Thread: Correlations Between Punting and Defense and Punting and Winning
View Single Post
Old 04-21-2018, 03:43 PM   #45
kccrow kccrow is offline
MVP
 
kccrow's Avatar
 
Join Date: Oct 2011
Location: Michigan
Quote:
Originally Posted by ntexascardfan View Post
No, it's not...because it's not answering the question I posed.

Which is how much better does a top 5 punter v. a replacement level punter effect the game.

Simplistically posting points and ypg and then the correlation between that and the number of times a punter puts the ball inside the 20 is a poor metric.

A very basic take down of that metric is the fact that teams with poor offenses will tend to have a higher usage rate of their punter, giving that punter, regardless of his quality, more chances to knock down punts inside the 20 yard line.

The Jets had the most punts in the NFL in 2017 and the 3rd most punts knocked down inside the 20. The Jets also had one of the poorest offenses in the NFL in 2017. The Jets also had one of the lowest % of punts inside the 20 in the NFL last season.

Your metrics are way too basic to be evaluating the impact a great punter has on a team. Which, is why the formula needs to be built to determine the effect that a great punter over replacement level has on a team. You can't directly marry punting average/IN 20, net, etc., to defensive rankings and call it a day.

A more compelling measurement would be the % of punts inside the 20 yard line. From a high level 8 of the top 10 teams in the NFL last year in that statistic were in the top 15 in the league in defensive points per game. 6 of the top 10 in PPG were also top 10 in % of total punts inside the 20.

You need to determine how much better or worse a team would be with a great punter. Does having a great punter mean instead of giving up 21.3 points a game the team now gives up 20.5? If so, what value does that give to the teams over the course of a season.

Your simplistic tables and expression of the statistics don't get the job done.
For starters, great idea. However, you've made absolutely false statements on rankings for 2017.

Here is PTS/G and %IN 20 Rank Sorted Low to High
Spoiler!


Here is YDS/G and %IN 20 Ranks Sorted Low to High
Team Yds/G %IN 20 Rank
Spoiler!


I also ran the correlations on PTS/G to %IN 20 and YDS/G to %IN 20 and they are here:

Spoiler!
Posts: 12,679
kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.kccrow 's phone was tapped by Scott Pioli.
     Reply With Quote