Your opinion please.

Moderator: Palmtana

  • Author
  • Message
Offline

Dr. Biocide

  • Posts: 37
  • Joined: Mon Jan 07, 2013 10:15 pm

Your opinion please.

PostSat Jul 16, 2022 5:07 pm

What is a homerun worth in singles. A triple, double, walk, also in singles.
Offline

MaxPower

  • Posts: 765
  • Joined: Sat Sep 24, 2016 2:12 am

Re: Your opinion please.

PostSat Jul 16, 2022 6:00 pm

No need for opinions, just math. The offensive events relative to each other:

SB 1
BB 2
1B 3
2B 5
3B 7
HR 9

https://tht.fangraphs.com/why-woba-works/
Offline

Dr. Biocide

  • Posts: 37
  • Joined: Mon Jan 07, 2013 10:15 pm

Re: Your opinion please.

PostSat Jul 16, 2022 8:01 pm

MaxPower wrote:No need for opinions, just math. The offensive events relative to each other:

SB 1
BB 2
1B 3
2B 5
3B 7
HR 9

https://tht.fangraphs.com/why-woba-works/

Sorry. Yes of course. I was more curious about how those numbers hold up within the, basically, all-star teams we field in ATG9. Is Strat "nailing" the math?
Offline

MaxPower

  • Posts: 765
  • Joined: Sat Sep 24, 2016 2:12 am

Re: Your opinion please.

PostSat Jul 16, 2022 11:54 pm

Ah, great question. Well an $80 team in a neutral park averages 4.85 r/g by my reckoning, which is about what MLB saw in 2019. So probably not too far off from Palmer's rule of thumb I posted earlier. Obviously at higher caps things start to distort more.

Pelletier (PBUH) claims that doubles in Strat are worth slightly less than in MLB. viewtopic.php?f=17&t=639204 I've never studied the issue systematically but I have no reason to doubt him given how accurate the rest of his data has proven to be.

I have a file I use that shows how run expectancy changes as run environment fluctuates. It's funny because this file is the basis for every calculation in my system, but I have no idea where I got it from. I found it on my hard drive years after downloading it. And then incorporated it into my Strat gaming years after that. But it's truly the Rosetta Stone. Anyway here it is: https://docs.google.com/spreadsheets/d/ ... t-F4RKvJM/
Offline

jet40

  • Posts: 353
  • Joined: Sun Aug 26, 2012 10:37 am

Re: Your opinion please.

PostSun Jul 17, 2022 6:48 am

One thing that should be mentioned, all these 'relative values' change pretty much every single plate appearance. Everything is situational. A walk is more valuable leading off an inning as opposed to two outs nobody on. A single is more valuable with Vince Coleman on second and two outs then Ted Klu standing at second. The beauty of strat is pretty much every league presents different problems for a manager.
Getting back to the math, what MaxPower provided is very similar to the simple system I currently use.

John
Offline

labratory

  • Posts: 428
  • Joined: Sat Sep 29, 2012 11:33 am

Re: Your opinion please.

PostSun Jul 17, 2022 3:20 pm

Pelletier (PBUH) claims that doubles in Strat are worth slightly less than in MLB. viewtopic.php?f=17&t=639204 I've never studied the issue systematically but I have no reason to doubt him given how accurate the rest of his data has proven to be.

I've tried to run some simulations and doubles and triples always come out lower in value than the standard values. I always attributed it to the crudeness of the simulation (not accounting for sacrifice flies, etc.).
I didn't realize that Marc also found the same thing.


There’s another way to explain the difference. There are essentially three types of negative impacts of making an out:

1. Removing a runner who is on base, which occurs during a caught stealing or double play.
2. Decreasing the value of a runner on base, because he now has fewer outs in which to score during the rest of the inning.
3. Reducing the potential number of runs a team can score in a game, by reducing the number of outs left in the game.

In the wOBA article, they refer to the "ticking clock" negative aspect of an out which is related to the value of the extra plate appearance that you lose at the end of the game. This is #3.
But the ticking clock also plays into each individual inning whether there are runners on or not. Yes, it decreases the value of any runners (#2). But it also decreases the run expectancy of the inning even if there are no runners on base.

Where is that factored in?
Offline

barrmorris

  • Posts: 363
  • Joined: Fri Mar 04, 2022 6:25 pm

Re: Your opinion please.

PostSun Jul 17, 2022 6:12 pm

Some additional information

Diamond Dope uses the formula
BR=.47x singles+.78x doubles+1.09x triples+1.4xHRs+.33xBB/HBP-.25xouts
Note that these are numbers you can calculate from a hitters/pitchers card (so it excludes SB)

I wondered where the factors came from and whether they were based on the SOM game or mlb. I did a least squares analysis of Barnstormers and Players Championship data combined and got the following:

BR=.46xsingles+.72xdoubles+1.19xtriples+1.37xHRs+.27xBB/HBP-.1xouts

Separately, Barnstormers and PC give similar factors. The problem with linear regression on this data is that the independent variables (singles, doubles, etc.) are not truly independent (they are correlated). There are modeling techniques to address this collinearity but I have not tried them yet.
Offline

MaxPower

  • Posts: 765
  • Joined: Sat Sep 24, 2016 2:12 am

Re: Your opinion please.

PostSun Jul 17, 2022 6:47 pm

labratory wrote:
Pelletier (PBUH) claims that doubles in Strat are worth slightly less than in MLB. viewtopic.php?f=17&t=639204 I've never studied the issue systematically but I have no reason to doubt him given how accurate the rest of his data has proven to be.

I've tried to run some simulations and doubles and triples always come out lower in value than the standard values. I always attributed it to the crudeness of the simulation (not accounting for sacrifice flies, etc.).
I didn't realize that Marc also found the same thing.


There’s another way to explain the difference. There are essentially three types of negative impacts of making an out:

1. Removing a runner who is on base, which occurs during a caught stealing or double play.
2. Decreasing the value of a runner on base, because he now has fewer outs in which to score during the rest of the inning.
3. Reducing the potential number of runs a team can score in a game, by reducing the number of outs left in the game.

In the wOBA article, they refer to the "ticking clock" negative aspect of an out which is related to the value of the extra plate appearance that you lose at the end of the game. This is #3.
But the ticking clock also plays into each individual inning whether there are runners on or not. Yes, it decreases the value of any runners (#2). But it also decreases the run expectancy of the inning even if there are no runners on base.

Where is that factored in?

The article doesn't do a great job of communicating it, but the post on outs by Tom Tango that Studeman links to makes more clear that the effects you're asking about are included in #3. In fact Tango calls #3 "Inning Killer" instead of "Ticking Clock."

Very interesting to learn of another data point confirming Pelletier's analysis of doubles. If I had to guess the reason, it would be the max setting "More Baserunning Decisions" creating too many DO** and not enough DO***. Triples, on the other hand, I can't think of any way they would be less valuable than in MLB. Their value should be purely tied up in run environment.
Offline

MaxPower

  • Posts: 765
  • Joined: Sat Sep 24, 2016 2:12 am

Re: Your opinion please.

PostSun Jul 17, 2022 6:56 pm

barrmorris wrote:Some additional information

Diamond Dope uses the formula
BR=.47x singles+.78x doubles+1.09x triples+1.4xHRs+.33xBB/HBP-.25xouts
Note that these are numbers you can calculate from a hitters/pitchers card (so it excludes SB)

I wondered where the factors came from and whether they were based on the SOM game or mlb. I did a least squares analysis of Barnstormers and Players Championship data combined and got the following:

BR=.46xsingles+.72xdoubles+1.19xtriples+1.37xHRs+.27xBB/HBP-.1xouts

Separately, Barnstormers and PC give similar factors. The problem with linear regression on this data is that the independent variables (singles, doubles, etc.) are not truly independent (they are correlated). There are modeling techniques to address this collinearity but I have not tried them yet.

Very cool stuff, love to learn about this kind of analysis.
Offline

MaxPower

  • Posts: 765
  • Joined: Sat Sep 24, 2016 2:12 am

Re: Your opinion please.

PostMon Jul 18, 2022 5:27 am

Actually thinking about it more, triples having less value in Strat could be explained by, for example, HAL being a bad 3rd base coach on sac flies (which he is, too conservative).
Next

Return to Strat-O-Matic Baseball: All-Time Greats

Who is online

Users browsing this forum: No registered users and 39 guests

cron