a disater

Ghost of Ben Clemens Past

2:31

do you personally have any conviction in how much coaching effects players? Thinking Brewers hitting, Reds pitching etc. does a team like the Nationals have more hope in disappointing production (Dylan Crews) or increased consistency (Wood/Abrams) due to a new coaching staff and philosophy? How much effect would something like this have on a farm system?

Ben Clemens

2:31

I don't feel strongly that I can measure it. particularly not before the fact

but I'm absolutely sure coaching affects players

2:32

now, differential coaching? like I said, super hard to measure. But I'm pretty sure that teams have different coaching staffs and philosophies and that those impact how good their players get. I think that generaly speaking, a lot of the way I look at the game (numbers) is downstream of that

2:33

I don't need to know what coaching tip gets given to say that the Brewers are good at figuring out how to improve catcher defense

or you know, similar stuff

Arathorn

2:33

For me, the hard part of evaluating ABS challenges is that certain times call for 50-50 challenges to be smart. If a player misses 3, was he misjudging or just unlucky? Probably the latter but how do we tell/evaluate in anything less than a few full seasons' of data?

Ben Clemens

2:33

indeed!

concur completely

Your Name

2:33

Thoughts on the error conventions? That a pop up can drop right next to a fielder and no error be called is silly. Just get rid of errors at that point.

Ben Clemens

2:34

i mean, for sure

ridiculous

that said.... I mostly look at RA9 over ERA for my actual analysis anyway

and advanced defensive metrics don't care whether something was scored an error or now

not*

2:35

so i guess I'd say that I agree with you that it's silly. But given that I'm not using it to do any analysis, it's hard for me to get too worked up about changing it. there's no way they could reformulate the error that would make it a 'good stat' in terms of handling all edge cases

Your Name

2:35

How can you have the opinion that full ABS isn't ready for prime time, but partial is? Any call can be challenged, therefore there's not a difference.

Ben Clemens

2:35

well, the calls don't get made in real time

there's time to do some kind of post-processing

cleraly not the same thing

2:36

i'm telling you that when they were testing these in the minors, the live instantaneous system didn't work as well as the challenge system

people who watched these minor league games will concur

Oaktown Blues

2:36

So if ABS is prone to errors, will we eventually see a wacky challenge that gets it obviously wrong?

Ben Clemens

2:37

no b/c of the post-processing, is basically my understanding. but you'd have to ask someone a bit more technically in the weeds of those cameras to undersatnd

Pythagoras of Cascadia

2:37

In re: ABS. I've kinda backed off of this opinion a little bit since yesterday watching the O's/Twins game, but I'll state it anyway. Catchers with good eyes are going to be able to show up umpires on missed calls, and umpires aren't going to want to be shown up, so they're going to call more fringe pitches strikes when they know they're behind a catcher with a good eye. (Batters aren't going to be able to challenge/counter this because their view of the zone isn't as good). Ergo the skill of stealing strikes as a catcher shifts a little from body (framing) to eye (knowing/seeing strike zone). But that essentially analogous skill should be essentially as valuable as it was before ABS. Highly simplified view of the world, but perhaps a good razor for prediction.

Ben Clemens

2:37

seems reasonable. My prior is that there will be very little actual effect on the game aside from decreasing missed calls

2:38

but I think that many priors have merit

that's why this is going to be interesting

bosoxforlife

2:38

I was impressed that the Red Sox tore down their preseason rotation plan, used their eyes, and put Early into the #3 spot to begin the season. He has impressed in every start in a way that is rare to see.

Connecting…

Register

Message History

On Off

On Off

On Off

On Off

Your current subscriptions

None

Register now to continue managing your subscriptions

E-mail Address	Chat Owner	Status	Remove

Enter your email address to be notified when Ben Clemens hosts a chat