Twitter: Analytics - Love it or Hate it

Denim Chicken

Well-Known Member
Messages
15,676
Reaction score
24,553
No rating system is perfect. Dak’s grade is probably negatively affected by his 9 runs for 6 yards and no 1st downs and of course the 4th and goal sneak no TD.

Football outsiders has Dak with the 6th best passing game of the week, but overall he was 11th due to his poor rushing stats.

Anyone watching the game, be it casual fan or Peyton and Eli, can easily surmise Dak is the far superior QB between him and Hurts.

Fumbles and sacks really hurt a QBs grade.
 

MWH1967

The Cook
Messages
7,114
Reaction score
9,464
At this point you guys have to be better than this. You have the book for crying out loud!:facepalm::)

Cunfucius say" Between the lines is the truth"..:grin:
 

DFWJC

Well-Known Member
Messages
59,870
Reaction score
48,650
CowboysZone LOYAL Fan
Analytics are fine. But wow, this stat is exactly why you have to be really careful when applying them to make judgements on players/teams...


That tells you his program is highly flawed.
 

JD_KaPow

jimnabby
Messages
11,069
Reaction score
10,833
CowboysZone ULTIMATE Fan
Citing PFF as analytics is incorrect, they allow subjectivity into it and the basis of analytics is all subjectivity is removed.
That's not quite true. You can do analytics on qualitative data just fine. But the results will only be as good as the qualitative data you started with and the model you have. The qualitative (subjective) data could be great, but if the model is bad, even if it's "objective," you'll get garbage results.

PFF uses qualitative data but they also have some underlying model of how the football world works. It appears that one aspect of their model is that all plays are equal, which is objective but bad. As for how good the qualitative data is (the play-by-play evaluation of player performance), well I can't speak to that but I'm pretty skeptical.
 

jterrell

Penguinite
Messages
33,858
Reaction score
15,962
Analytics is very much a work in progress for the NFL.
It might be 10 more years before a very good algo is discovered for QB play because it is very, very complicated.

Dak was flat phenomenal last night.
Anyone who watched saw that and he controlled the game entirely with audibles and player movement.
A couple of his best throws didn't even count and he was still the guy who ran the game.
He fed a rush game that crushed the Eagles spirits and picked them apart with accurate passes.

That was a Rodgers level game. Game outcome was never in doubt.

Dallas shut the game down two series early having already scored 40 and we all know McCarthy gave up one series with no desire to manage the first half clock.
But you have an adjusted yards per attempt of 11.46 and 3 TDs with 0 INT on 80% completion percentage??
Philly couldn't get him off the field.
 

CowboysFaninHouston

CowboysFaninDC
Messages
33,880
Reaction score
19,456
where is @khiladi to give us the updated stats against zone defenses between the 5:32 mark of first quarter and 4:18 mark of 3rd quarter, in games played on grass west of Austin, north of mississipi, when the ball is between the 27 yard line and the 18 yard line of the opponent and how much dak struggles in that specific range....
 

Hoofbite

Well-Known Member
Messages
40,796
Reaction score
11,417
They actually have Hurts with a higher grade on the season. I’ve no idea what they’re watching.

Aside from my beliefs on how they seemingly manage to grade every player's snaps by Tuesday, they're obviously giving him a nudge for his rushing numbers.
 
Top