What Does the Wrong Coach Look Like?
What Does the Wrong Coach Look Like?
Mad Son takes a closer look at penalties to see how they reflect on how good a coach can be
Comments
-
I know many coaches don't find penalties have any correlation to winning, Petersen has said so too. Really interesting to see the data on it. I think they consider them a hindrance at times, and there are some penalties that are just stupid.
I wonder if the increased pace of play factors into the mindset of coaches not finding them to be related to winning and losing? -
Excellent read. You shit better articles than what the goons at DM write.
-
I believe that all good coaches realize that if offensive holding or defensive pass interference could be called every play, it's sometimes gonna get called and they take those in stride.
Bad coaching regarding penalties shows itself as repeated false starts, illegal formations, delays of game, offsides or encroachment, and making PFs that are willful, e.g. Taunting, spearing, retaliating, etc., with no discernible improvement over time.
I bag on Peenerman all the time, penalties aren't one of the things about him I call out.
-
R-squared isn't an indicator of correlation. In this case It tells you how well variance in penalties (and only penalties) does to explain variance in win%. The low R2 value tells you: not very well.
There are other variables you would have to control for before making that conclusion (SOS, talent, Experience, conference, etc).
-
When I clicked on the link, I was expecting a giant pic of Willingham.
When you ran the analysis, did you also run a Spearman correlation? I saw you posted the rho squared value for Pearson (linear) correlation, but I'm wondering if you also did a Spearman to see if there is a non-linear relationship.
Edit: Idk what the data looks like, but if the data can be categorized, I'd be curious if you could stratify the data by type of penalty or when penalties occur (which down, whether it's offense, defense, etc.). I'm wonder if there's some correlation between a subset of penalties rather than penalties in general. -
Yes, I recognize what you are saying is technically true. I am assuming I can draw that conclusions as a practicality.BallSacked said:R-squared isn't an indicator of correlation. In this case It tells you how well variance in penalties (and only penalties) does to explain variance in win%. The low R2 value tells you: not very well.
There are other variables you would have to control for before making that conclusion (SOS, talent, Experience, conference, etc). -
The PAC 12 is the world wide leader in penalties. Right or wrong that is a statistical fact. Numerous Pac teams are in the top 20 for penalties per game.
And Canard is right about the types of penalties. If you can't even line up right something is wrong -
It's like revenge of the nerds in here
-
When you first started poasting here I though you were OBK, clearly you aren'tBallSacked said:R-squared isn't an indicator of correlation. In this case It tells you how well variance in penalties (and only penalties) does to explain variance in win%. The low R2 value tells you: not very well.
There are other variables you would have to control for before making that conclusion (SOS, talent, Experience, conference, etc).
-
I'm actually a Bruin (Broog?) who lurked around here until I got dialed in. I'm a big Jordan Zummwalt fan.AZDuck said:
When you first started poasting here I though you were OBK, clearly you aren'tBallSacked said:R-squared isn't an indicator of correlation. In this case It tells you how well variance in penalties (and only penalties) does to explain variance in win%. The low R2 value tells you: not very well.
There are other variables you would have to control for before making that conclusion (SOS, talent, Experience, conference, etc).
So now you know I'm clearly clearly not OBKFS.
