Game Thread GAME# 63: WIZARDS @ BLAZERS - MARCH 4, 2020 - WEDNESDAY, 7:00, NBCSNW

Here's a look at Offensive Rating vs Wins. Here you can see a correlation.

 

Oh come on! How was I supposed to know that @BonesJones was actually going to pass for once?

 

Again, I am strictly talking about an individual team. Every team is comprised of different players with different strengths and thus has to be looked at accordingly.

It was not a "random" number I picked out of a hat, it was a stat shown on the broadcast last night that is pretty close to their seasonal average.

I have never said that being among the leaders in assists per game leads to winning but in the Blazers case they sure win a lot more games when they don't have terrible assist numbers.

 

Next do a graph that shows scoring more points than the other team vs wins!!! That will prove about as much as you're doing.

 

Why would I pass to someone who isnt ready for a pass? Or out to someone who airballs from 15 feet? Expecting good answers to these questions.

 

Yo I coulda sworn the final score was 104-124

Was there a score keeping error?

 

That man gave me so much shit for my passing for no reason, then he almost drops a dime put of bounds...

People in my league would laugh at yall making fun of my passing because all theyve seen from me is dimes.

 

So graphs with large data sets don't prove much, but the much smaller sample size you are using proves a lot?

 

You're saying that data results from 30 different teams is more relevent to the Portland Trail blazers than data based only on the Portland Trail Blazers?

 

That would be interesting to look at. Does "ball movement", assists, or something along those lines matter to one team, but not others. I'd be curious on the causation of it, but I suppose it's possible.

 

I'm so sick of the "ball movement doesnt matter" narrative.

 

Some teams move the ball well and get to the line more than the average team because of the driving lanes it opens up that they attack...

In that scenario, theres no assist registered even though ball movement is helping the offense get a more efficient result (two free throws). Thatd help create a lack of correlation between offensive rating and assists, and is just one of many examples why assists cant be used to gauge ball movement, and why you cant look at the data sets your looking at as a legitimate gauge of what is helpful for NBA teams.

 

It would be very rare for a measurable to have meaning to one team and not 29 other teams. I guess in general, I would say yes it would make more sense to look at league wide data to see if it is a measure that holds any value. Specifically, regarding assists, I'm not sure why they wouldn't have had a strong correlation league wide for many years, but would have a strong correlation to one team.

There are real flaws with using assists as a measure of "quality".

 

How? All teams are different, so data derived from one team isnt as relevant to any of the other 29 teams, and averaging doesnt make it more relevant. And thats assuming it isnt flawed.

 

Yes, yes, I totally agree. It's why I would never point to assist numbers as proof of anything regarding the efficency of an offense.

Just for the record, I have never said "ball movement doesn't matter". Also to be far, I can't say I've seen any data the proves it does matter either. I would HOPE it matters.

 

Why wouldnt it have a strong correlation?

Some teams thrive in isolation.
Some teams play a slower pace.
Some teams get to the line a ton.
Some teams rely on 3s more.
Etc...

 

Why do you need data to prove it matters? You dont! Dont focus on stats to prove everything that your eyes should tell you.

 

All teams are different, I agree. It sounds to me that you're making a case because all teams are different, you can't compare data from one team to another. I could be wrong, but I believe you compare Portland's rank or data with other teams as a way to back up points you're making.

 

You're asking a great question about why something does or doesn't have a strong correlation and I could only guess. The fact that there is not a strong correlation between wins and assists is not something I thought would be the case, but the data is clear from a mathmatical standpoint.

 

Never said that. Simply said that team based stats are more relevant to that team than the stats of other teams when determi ing what works for said team... Thats different than measuring the quality of an area of the game for a team through comparison.