Passing Yards and Receiving Yards

[TheSix]

Hey guys, just wondering on how some of you guys calculate edges on passing and receiving yards for football props?

[bangminton]

Six, there may be different ways to calculate an edge for Passing Yards and there are a number of variables for NFL so I don't believe there is a foolproof approach like ProLine or PointSpread but here is something that I have been using for a number of years, with a solid rate of return.

It is not much different than finding an edge for sacks, except for the Poisson part. You should be able to find all of the stats below from NFL.com

1. Look at the QB stats and tabulate his passing yards for each game. We are in only Week 4 of the 2015 season so you may want to use 2014 or even 2013 numbers for a bigger sample size. You can start using this year number as the season progress and there is more data for analysis.

2. Find the median of those passing yards. Don't just rely on the average because there can and will be outliers.

3. Find the average passing yards the opposing team is allowing per game. Once again, use previous year's stats if you don't have enough data from the current year.

4. Find the league average for passing yards allowed.

5. Expected QB passing yard = Value you find in #2 * (Value you find in #3 / Value you find in #4)

6. Compare against what OLG offers and correlate that with the WRs/TEs

Don't forget to check the Pointspread and the game total from Pinny. If we were to believe Pinny's line is efficient, the big favorite may not pass as much as the big underdog, although this is certainly not a given.

You may also want to check the injury report as a situational angle.

[richhhh]

Problem with above method is it does not provide a win% needed to quantify a edge. It spits out a predicted average which most time's will be similar to the line you are handicapping.

[bangminton]

I find the books, including OLG, use the average for #2 a lot of the times, when the median is more accurate.

I do have a way to quantify the edge from my method but since nobody replied to TheSix for over a week, I thought I would chime in.

If you think your approach is better, feel free to share the love, rich.

[richhhh]

It was a good post bang and without a doubt will make others sharper.. Even without a strong estimate of probability one could likey be profitable depending how off olg lines are and how strong the correlation's are.

[bangminton]

If you see rooms for improvement, I am all ears, rich. This old dog never shies away from learning new tricks.

Curious to hear what approach others are taking. You and I can't surely be the only ones calculating Props edges.

[TheSix]

Hey Bang, at point should I be only looking at this years stats?

[bangminton]

Hey Bang, at point should I be only looking at this years stats?

The short answer is Yes and the long answer is it depends on the type of Props, Six.

7 week of stats is definitely not the ideal sample size that we need to run any kind of effective model. Having said that, a 16-week regular NFL season schedule (I am not including the playoff) will always be a point of contention because the sample size is nowhere near that of baseball, hockey or basketball.

We have to make do with what we got. You probably know this already but we also need to make proper adjustment along the way. The Peyton Manning of 2015 is, simply put, not the same Peyton Manning of 2012/2013 when he first joined the Broncos. Father Time will eventually catches up with us all

[TheSix]

Thanks Bang