Heaps' Passing Efficiency: A matched sample approach

[pelagius]

What does the graph show

The blue bars show heaps' actual passing efficiency for each game this season. Heaps' performance (in term of passing efficiency) is compared to a matched sample (the red bars). The sample is constructed using non-glory years BYU quarterbacks playing in their sophmore years: Walsh (1993), Feterik (1997), Beck (2004), and Hall (2007). The sample is constructed by computing the average passing effficiency for games played by those quarterbacks that had a similar computer model ranking as the team the BYU/Heaps played against this season (I could sharpen this part up a bit but it is hard to do at this point season ... the model rankings for 2011 are pretty noisy at this point). I also adjust efficiency for home field advantage (note for what every reason, after adjusting for opponent quality the matched sample of quarterbacks actually has higher efficiency on the road).

The computer model rankings are computed each year using a regression based margin of victory model (very similar to the Sagarin predictor model). I make two adjustments; first I throw out all games played by BYU in the model. This avoids a bias relative to whether BYU won or lost the game. Second, I only rank Division 1-A teams; this keeps the number of teams ranked relatively constant.

It is worth noting what this doesn't do. It doesn't control for the supporting cast of the team. If the matched sample have far superior supporting casts then inferences are likely to be suspect.

[cougjunkie]

What does the graph show

The blue bars show heaps' actual passing efficiency for each game this season. Heaps' performance (in term of passing efficiency) is compared to a matched sample (the red bars). The sample is constructed using non-glory years BYU quarterbacks playing in their sophmore years: Walsh (1993), Feterik (1997), Beck (2004), and Hall (2007). The sample is constructed by computing the average passing effficiency for games played by those quarterbacks that had a similar computer model ranking as the team the BYU/Heaps played against this season (I could sharpen this part up a bit but it is hard to do at this point season ... the model rankings for 2011 are pretty noisy at this point). I also adjust efficiency for home field advantage (note for what every reason, after adjusting for opponent quality the matched sample of quarterbacks actually has higher efficiency on the road).

The computer model rankings are computed each year using a regression based margin of victory model (very similar to the Sagarin predictor model). I make two adjustments; first I throw out all games played by BYU in the model. This avoids a bias relative to whether BYU won or lost the game. Second, I only rank Division 1-A teams; this keeps the number of teams ranked relatively constant.

Seems like a lot of mumbo jumbo just to point out the obvious. Heaps has not been very good, especially compared to other QBs at this stage in their careers.

[byu71]

What does the graph show

The blue bars show heaps' actual passing efficiency for each game this season. Heaps' performance (in term of passing efficiency) is compared to a matched sample (the red bars). The sample is constructed using non-glory years BYU quarterbacks playing in their sophmore years: Walsh (1993), Feterik (1997), Beck (2004), and Hall (2007). The sample is constructed by computing the average passing effficiency for games played by those quarterbacks that had a similar computer model ranking as the team the BYU/Heaps played against this season (I could sharpen this part up a bit but it is hard to do at this point season ... the model rankings for 2011 are pretty noisy at this point). I also adjust efficiency for home field advantage (note for what every reason, after adjusting for opponent quality the matched sample of quarterbacks actually has higher efficiency on the road).

The computer model rankings are computed each year using a regression based margin of victory model (very similar to the Sagarin predictor model). I make two adjustments; first I throw out all games played by BYU in the model. This avoids a bias relative to whether BYU won or lost the game. Second, I only rank Division 1-A teams; this keeps the number of teams ranked relatively constant.

That is pretty dang cool.

Cool you add in an adjustment for the efficiency of the O'lines, RBacks, recieivers and OC's they had too.

[MartyFunkhouser]

Seems like a lot of mumbo jumbo just to point out the obvious. Heaps has not been very good, especially compared to other QBs at this stage in their careers.

When even cougjunkie is ready to admit his Jesus QB hasn't been very good, it speaks volumes about how pathetic his performance has been.

[pelagius]

Seems like a lot of mumbo jumbo just to point out the obvious. Heaps has not been very good, especially compared to other QBs at this stage in their careers.

That's fair. It does I control for opponent quality fairly well which might be useful to some.

[pelagius]

That is pretty dang cool.

Cool you add in an adjustment for the efficiency of the O'lines, RBacks, recieivers and OC's they had too.

This is, of course as I mention, the identifying assumption.

[MarkGrace]

If you can come up with something more interesting than pass efficiency that would be cool. You could be the next Bill James!

[pelagius]

If you can come up with something more interesting than pass efficiency that would be cool. You could be the next Bill James!

Actually, this overcomes most of your previously stated objections to passing efficiency.

[MarkGrace]

Actually, this overcomes most of your previously stated objections to passing efficiency.

Maybe I misunderstood my previously stated objections.

[TripletDaddy]

pelagiu$, what are some of the equivalent teams flowing into your model? For example, which teams are the Texas analog?

[pellegrino]

I love the title of the thread, very academic. What conference are you presenting this at?

[pelagius]

pelagiu$, what are some of the equivalent teams flowing into your model? For example, which teams are the Texas analog?

    +----------------------------------------------------+
     | year   game     opponent   scored   allowed   rank |
     |----------------------------------------------------|
  5. | 1993      5         UCLA       14        68      9 |
  6. | 1993      6   Notre Dame       20        45      6 |
 12. | 1993     12      Ohio St       21        28     11 |
 13. | 1997      1   Washington       20        42      5 |
 14. | 1997      2   Arizona St       13        10     15 |
     |----------------------------------------------------|
 22. | 2004      3          USC       10        42      1 |
 23. | 2004      4     Boise St       27        28      8 |
 30. | 2004     11         Utah       21        52      4 |
     +----------------------------------------------------+

So in the Texas case, Heaps is only compared to games played by Walsh, Feterik, and Beck. Texas has the thinnest comparison group because (at least right now) it has a very high ranking.

[pelagius]

I love the title of the thread, very academic. What conference are you presenting this at?

Probably, should have went with "Did you know Heaps sucks this bad?"

[YOhio]

Very interesting. You didn't mention any controls for OC, but did you see any trends b/w Chow, Anae and Doman?

[MarkGrace]

Maybe I misunderstood my previously stated objections.

Not really. Pelagius, I remember thinking that based on your response at the time you had read my post wrong, even though I didn't comment.

http://cougaruteforum.com/showthread...242#post664242