Winning...What works

Some years ago the NFL stats were compared directly against the winning percentage. It was discovered that the number one stat of a qb that correlates with winning is the yards per attempt stat. Steve Young had an impressive ypa in the years that he won. Here are some interesting stats gotten off some other board.



...... .......Passing Cmp Att Pct Yds Yd/A

1. Riley Nelson 99 162 61.1 1467 9.1

2. Jake Heaps 144 252 57.1 1452 5.8



..............Rushing Att Yds Avg TD

1. JJ Di Luigi 106 546 5.2 3

2. Michael Alisa 85 455 5.4 3

3. Riley Nelson 75 376 5 1

4. Josh Quezada 77 274 3.6 1

5. Bryan Kariya 67 271 4 6



...........Receiving Rec Yds Avg TD

1. Cody Hoffman 53 821 15.5 7

2. Ross Apo 34 453 13.3 9

3. JD Falslev 29 299 10.3 2

4. JJ Di Luigi 25 259 10.4 0

5. Marcus Mathews 25 265 10.6 1



This says to me that Riley would have won more just from this analysis, and by quite a bit. Jake had tools, but not the head for it. I'm glad Riley won out in the end. We need someone who produces. His Yd/a was bout as good as Young's in his best years btw. It is amazing that Nelson had 90 less attempts than Heaps, yet had more yards.

Good post.



I would like to see SpiffCoug do an analysis comaring winn % for team with the higher YPA and see if that is a better indicator than the entire pass efficiency formula.

Yards per attempt is part of the PER formula.

The irony of this is that Heaps has the arm to go long, can throw the football 70 yds? Yet, he did not complete many long throws, except against UTEP last year. Nelson who hasn't a real strong arm gets more yards per throw than the guy with the rocket arm.



I think we all knew something was very wrong with Heaps (whatever it is), when he missed wide open receivers who were not even that deep. Austin Holt is behind the secondary, only about 25 yds down field and Heaps throws a bullet ten feet over his head.



I'm curious, in 2013, to see how Heaps does on a very bad Kansas team, that I have doubts will give him as much throwing time as he got at BYU. But maybe the year on the sidelines will make him what we all thought he was.

SpiffCoug wrote:
Yards per attempt is part of the PER formula.

I know, but it is possible for the QB on Team A to have a higher YPA than the QB on Team B, but for the QB on Team B to have a higher PER than the QB on Team A.



Maybe 90% of the time the QB with a higher YPA also has the higher PER, but lots of TD passess OR a very high completion percentage can cause a QB with a lower YPA to have a higher PER.



So, I am curious now to know if the YPA is a stronger indicator of winning than the complete PER formula.

Going back to 2003 (that's as far back as I have the box scores input on my tracker) here's what I found out.



When BYU wins the PER battle (i.e. BYU's PER is greater than their opponent) their record is: 95-17 (.848) (if you go back to 2005 the start of Bronco, BYU's PER W/L is 77-12 (.865)



When BYU wins the YPA battle (i.e. BYU's YPA is greater than their opponent) their record is: 71-41 (.634). Under Bronco the YPA W/L is 60-29 (.674)



Overall, BYU's W/L record is: 74-38 (.661) (65-24 [.730] under Bronco).



So, the data is a resounding NO. YPA is NOT a better predictor of victory of W/L than PER.



PER wins again!



This makes sense. Teams with a a good YPA will also be efficient, but the real bonus of PER is what you get from throwing TDs. YPA is only a portion of what makes a team's passing game efficient.

I think my math may have been wrong on that last post.



Since 2003, BYU's has played 112 games. In those games, their overall record is 74-38 (.661).



In those 112 games, they had a better YPA than their opponent in 71 of them. BYU's record in those 71 games was 59-12 (.831).



In those same 112 games, BYU had a better PER than their opponent in 76 of them. BYU's record in those 76 games was 66-10 (.868).



The team that had the better PER in those 112 games was 95-17 (.848).

The team that had the better YPA in those 112 games was 87-25 (.777).

It really makes a lot of sense. The PER formula works well, and so of course the YPA stat is also a good indicator of success, but not as robust as everything that goes into PER.

136 Baby - Just do it!



This is a combination of Al Davis - Just Win Baby and Nike with PER added for a special twist.



[ Post made via Mobile Device ] http://cougarcorner.com/images/mobile.png" alt="" loading="lazy" style="max-width:100%">

SpiffCoug wrote:
I think my math may have been wrong on that last post.



Since 2003, BYU's has played 112 games. In those games, their overall record is 74-38 (.661).



In those 112 games, they had a better YPA than their opponent in 71 of them. BYU's record in those 71 games was 59-12 (.831).



In those same 112 games, BYU had a better PER than their opponent in 76 of them. BYU's record in those 76 games was 66-10 (.868).



The team that had the better PER in those 112 games was 95-17 (.848).

The team that had the better YPA in those 112 games was 87-25 (.777).

Thanks, Spiff.



It is always nice to see the numbers. Very significant difference.