Warehouse Wire 2037.07: AAA Callups

GM: Brandon Slouck

Moderator: bcslouck

User avatar
bcslouck
General Manager
Posts: 1788
Joined: Sun Oct 02, 2016 10:09 am
Location: Glen Burnie, MD
Has thanked: 35 times
Been thanked: 28 times

Warehouse Wire 2037.07: AAA Callups

Post by bcslouck » Tue Jan 08, 2019 8:05 am

Early into June, the Jimmies find themselves under .500 at 27-35 but only 6.5 games back of the JLA Division and 5.5 games back of the last Wild Card spot. That isn't a bad spot to be in considering the disaster last year was but the team doesn't plan to make any major moves to make a push toward getting back into the playoffs. With that in mind, they are eyeing toward shuffling the roster a bit in an effort to see if some of the young guns can give a jolt to the team.

This leads to the announcement that 1B Joaquin Hebner and SP Clifford Bell have been called up from AAA to make their ML debuts.

Hebner has moved through the system quickly. His numbers are down due to a slow start in June, but the team feels confident he can handle the jump to the ML roster. He provides a defensive upgrade at 1B while sending Luis Miguel Carrillo to the bench. Carrillo has been a disaster this season and may be a DFA candidate in the near future.

Bell has been solid at AA over the last 2 seasons. Despite only being 20 years old, he already has 146.2 innings under his belt. While there were other candidates to come up and start, the team opted for the young lefty.

The team is hoping these 2 will start the young movement on the ML team. Spending the last few seasons acquiring and collecting talent, some of that talent is close to ready and the team would like the youngsters to get some experience in Charm City before being handed the reigns in 2038. Expect more callups in the coming weeks.

In corresponding moves to clear space on the 27 man roster, Guenther Poppmeier has been DFA'd and Carlos Estrada has been optioned to AAA.
GM Brandon Slouck
Charm City Jimmies: 2029 - present (707-753)
2032 Johnson League Champions
2032 Johnson League Atlantic Division Champions
3x Johnson League Wildcard Berth (2031, 2033, 2034)
2x Johnson League GM of the Year (2032, 2034)

User avatar
Lane
PR Director/Historian
Posts: 4126
Joined: Tue Feb 02, 2016 8:18 am
Location: Los Angeles
Has thanked: 25 times
Been thanked: 35 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by Lane » Tue Jan 08, 2019 10:47 am

it seems that the distance from 6 contact to 5 contact for RHB is about as deep as the Grand Canyon.
Stephen Lane
PR Director / Historian / Tech Support
General Manager, Long Beach Surfers
Since 2026

Image


Ex-GM, Amsterdam Neptunes
2025 EBA Champions

User avatar
agrudez
General Manager
Posts: 7591
Joined: Tue Nov 20, 2012 10:30 am
Has thanked: 15 times
Been thanked: 29 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by agrudez » Tue Jan 08, 2019 12:35 pm

Lane wrote:
Tue Jan 08, 2019 10:47 am
it seems that the distance from 6 contact to 5 contact for RHB is about as deep as the Grand Canyon.
*Looks up into the sky and sees a bright light shaped like an "L" and immediately runs towards it*

You called?
League Director: Kyle “agrudez” Stever*
*Also serves as chief muckraker
-Ron, 2025 media guide

Image

User avatar
recte44
Commissioner
Posts: 26419
Joined: Tue Mar 30, 2010 12:14 pm
Location: Oconomowoc, WI
Has thanked: 25 times
Been thanked: 70 times
Contact:

Re: Warehouse Wire 2037.07: AAA Callups

Post by recte44 » Tue Jan 08, 2019 2:46 pm

Oh man, .141????
:eek: :eek: :eek: :eek:
Matt Rectenwald
Commissioner, GM: Las Vegas Hustlers

User avatar
bcslouck
General Manager
Posts: 1788
Joined: Sun Oct 02, 2016 10:09 am
Location: Glen Burnie, MD
Has thanked: 35 times
Been thanked: 28 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by bcslouck » Wed Jan 09, 2019 6:36 am

Yeah. I would of taken his line last year. Oh well.
GM Brandon Slouck
Charm City Jimmies: 2029 - present (707-753)
2032 Johnson League Champions
2032 Johnson League Atlantic Division Champions
3x Johnson League Wildcard Berth (2031, 2033, 2034)
2x Johnson League GM of the Year (2032, 2034)

User avatar
Ted
General Manager
Posts: 4406
Joined: Thu Jun 04, 2015 2:50 pm
Has thanked: 92 times
Been thanked: 62 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by Ted » Wed Jan 09, 2019 8:33 am

It's worth pointing out that his BABIP is a paltry .180, whereas his career mark is around .250. More recently he's been in the .220-.230 range, and he has had a slight downward trend, but this year is a huge departure. His K and BB rates don't really seem much different, and his ratings seem about the same across the board. If I had to bet, I'd say this is all BABIP luck rather than aging, but I wouldn't be sure.
Ted Schmidt
California Crusaders (2021-present)
Image

User avatar
RonCo
League Director
Posts: 9577
Joined: Sat Nov 14, 2015 10:48 pm
Has thanked: 141 times
Been thanked: 78 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by RonCo » Wed Jan 09, 2019 12:05 pm

There are a lot of guys who are struggling like that right now--prompting my "Worst" list. :)
Yellow Springs Nine
Playoff Regular, Landis Averse

User avatar
RonCo
League Director
Posts: 9577
Joined: Sat Nov 14, 2015 10:48 pm
Has thanked: 141 times
Been thanked: 78 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by RonCo » Wed Jan 09, 2019 12:24 pm

here is an interesting chart regarding the Contact rating (which I still say is being mis-used even by those who understand it).

This is a quick Excel pivot table from the stats listing in the game, classifying all hitters by their Contract Rating, batting handedness, and their BABIP/Batting average. Bottom line: a 5s BABIP does not change deeply, but their batting average often does. You can say that about most steps. More on this later as I get a chance...but for now, here's the chart in question.
2037-CONTCTvBABIPvAVG-June.PNG
Yellow Springs Nine
Playoff Regular, Landis Averse

User avatar
RonCo
League Director
Posts: 9577
Joined: Sat Nov 14, 2015 10:48 pm
Has thanked: 141 times
Been thanked: 78 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by RonCo » Wed Jan 09, 2019 1:08 pm

Here's a deeper-dive into hitters with a rating of "7" contact, split by their AVK ratings. Given the state of our league (mid-June), we have a little sample size noise. But you can see the basic trend--especially when you look at switch hitters (who always have the platoon advantage).If I get a bit more time I'll try to do some data pulls for LHB and RHB that split by pitcher handedness to see that, too...but here we have what we have because the data is super-simple to take.

Everyone seems to know that "Contact" is not a real rating. It's there to supposedly predict batting average. The problem is that it's horrible at doing this prediction (and who cares about batting average anyway).

The real value of Contact to a OOTP GM is that it gives you a way to predict BABIP (which is a real skill in OOTP terms).

Here's the chart I've created for this little discussion...I suppose I should make this a feature to hope more people see it:

2037-CON-AVK-BABIP-AVG-handedness-june.PNG

STUDYING THE DATA

Let's look at switch hitters, because the data is smoother due to less platoon noise (I assume). Once you see this concept here, you can apply it to the RHB and LHB charts.

For "7" contact hitters their BABIP averages .302, and scales up as their AVK goes down. This makes sense. But look at the batting average column. Theoretically, this number should stay constant (beacus they are all "7" contact hitters and contact predicts batting average, but as AVK goes down the players' averages go down.

You could argue that the sample size is moderately low, and that the slope is small enough (.258-.267) that it could just be random noise. There's some good reason to see it that way when you're looking at the data the first time. Skepticism is always warranted. I suggest, however, that when you do this over and over again (as I have done), the same pattern eventually emerges. I suggest you look at the data for RHB and LHB now, and see what is essentially the same pattern, but realize that the pattern is a little obscured by a combination of sample size (only 209 AB for 7/10 hitters) and platoon advantage issues that I can't see into with this data set.

WHY IT HAPPENS:

OOTP's CON algorithm isn't actually "horrible." But it does seem to assume a certain scatter of ratings inside the game. Our league, like many online leagues, has our pitcher Stuff ratings insanely high (some of it self-inflicted due to our reliever conversions). This part is complicated, but given the order of events inside the engine this adds up to mean we get to that BABIP rating less often than the game thinks we will. So it's estimate of Batting average (the Contact rating) is wrong--and it's more wrong as AVK goes down.

What it means for Carrillo

I'm not sure. But at the end of the day, as Ted said, Carrillo's BABIP is coming back into line. He's striking out more than he has recently, though, so his Batting average is way, way down. At 32, will he bounce back? Again, I don't know. But guys like Carrillo are dangerous because when they do fall into a bad "random" streak, they are less than worthless.
Yellow Springs Nine
Playoff Regular, Landis Averse

User avatar
Ted
General Manager
Posts: 4406
Joined: Thu Jun 04, 2015 2:50 pm
Has thanked: 92 times
Been thanked: 62 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by Ted » Wed Jan 09, 2019 1:16 pm

Neat stuff, Ron. Although it just makes me hate the BABIP/AVK inverse relation/setup that exists in the first place. On the other hand, that relationship only exists if I'm looking at contact. Solution? Stop caring about the contact rating except as a means to estimate BABIP. Time to make a spreadsheet that lets me input all the players and churn out BABIP/gappower/power/Eye/AVK. Goodbye contact, you garbage rating, you.
Ted Schmidt
California Crusaders (2021-present)
Image

User avatar
RonCo
League Director
Posts: 9577
Joined: Sat Nov 14, 2015 10:48 pm
Has thanked: 141 times
Been thanked: 78 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by RonCo » Wed Jan 09, 2019 1:28 pm

I admit I've never understood what you and Kyle (and others) mean by "Inverse AVK/BABIP" relationship. There is no inverse relationship in the ratings or in the data by themselves. The best I can get to is that it's valuable to understand that:

CON = Function(BABIP,POW,AVK)

So for a constant CON, as AVK goes up, BABIP is going down ...

But that's like saying that when the sun goes down, darkness increases. So (not to pick on you particularly, because I hear this often from several directions) saying that you hate this inverse relationship as regards to the baseball stat "Batting Average" is like saying you dislike the basic math behind fundamental baseball stats.

Bottom line: if two real life batters hit for the same average and power rate, but one is better at avoiding strikeouts, the other is better at getting hits on balls in play. That's what OOTP's model does.

The "issue" with oddities around the OOTP CON rating are due to the combination of the engine and our skewed STUFF ratings.
Yellow Springs Nine
Playoff Regular, Landis Averse

User avatar
RonCo
League Director
Posts: 9577
Joined: Sat Nov 14, 2015 10:48 pm
Has thanked: 141 times
Been thanked: 78 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by RonCo » Wed Jan 09, 2019 1:51 pm

Continuing from above:

After all, you can say the same thing about all three components: BABIP, AVK, POW.

All three are positively correlated to Batting Average. As Power goes up, CONTACT goes up. As BABIP goes up, CONTACT goes up. As AVK goes up, CONTACT goes up. That's because for a hitter, Batting Average is an output of three skills: AVK, POWER, and BABIP. The three inputs are not correlated in any way, but combine to create a statistical output that we can measure.

This is the right physical equation for the way these three baseball skills work.

The game, rather than giving us the three inputs in a clean fashion, gives us two inputs and a result (because Markus thinks the average fan wants to see "batting average" as a rating). Then it asks us to do a middle-grade algebra problem to estimate/calculate the third input.

I admit I'd like to see Markus just give us the BABIP rating. But while I know this BABIP thing _is_ controversial, I personally have never understood why it would be so.
Yellow Springs Nine
Playoff Regular, Landis Averse

User avatar
Ted
General Manager
Posts: 4406
Joined: Thu Jun 04, 2015 2:50 pm
Has thanked: 92 times
Been thanked: 62 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by Ted » Wed Jan 09, 2019 2:20 pm

RonCo wrote:
Wed Jan 09, 2019 1:28 pm
I admit I've never understood what you and Kyle (and others) mean by "Inverse AVK/BABIP" relationship. There is no inverse relationship in the ratings or in the data by themselves. The best I can get to is that it's valuable to understand that:

CON = Function(BABIP,POW,AVK)

So for a constant CON, as AVK goes up, BABIP is going down ...

But that's like saying that when the sun goes down, darkness increases. So (not to pick on you particularly, because I hear this often from several directions) saying that you hate this inverse relationship as regards to the baseball stat "Batting Average" is like saying you dislike the basic math behind fundamental baseball stats.

Bottom line: if two real life batters hit for the same average and power rate, but one is better at avoiding strikeouts, the other is better at getting hits on balls in play. That's what OOTP's model does.

The "issue" with oddities around the OOTP CON rating are due to the combination of the engine and our skewed STUFF ratings.
It's only "inverse" for a constant contact rating or constant average, as you pointed out. It's also a very artificial construct. BABIP is function of where a guy hits a baseball and how fast he runs. Avoiding strikeouts is about pitch selection and ability to get the bat head on the ball. These are relatively independent skills as much has anything a hitter does can be.

"Bottom line: if two real life batters hit for the same average and power rate, but one is better at avoiding strikeouts, the other is better at getting hits on balls in play. That's what OOTP's model does."

This is exactly what I (and I think Kyle) have a problem with. While this is mathematically correct, it is not really that important in terms of outcomes. It wouldn't even really be a problem, but the game provides us with "contact/avg" as a thing to track/value, so I have to deal with the CON=Function(BABIP,POW,AVK) which you described.

Frankly, I don't care about CON. No one should care about CON. CON gets us a less meaningful intermediate value that is very noisy. And if the game is using CON to drive other decisions internally, then it is doubly wrong.

It all goes back to an over-reliance on batting average we are all conditioned to have. CON as a surrogate for batting average, is displayed to make the game familiar to us (especially traditionalists) but it's far less important than some combination on on base percentage and extra base hits. It tells us far less.

I disagree with using BABIP as in input. That's fundamentally wrong, or at least a huge oversimplification that takes the fun out of things, but I'd rather they gave me BABIP than CON (holding the same relationship as you described above). I'm aware of the mathematical relationship and do not care much for the outcome it gives me. Far less than I would about BABIP (or the components of hitting that it is calculated from) .

So anyway, that's why I'm annoyed. I'm annoyed that I'm forced to infer meaningful ratings by holding constant a known number to form an inverse relationship yielding the result I want. I also fear that the game engine uses CON as an input for things, which is even worse as it is an output of something else that the game should also be using as an output.

Were I starting over, I would make the ratings as such, as I feel it would be more "reaL"

OLD rating (what it is)->New rating (what it is)
CON (the function you described above, a batting avg surrogate)->CON (the hit tool, essentially how good a guy is at getting the good part of the bat on the ball)
Gap Power (how many doubles/triples) -> Launch angle
Power (how many home runs hit) -> Exit velocity
Eye (how many BB) -> patience/approach, selection
AVK (how good at not striking out)-> Adjustment to two strikes (how much does the player cut down with 2 strikes)

As you can see, most of our current ratings don't represent actual player ability, but rather are either numerical inputs or specific values to counter pitcher abilities. I'd prefer they actually represent player abilities. As we've discussed before, this would require a complete overhaul of the game.

But what we have doesn't make sense. How the hell are EYE and AVK separate ratings? 10 eye, 3 AVK? Okay so the guy is a savant at balls vs strikes but can't put the bat on the ball when it's in the zone to save his life. So, the best way to rationalize this is that EYE is pitch selection/patience, and AVK is actually what most of us think of as "contact". That makes CON annoying and really just a gateway to BABIP. And this is why I'm annoyed with the setup.
Ted Schmidt
California Crusaders (2021-present)
Image

User avatar
Ted
General Manager
Posts: 4406
Joined: Thu Jun 04, 2015 2:50 pm
Has thanked: 92 times
Been thanked: 62 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by Ted » Wed Jan 09, 2019 2:21 pm

RonCo wrote:
Wed Jan 09, 2019 1:51 pm
Continuing from above:

After all, you can say the same thing about all three components: BABIP, AVK, POW.

All three are positively correlated to Batting Average. As Power goes up, CONTACT goes up. As BABIP goes up, CONTACT goes up. As AVK goes up, CONTACT goes up. That's because for a hitter, Batting Average is an output of three skills: AVK, POWER, and BABIP. The three inputs are not correlated in any way, but combine to create a statistical output that we can measure.

This is the right physical equation for the way these three baseball skills work.

The game, rather than giving us the three inputs in a clean fashion, gives us two inputs and a result (because Markus thinks the average fan wants to see "batting average" as a rating). Then it asks us to do a middle-grade algebra problem to estimate/calculate the third input.

I admit I'd like to see Markus just give us the BABIP rating. But while I know this BABIP thing _is_ controversial, I personally have never understood why it would be so.
I think I just used more words to say this. So we are agreed. Good. Markus needs to prioritize re labeling ratings to what they are.
Ted Schmidt
California Crusaders (2021-present)
Image

User avatar
RonCo
League Director
Posts: 9577
Joined: Sat Nov 14, 2015 10:48 pm
Has thanked: 141 times
Been thanked: 78 times

Re: Warehouse Wire 2037.07: AAA Callups

Post by RonCo » Wed Jan 09, 2019 2:35 pm

Yes, part of the issue is that you want to change the base inputs of the whole model--which I agree with in general. But we can have a whole mess of conversations about how to model more realistic inputs. BABIP has several inputs if you get into the neep of things.

For reasons that make very good sense on the whole (it's tied to Log5 equations for its outputs), OOTP uses stat expectations for inputs. In that light, a batter's skills are defined as:

Walk rate (Eye)
Strikeout rate (Avk)
Home run rate (Power)
BABIP
Double Rate (Gap)
HBP

There are also a few other things involved.

All of these statistical outputs follow real life hitters, so qualify as real skills at the level the game works. As a counter, BABIP is not a pitcher skill (it does not follow pitchers year to year in most cases). In that light, BABIP is not generally an OOTP pitcher's skill.

So, I argue that in context of the game's design as it is, these "make sense."

Though, again, I would prefer Markus to show us the BABIP rating...and in the big picture I agree the game would be cooler if Markus broke things down with a more modern take physical skills rather than "statistical" ones.
Yellow Springs Nine
Playoff Regular, Landis Averse

Post Reply Previous topicNext topic

Return to “Charm City Jimmies”