Kerry Byrne:Stats write different story for ‘D’

[SVN]

http://www.bostonherald.com/sports/...ots/2015/01/stats_write_different_story_for_d

 

[SHOWTIME15]

CHFF has cleveland ranked number 1 in DPR...but for some reason i don't think they'll win the super bowl this year.

http://www.coldhardfootballfacts.com/stats/2014/Final/DPR/

 

[RobertWeathers]

Good info. Makes sense. I don't think Kerry took ino account how different (and more difficult to defend) the passing game is now when compared to the 01-04 teams...

 

[Sicilian]

I'm sorry I wasted a click on that. Every year, the "measuring stick" for a good defense seems to change depending on which stat makes the Pats look worst. Pats don't give up many points? Doesn't matter, look at all the yards. Don't give up many points and shore up the yardage numbers? Doesn't matter, DPR is REALLY where it's at.

 

[SVN]

I'm sorry I wasted a click on that. Every year, the "measuring stick" for a good defense seems to change depending on which stat makes the Pats look worst. Pats don't give up many points? Doesn't matter, look at all the yards. Don't give up many points and shore up the yardage numbers? Doesn't matter, DPR is REALLY where it's at.

I dont think its that . Every year the measuring stick changes depending on who wins the SB

 

[Oswlek]

Kerry Byrne is good at collecting data but lousy at analyzing it. He's also a plagiarist, but I'll let that slide for now.

More to the point, check out this three game run of pass ratings by opposing QBs:

Chicago: 108.6
Denver: 80.9
Indy: 90

Yes, Kerry, NE's defense was horrible during that stretch. Clearly if they try to bring that unit into the playoffs they'll be torched, right?

 

[PatsBoy12]

I'm sorry I wasted a click on that. Every year, the "measuring stick" for a good defense seems to change depending on which stat makes the Pats look worst. Pats don't give up many points? Doesn't matter, look at all the yards. Don't give up many points and shore up the yardage numbers? Doesn't matter, DPR is REALLY where it's at.

I wasted my click as well. I should have known just based on the title of the article not to click, but I still did. Whatever.

 

[ctpatsfan77]

Haven't clicked, but all I feel I need to say is this: even a merely good D playing bad Os won't hold opponents to just 1 TD and 5 FGs in the second half in a seven-game span.

 

[Oswlek]

Haven't clicked, but all I feel I need to say is this: even a merely good D playing bad Os won't hold opponents to just 1 TD and 5 FGs in the second half in a seven-game span.

Ha! I had to read this like seven times to figure out you weren't talking to me. Doh!

 

[SalemPats]

http://www.bostonherald.com/sports/...ots/2015/01/stats_write_different_story_for_d

This is where 'analytics' are stupid. Our defense got better every week. You have too look at the schedule in terms of competition and how the games went. How many garbage yards (like Cutler and Manning in the 2nd half when we were up by over 20). We Completely shut down Stafford and Rivers. You need to watch the tape and understand how the games went.

Week 8 - 15 we played against very good offenses and went 6-1 while shutting down every offense (but the Packers in the 1st half)

8 Oct 26 CHI 23 @ NE 51 Final 68,756 Tom Brady
354 Yds (NE) Matt Forte
114 Yds (CHI) Rob Gronkowski
149 Yds (NE)
9 Nov 02 DEN 21 @ NE 43 Final 68,756 Peyton Manning
438 Yds (DEN) Jonas Gray
33 Yds (NE) Emmanuel Sanders
151 Yds (DEN)
10 Bye
11 Nov 16 NE 42 @ IND 20 Final 66,751 Andrew Luck
303 Yds (IND) Jonas Gray
201 Yds (NE) Coby Fleener
144 Yds (IND)
12 Nov 23 DET 9 @ NE 34 Final 68,756 Tom Brady
349 Yds (NE) LeGarrette Blount
78 Yds (NE) Brandon LaFell
98 Yds (NE)
13 Nov 30 NE 21 @ GB 26 Final 78,431 Aaron Rodgers
368 Yds (GB) Eddie Lacy
98 Yds (GB) Davante Adams
121 Yds (GB)
14 Dec 07 NE 23 @ SD 14 Final 68,815 Tom Brady
317 Yds (NE) LeGarrette Blount
66 Yds (NE) Julian Edelman
141 Yds (NE)
15 Dec 14 MIA 13 @ NE 41 Final 68,756 Ryan Tannehill
346 Yds (MIA) Jonas Gray
62 Yds (NE) Mike Wallace

 

[IllegalContact]

over the 2nd half of the season, only the seahawks gave up fewer point........(104 to 138) however, over that span, the pats played 3 of the top 6 offenses (denver, indy, green bay).....over that same time, the seahawks faced no such quality (the best 2 offenses were a mark sanchez eagles and a eli manning giants....the rest included 5 games against bottom 10 offenses)

the defense is going to make flacco look bad.

 

[patsfan-1982]

here is the way i see this defense it is far from a top 5 defense but they have the talent at all levels of the defense to get a stop with the game on the line something this team has not had since 2007 so as long as the offense can get the lead late they got a shot to stop anyone and that all you can ask for in the playoffs

 

[ctpatsfan77]

here is the way i see this defense it is far from a top 5 defense but they have the talent at all levels of the defense to get a stop with the game on the line something this team has not had since 2007 so as long as the offense can get the lead late they got a shot to stop anyone and that all you can ask for in the playoffs

Really? Who would you put above them?

[Seriously, if this isn't a top 5 defense, it's damn close.]

 

[Oswlek]

here is the way i see this defense it is far from a top 5 defense

It is? By nearly every metric they are in or pretty close to the top 10, and if you are in the top 10 that means you are close to being top 5.

FWIW, as of week 15 (Miami game), NE had the second best defense by DVOA since week 10. Very small sample, I know, but it is still encouraging, particularly since the teams faced over this time were Indy, Det, GB, SD and Miami, with three coming on the road.

They struggled a little against NY, but not so much that they would fall out of out of top 5, particularly since their second best rating was more than double the #10's efficiency.

It could be countered that NE's DVOA for the first 10 weeks was 20th, but their games against Miami and KC - clearly non-predictive at this point - drag those numbers down a great deal. DVOA also has a hard time with NE's defense when holding a lead. They actually gave Chicago and Denver positive offensive ratings despite halftime deficits of 38-7 and 27-7, respectively, with neither team ever getting closer than 13 points afterward.

Yes, in two paragraphs I just used DVOA to prove NE is an elite defense and then demonstrated why DVOA is faulty. That's just how I roll.

 

[FirstAndGoal]

Doesn't this guy write about food?

 

[patsfanincleveland]

It is? By nearly every metric they are in or pretty close to the top 10, and if you are in the top 10 that means you are close to being top 5.

FWIW, as of week 15 (Miami game), NE had the second best defense by DVOA since week 10. Very small sample, I know, but it is still encouraging, particularly since the teams faced over this time were Indy, Det, GB, SD and Miami, with three coming on the road.

They struggled a little against NY, but not so much that they would fall out of out of top 5, particularly since their second best rating was more than double the #10's efficiency.

It could be countered that NE's DVOA for the first 10 weeks was 20th, but their games against Miami and KC - clearly non-predictive at this point - drag those numbers down a great deal. DVOA also has a hard time with NE's defense when holding a lead. They actually gave Chicago and Denver positive offensive ratings despite halftime deficits of 38-7 and 27-7, respectively, with neither team ever getting closer than 13 points afterward.

Yes, in two paragraphs I just used DVOA to prove NE is an elite defense and then demonstrated why DVOA is faulty. That's just how I roll.

The reality is if someone isn't trying to quantify situation football, the stats are used to skew the debate to "prove" what someone all ready thinks.

That's why people think it's a big deal that Vince Young throws for 400 yards down 40 points.

 

[BradyManny]

Data's only as useful as the analysis performed on it...not sure I agree with what Byrne takes from the data. I agree the stats he cites as typically meaningful are generally pretty meaningful over the long run. And that's exactly what he does - is cite how its useful over the long run, he looks at the entire Super Bowl era.

In one 16 game season, one or two games is enough to swing those numbers, and the fact that we're in the top 5 tells me we're in the conversation for teams that should win the Super Bowl. Yeah, maybe Seattle's a little bit better. That's not news to anyone. We've known for a while that it was going to take 3 tough games to get us a title, with the 3rd of those being by far the most difficult.

The analysis also ignores special teams, where we are strong. The point being we're a team that can win in all 3 phases this year, and that really increases your chance of winning in the post-season.

 

[TheBostonStraggler]

It has the Patriots ranked tenth, not bad at all. Patriots played 5 of the 9 teams above them in rank and went 4-1. We're 8th in comp % (when is the last time we were that high in comp%???).

Does it measure additional things that I believe are important to a precise metric?
(A) meaningless stats/points -- I don't know what other teams had, maybe similar, but the Patriots had plenty of completions/yards on them when the game was over - where the D was leaning prevent (garbage time)
(B) Passing defense/stats on your own side of the field most especially in the red zone (a very under valued stat IMHO).
(C) strength of opponents faced


I'd be curious if these are factored in where the Patriots are.
I'd also like to see these comprehensive D stats taking only for Oct-Dec. The Patriots D was definitely not peaking in Sept yet played pretty consistently since then (IMHO Oct-Dec is a significant enough sample to make a determination of the D even though it is discarding September).

Even if the added stats doesn't end up improving the Patriots D metric, again this rank has them at #10. #10 out of 32 is not a bad thing overall. And the difference between remaining playoff teams (4 of them) ranked better than the Patriots is a spread of about 3.5 or less. That isn't exactly frightful considering the point CHFF made -- which is past winners had a dominant ranking -- doesn't seem to apply to a difference of 3.5 points or less.

 

[NSPF]

Gotta love a good satire piece!