QB Hits - Spencer and Ware

[Cowboy Brian]

And watch like Dumerveil finish ahead of him this year. I like Dumerveil but cmon.

Im cool w/ Revis or Woodson winning it, but Im not seeing Ware even mentioned.

Last year the excuse was the overall D wasnt good enough, what about this year?

Hes not good in coverage will be their excuse.

 

[adbutcher]

54 QB Pressures is beyond amazing.

The pressure is definitely altering the way opposing offenses attack us. I would love to see how many 3 step drops we face compared to other defenses. I bet Wade is licking his chops when he get the tendency report. It is only so many route combinations from a primary 3 step drop attack. This could explain the way our defense is starting to look dominate. We are starting to finally dictate what the offense can do and I am loving it!

 

[burmafrd]

say a QB gets blindsided just after he throws the ball; he is hit but its not a pressure since he never saw it coming.

 

[theogt]

say a QB gets blindsided just after he throws the ball; he is hit but its not a pressure since he never saw it coming.

Or he could merely get hit just after he throws, but the throw wasn't rushed.

 

[percyhoward]

if there could be a formula to add up those statistics with the sack statistics.... would it show an accurate rating?... and what would it be. i agree interesting thread.

I was just thinking the same thing. You could assign 3 points for a sack, 2 for a pressure, and 1 for a hit, for example.

It seems like stats in general are either: nowhere near as complete as they could be, or made much too complicated than they need to be by having silly and often subjective elements injected into them.

theo, great thread

 

[EPL0c0]

It doesn't reflect in the sack #s, but Spencer/Ware are having one heck of a year.

I think this really says a lot about this year: it's not about individual stats, it's not about ONE guy, it's a complete and total team effort and everybody's putting in the effort for the team.

I'd rather have a team full of quality no-names than scrubs + 1 or 2 superstars

 

[theogt]

I was just thinking the same thing. You could assign 3 points for a sack, 2 for a pressure, and 1 for a hit, for example.

It seems like stats in general are either: nowhere near as complete as they could be, or made much too complicated than they need to be by having silly and often subjective elements injected into them.

theo, great thread

You could run a variance study on each of the stats and come up with a divisor that brings each group to a similar variance. That is what they did to come up with QB rating.

I'm too lazy to do that today. It's my day off.

 

[BAT]

Orakpo is neither of those lists b/c he is neither a 3-4 OLB or a 4-3 DE.

 

[percyhoward]

You could run a variance study on each of the stats and come up with a divisor that brings each group to a similar variance. That is what they did to come up with QB rating.

I can't appreciate stats for their own sake, but I'm sure what you said makes sense.

 

[theogt]

I can't appreciate stats for their own sake, but I'm sure what you said makes sense.

Here's what I did. I took each of set of statistics and ran a variance study on each on (using Excel). This tells you how each of the players vary within each set of statistics (sacks, pressures, hits). The idea is to develop a formula in order to make each set of statistics comparable. The way to do this is to divide each variable by a number such that the variance is roughly equal on each.

The numbers I came up with are 17 for sacks, 23 for hits, and 50 for pressures. Once you divide each player's total number of sacks, hits and pressures by those numbers, a variance study shows that the sets of numbers all have an equal variance of roughly .05. This also happens to be the same variance that each variable of QB rating (yards per attempt, TDs, per attempt, and INTs per attempt) tends to have if you run the study over any given season.

Now all you have to do is add the numbers up and multiple by some random number to give you a large enough number to compare more easily.

And you end up with (for 3-4 OLBs):

Name		QB Sk	QB Ht	QB Pr	QB Sk	QB Ht	QB Pr	Passrush Formula
[COLOR="Blue"][B]DeMarcus Ware	12	16	54	0.71	0.70	1.08	124.08[/B][/COLOR]
Elvis Dumervil	17	7	29	1.00	0.30	0.58	94.22
[COLOR="Blue"][B]A. Spencer	5	26	21	0.29	1.13	0.42	92.23[/B][/COLOR]
James Harrison	10	12	31	0.59	0.52	0.62	86.50
Lamarr Woodley	10	11	27	0.59	0.48	0.54	80.32
Tamba Hali	8	11	30	0.47	0.48	0.60	77.44
Clay Matthews	10	9	24	0.59	0.39	0.48	72.98
K. Wimbley	7	7	21	0.41	0.30	0.42	56.81
Manny Lawson	6	9	19	0.35	0.39	0.38	56.21
Parys Haralson	5	7	26	0.29	0.30	0.52	55.92
Shaun Phillips	7	8	16	0.41	0.35	0.32	53.98
Clark Haggans	6	7	21	0.35	0.30	0.42	53.86
Jason Taylor	8	4	19	0.47	0.17	0.38	51.23
T. Banta-Cain	7	7	15	0.41	0.30	0.30	50.81
Aaron Kampman	4	12	11	0.24	0.52	0.22	48.85
Joey Porter	8	5	11	0.47	0.22	0.22	45.40
S. Merriman	4	4	13	0.24	0.17	0.26	33.46
Robert Ayers	0	5	17	0.00	0.22	0.34	27.87
Brad Jones	4	3	9	0.24	0.13	0.18	27.29
Chike Okeafor	3	2	13	0.18	0.09	0.26	26.17
Mike Vrabel	2	2	13	0.12	0.09	0.26	23.23
Matt Roth	3	2	7	0.18	0.09	0.14	20.17
David Bowens	1	6	2	0.06	0.26	0.04	17.98
Mario Haggan	1	3	7	0.06	0.13	0.14	16.46
Derrick Harvey	0	3	6	0.00	0.13	0.12	12.52
Clint Ingram	1	2	4	0.06	0.09	0.08	11.29
Jason Trusnik	2	2	1	0.12	0.09	0.02	11.23
Brady Poppinga	1	0	2	0.06	0.00	0.04	4.94

The formula is: [(Sacks/17) + (QB Hits/23) + (QB Pressures/50)] * 50

A couple things to note. The variance studies should be done on a larger body of statistics, but this is probably pretty darn close. Also, I haven't limited the maximum value of each variable as done in the passer rating formula.

 

[theogt]

Orakpo would be a 69.86.

 

[percyhoward]

A couple things to note. The variance studies should be done on a larger body of statistics, but this is probably pretty darn close. Also, I haven't limited the maximum value of each variable as done in the passer rating formula.

Nice work, again.

(Are number of opponents' pass plays figured into this?)

This is the kind of info the pro's should be giving us, obviously. We could take these ratings, then and look at splits, such as down-and-distance, by what packages were being used, in close games/blowouts, field position, etc.

We'd know a lot more about who's doing what on the field, and there would be less speculation. It would basically be a clearer picture. "High-Def Stats."

 

[Zaxor]

nice work do you have the same info on Ratliff

 

[casmith07]

Interesting that Orakpo doesn't appear on either list.

But somehow made the Pro Bowl.

Pro Bowl =

 

[Hoofbite]

Do the Chiefs play a 3-4?

 

[Hoofbite]

But somehow made the Pro Bowl.

Pro Bowl =

Playing well and having a large rabid fanbase will do that.

 

[Hostile]

I remember when we needed to think about trading Spencer and Draft picks for Shawne Merriman to have a real bookend for Ware. Then again I remember when Merriman was better than Ware and we blew that Draft pick.

 

[NeonNinja]

I remember when we needed to think about trading Spencer and Draft picks for Shawne Merriman to have a real bookend for Ware. Then again I remember when Merriman was better than Ware and we blew that Draft pick.

:laugh2:

 

[Hostile]

:laugh2:

That was mean of me wasn't it?

 

[Eskimo]

That's interesting work you did with the variances but I just question the weighting of sacks versus hits which are pretty close to equally valued in this system.

Personally I would rank them sacks more important than pressures (since they definitely alter the play) which are more important than hits (in general).

I do think you get a more valid and reproducible quantum out of this measurement than you do out of the simple sacks number. I wonder if this was run over the last several years if we saw this QB pressuring stat you have created is more stable (less variability) than sacks. One thing that you could do if you haven't already is normalize everything per game played if you so desired so that it was more about rates than seasonal statistics - this would likely make the construct more stable as well.