2023-2024 NFL Season Thread • Page 1556

i feel like you think advanced stats are just random made up formulas and not hyper-specific focus on individual metrics of the game.

 

And that's because

They can be focused AND people can use them without knowing what they actually are and that's often what happens

 

I feel like that's just reason to work on understanding them better, not rejecting them outright and defaulting to the most basic metrics that have been proven time and time again (ESPECIALLY at the college level) to not matter.

 

Well yeah, colleges are not always playing the same level of competition so of course stats are wonky. They’re almost playing an entirely different type of football when it’s against an overmatched opponent that plays shitty defense or some shit. in the nfl the stats measure equal territory. Of course Hendon Hooker’s stats against Austin Peay are misleading lol

 

Steelers won so they should have won.

 

Waddle practicing wahoooo

Connor Williams and Jaelan Phillips not practicing qqqqqqqqqq

 

advanced stats can't measure that 'dawg' in each team

dqm

 

just assuming that's tank dell

 

Frankly we are at a state where it's already proven that's too big of an ask

 

EPA is literally expected points added. What the fuck is that aside from a nebulous widespread “is you adding to your team or is you ain’t?”

Defense adjusted value over average (DVOA)… so you’re adjusting based on how good the defense is across from you. Except some defenses are better at defending certain things than others, and a bad team doing the thing they struggle with can certainly have success against them. Also, except division rivals play each other tougher cause they watch more film on each other and can predict tendencies better. Also, except good defenses have bad games sometimes. Does it factor in all that? How much? Who’s to say? How do you even measure that in a way that’s more objective than PFF grades, which are subjective af?

CPOE “completion percentage over expected” is more specific, but again, what is an “expected completion”, is this any throw that looks hard to complete? How do you measure that? Do you factor in catch probability (which itself is flawed, a lot of passes that LOOK low percentage were actually very intentionally placed there and very likely to be completed because the guy was initially open and the defenders got there late)? What about route win rate or contested catch rate, is a contested catch a completion over expected if you kinda DO expect that WR to poster corners back shoulder all day long? Which of these stats do you leave out and why? What’s the right balance to weigh them? It’s just advanced stats that rely on more advanced stats ad nauseam, anywhere in the chain there can be misleading data throwing the whole thing off.

 

Love something as much as Randall hates EPA

 

impossible

 

I’m not criticizing all advanced stats, more specific ones are good, I just hate the ones that try to measure too much all at once and armchair analysts on Twitter treat them like the Bible and can’t be disputed because it’s numbers

 

I have continuously pushed back aganist EPA, although you not understanding something is not justification for it being bad. It's derived from baseball statistics like WAR. It is useful but fantasy twitter's current obsession with it is absurd and stupid.

I also have not brought this up but again, you not understanding something is not a criticism of it.

Third time, same thing.

 

(I wonder if the pushback would be this hard if Lawrence was higher in EPA rankings)

/ducks

 

Lol this is so dumb

 

And at no point have I used any of those things you just ranted about.

 

I’m making an advanced stat called Dustin Points, or DP for short

 

love me a good ole dp

 

You're not alone

I do find PFF and others like them valuable to a degree, but they can get too far into the weeds at times'

Plus the #1 thing I ALWAYS remember is they do not fully know every player's assignment/responsibility on a given play

 

I like that you just assumed you’re being referred to as an armchair analyst based on Randall’s post

 

I say this all the time, but PFF does a lot of things well as far as aggregating data.

Player grades are not one of them. Those are meaningless.

 

There's someone trying to make "scripted play grade" happen and i'm like dude come on lmao