Whitner still looking for a job

[NewEra]

What info would you use to prove your point?

 

Not like it will matter, I have seen every game of both of their careers. One is much better than the other.

 

I won't use any info, because there isn't any that can prove it either way. I use it by watching play. I've also seen every game of both of their careers. Haven't missed a game since 95. I like Wilson a lot. I like whitner, not a lot due to his inability to keep his mouth shut. My feeling, as stated previously is, whitner = better vs the run, Wilson = better vs the pass. Different players that do different things well. I disagree that he is "much better". If he was much better, he would've been starting and whitner would've been on the bench IMO. Sure, one is a 1st rd pick and the other a udfa, but last years coaching staff had zero ties to either. They didn't draft whitner. They were there to put the best players on the field in order to help them keep their jobs as coaches. Maybin was a 1st rd pick. I didnt see him getting any time over moats or Coleman.

 

You feel Wilson is much better than Whitner. I disagree. End of story.

[ICanSleepWhenI'mDead]

Should we throw him a bone?

What makes you think he'd catch it?

[QB Bills]

What makes you think he'd catch it?

hahahaha

 

nicely played, sir

[Orton's Arm]

Interesting... But what is the time period of the analysis? If he time period is the history of the NFL i wouldn't necessarily have statistical significance for current trends, say the last 5 or 10 years. If it is 5 or 10 years, it's pretty telling. However, I could get a high r squared too with a little work!!!!!!

I don't recall the exact period of analysis, so don't quote me on this. But I'm reasonably certain the period of time they examined was somewhere between one year and five years long--certainly not longer than that.

 

The analysis itself seemed like a fairly standard-issue multiple linear regression analysis. That being said, I have to compliment the author of the piece for writing something which a) had no obvious errors or logical fallacies, and b) which contained some insightful statements about cause and effect and correlation.

 

One of the things I found interesting about the model was that it explained 80% of the variation in teams' winning percentages, even though it didn't account for special teams. This means that at most variation in special teams play explains 20% of differences in winning percentage, and very probably less. So when Marv says that special teams is 1/3 of the game, he's wrong! (Not that this is the only thing about which Marv has been wrong!)

[boyst]

One possible definition of "better player" is "a player who's more able to help your team win games." Using that as a definition, the best QB ever is a better player than the best punter ever, because good quarterbacking is more important than good punting.

 

Another possible definition of "better player" would be "demonstrates a higher level of play at his position." Using this definition, the best punter ever might be roughly as good as the best quarterback ever. Using this second definition, one could say that Whitner and Wilson are about equal as players. But using the first definition, Wilson is clearly a step above Whitner.

Isn't it sad, then, that our best Bill over the last 5 years has been our punter?

[dave mcbride]

I agree with your assessment that Wilson is better against the pass, and Whitner is better against the run.

 

According to a multiple regression study performed by The New York Times, an improvement in pass defense is three times as effective as an analogous improvement in run defense. When the effect of increased interceptions is added in, the margin increases to 4-to-1.

 

If George Wilson is a 50 out of 100 in run defense and a 70 out of 100 in pass defense, and if Whitner is the reverse of that, then Wilson is the significantly better player.

 

Edit: Read the following only if you're interested in statistics speak. The dependent variable of the New York Times' multiple regression analysis was a team's number of wins. Collectively, the independent variables explained 80% of the observed variation in the dependent variable. The independent variables included average yards per pass play (offense), average yards per pass play (defense), average yards per running play (offense), average yards per running play (defense), average INTs per game (offense), average INTs per game (defense). A one standard deviation improvement in your team's average yards per pass play (offense or defense) was three times as effective as a one SD improvement in your team's average yards per running play (offense or defense). A one SD improvement in your team's interception rate (offense or defense) was as effective in producing more wins as a one SD improvement in your team's average yards per running play (offense or defense).

Great info, and it rings true. Thanks.

[eball]

One of the things I found interesting about the model was that it explained 80% of the variation in teams' winning percentages, even though it didn't account for special teams. This means that at most variation in special teams play explains 20% of differences in winning percentage, and very probably less. So when Marv says that special teams is 1/3 of the game, he's wrong! (Not that this is the only thing about which Marv has been wrong!)

Makes sense. Look how good the Bills' ST were with Bobby April (always top 5) and the W-L record still didn't get there. While ST MAY make a difference in a game, they don't USUALLY impact the outcome -- the team with the best O and D generally wins.

 

Which really means -- you just need to make sure your ST are adequate (i.e., not piss poor) and focus on the meat and potatoes of football. I think that is probably common sense, but we've been beaten over the heads for the last 15-20 years with how "important" ST are to being a good football team.