This Has My Attention: New Questions

This is quite a long thread -
Did manage to read a few pages of it, got the gist -
It’s the usual thing- when folks look at the math - they always can come back to its possible - as long as its not x standard deviations away, they dont find anything suspicious- so you can never really get anywhere with these sorts of hypotheses.
Tangentially, my question to the folks who so fervently deny that there are black box stuff occurring is why they need to look
Actuals - leagues played,
Rather than simply do the card maths?

Here's my thing ... Why is it that those who believe in Black Box stuff have to prove it beyond any reasonable doubt, anecdotal evidence is never enough.

But ... Those who say there is no Black Box stuff going on don't have to prove anything. All they have to do is say it doesn't exist based upon their own anecdotal evidence.

This is quite a long thread -
Did manage to read a few pages of it, got the gist -
It’s the usual thing- when folks look at the math - they always can come back to its possible - as long as its not x standard deviations away, they dont find anything suspicious- so you can never really get anywhere with these sorts of hypotheses.
Tangentially, my question to the folks who so fervently deny that there are black box stuff occurring is why they need to look
Actuals - leagues played,
Rather than simply do the card maths?

Here's my thing ... Why is it that those who believe in Black Box stuff have to prove it beyond any reasonable doubt, anecdotal evidence is never enough.

But ... Those who say there is no Black Box stuff going on don't have to prove anything. All they have to do is say it doesn't exist based upon their own anecdotal evidence.

Fwiw - the way I see it, it's basically the same as being presumed innocent until proven otherwise. If Strat is being accused of somehow artificially manipulating game outcomes to benefit a certain group of managers over another, then in my view the onus is on those doing the accusing to offer up some type of real evidence, backed by a significant amount of data.

Its not that I don't get that argument, or rather its not to say thats not a valid counter argument
but getting 'proof' is virtually impossible-
Im not even sure what it would take to do so;
you'd have to run a whole bunch of experiments with your credits to try and figure it out.

But also, that still doesn't really answer the question of why (some) people concurrently claim there aren't any real black box stuff happening while going to actuals in leagues played to mine stats.

Its been bugging me for a bit now- bc it just doesn't make sense.

IF one believes its not happening just use the cards data, no?

This thread moves along its course depending on who’s posting and how they’re feeling that day. I did not open this thread to reignite the black box discussions, however, I can see how we keep coming back to that. I opened this up to discuss one very simple observation:

Why does the expected win rate (pythag) no longer match up with my run differential, like, ever?

There was a time when the 365 pythag matched MLB pythag, in the sense that it fell largely within +/- 3, but more importantly there was a bell curve around that +/- . In other words, my +200 run diff would amount to 100 wins (or whatever it is), but my actual wins could be 105.

That is no longer the case. After 14 pages, I guess because we can’t figure out what might be happening, the conversation creeps back to the black box stuff because there’s nothing left to point to or talk about.

In another thread, a manager posted how the double play information doesn’t align correctly. I also posted an example of league runs scored not matching league runs allowed. I don’t think any of this is nefarious, but it does beg the question of what else might be functioning incorrectly.

A few words on anecdotal evidence too. Posting actually occurring examples is not “anecdotal”. You could argue that it’s well chosen, or not statistically significant, or whatever. But it is REAL evidence. Saying, “well, I ran the numbers and everything is fine” is the anecdotal argument.

This thread moves along its course depending on who’s posting and how they’re feeling that day. I did not open this thread to reignite the black box discussions, however, I can see how we keep coming back to that. I opened this up to discuss one very simple observation:

Why does the expected win rate (pythag) no longer match up with my run differential, like, ever?

There was a time when the 365 pythag matched MLB pythag, in the sense that it fell largely within +/- 3, but more importantly there was a bell curve around that +/- . In other words, my +200 run diff would amount to 100 wins (or whatever it is), but my actual wins could be 105.

That is no longer the case. After 14 pages, I guess because we can’t figure out what might be happening, the conversation creeps back to the black box stuff because there’s nothing left to point to or talk about.

In another thread, a manager posted how the double play information doesn’t align correctly. I also posted an example of league runs scored not matching league runs allowed. I don’t think any of this is nefarious, but it does beg the question of what else might be functioning incorrectly.

A few words on anecdotal evidence too. Posting actually occurring examples is not “anecdotal”. You could argue that it’s well chosen, or not statistically significant, or whatever. But it is REAL evidence. Saying, “well, I ran the numbers and everything is fine” is the anecdotal argument.

Yes,
You are right...
the discussion comes back to 'black box' bc its the explanation for why things are 'weird' statistically.
Its underlying all the questions of why you can get a team of lefty killers struggling against lefty pitchers for example.
Pythag is another one of similar stuff-
Ive seen teams with minus 50 runs have a better record than a team with plus 150 - (thats a 200 run difference)
Have seen the team with the best differential miss the playoffs multiple times.

Yes,
You are right...
the discussion comes back to 'black box' bc its the explanation for why things are 'weird' statistically.
Its underlying all the questions of why you can get a team of lefty killers struggling against lefty pitchers for example.
Pythag is another one of similar stuff-
Ive seen teams with minus 50 runs have a better record than a team with plus 150 - (thats a 200 run difference)
Have seen the team with the best differential miss the playoffs multiple times.

Because of my style of play, it is not uncommon for my team to lead the league in R/D and it is also just as common for my teams to NOT live up to their pythag record and I too have had teams miss the playoffs despite a great R/D

Tangentially, my question to the folks who so fervently deny that there are black box stuff occurring is why they need to look Actuals - leagues played, Rather than simply do the card maths?

Sure it makes sense to avoid doing the math in a statistical simulation.

BTW--the person who poses a hypothesis gets to provide the proof that makes it plausible. Not the other way around.
So I see the rain stick and the dance, and then the rain. Problem is I know better. And apparently that is my problem. Fair enough.

FWIW, I'm not bothered at all by the posing of any types of theories or thoughts suspicions etc.

But when a specific question is continually posed with the but...but...but... when the answer received is not favorable, then it gets tiring so sometimes the answers become less charitable. Sorry.

Tangentially, my question to the folks who so fervently deny that there are black box stuff occurring is why they need to look Actuals - leagues played, Rather than simply do the card maths?

Sure it makes sense to avoid doing the math in a statistical simulation.

BTW--the person who poses a hypothesis gets to provide the proof that makes it plausible. Not the other way around.
So I see the rain stick and the dance, and then the rain. Problem is I know better. And apparently that is my problem. Fair enough.

Frankie-
If you believe there are no black Box things occurring, then it simply becomes a very straight maths problem- the simulations should/would give you meaningless, random data- you could literally go by what’s on the cards, with zero thought to what the simulation actually does- since its not affected by anything other than said cards.
What am I missing here?
Further, the idea that one has to mathematically prove something before it can be accepted as a reasonable/strong hypothesis also isn’t really correct.

Frankie-
If you believe there are no black Box things occurring, then it simply becomes a very straight maths problem- the simulations should/would give you meaningless, random data- you could literally go by what’s on the cards, with zero thought to what the simulation actually does- since its not affected by anything other than said cards.
What am I missing here?
Further, the idea that one has to mathematically prove something before it can be accepted as a reasonable/strong hypothesis also isn’t really correct.

Do you not utilize settings on do not pinch hit, quick relief, quick hook or slow hook, how aggressive to steal bases or conservative (percentages of success rate can go from 60 percent to 85 percent) how aggressive to be on basepaths, set up for how long starters stay in, either by getting s5 pitchers or maybe s9 pitchers, how strong to put outfield arms in, where to put error prone outfielders? BP Hr's on cards depending on the team and league ballparks. When is anything the cards?

Anyone that expects in this game expected wins to represent how good their team is, why not just give points for margin of victory and be done with this win loss BS. Get 3 points for a 25-0 win and .1 point for a 1-0 win, or whatever the mathematical model would generate to get expected wins from run differential and actual results. This of course will mean people who get in a division with a truly terrible team will get a lot of points but it would be fair and equitable.