The Regression of Patriots Special Teams in 2018:

[maineman209]

Just like Lewis, Cooks, and Amendola were replaced by inferior players.

I guess it always bears repeating that the Pats are constrained by an externally imposed budget and head count limit that need to be carefully managed in coordination, and with consideration for future "indebtedness" and future ROI. Which is just a fancy way of saying that the team simply can't afford to keep and pay everyone's favorite players ad infinitum.

I'm disappointed that the Pats couldn't keep DLew and Danny, but I accept the reality of the circumstances as I do with Cooks' departure. WRT their "replacements being inferior", I can't even comprehend how that statement works at all. DLew, Danny and Cooks were/are unique in how they played and in what their roles were, so I don't see how there's any one-to-one correspondence with any new Pats players and their roles on which to base any superior/inferior judgement to begin with.

I suppose the closest one might get to a one-to-one comp is to assume that Hill and Michel "replaced" DLew. But neither of them even come close to being the same "type". Neither has the same running style (both are more about power than elusiveness), and neither has the receiving chops. Nevertheless, Michel is producing more rushing yards per game, when available. And, of course, we'll probably never know how much Hill may have contributed, since he went on IR almost immediately after the season began.

Yeah, yeah ... "Injury prone! Should have known!" Seriously, anyone who think that the Pats should have foreseen Develin rolling up on Hill's leg is ascribing god-like prescience to them that no human being has ever possessed and then trashing them for not actually posessing it. This is probably the ultimate straw-man argument.

Burkhead probably comes closest to DLew in overall skill set, but he's still more power-oriented than elusive. No, he wasn't as productive last season as DLew, but then, he was behind DLew. Of course, Burkead has also been injured missed 21 games so far in his first two seasons. But then, DLew also missed 21 games (regular & post-season) in his first two seasons with the Pats.

It's much more difficult for me to discern any one-to-one replacement correspondence for Cooks and Danny.

It sometimes seems like the Pats are using Gordon in Cooks' 2017 "deep threat" role, although, again, two different types of players. There's a parallel there to Cooks taking over the "deep threat" role that Hogan performed (very well) in 2016. As far as I can see, their only point of overlap is in how they've been deployed a lot of the time (though not all of the time). I don't know how to make any superior/inferior judgement from that especially considering that Gordon probably hasn't had enough time (in seven games played) to have run all the route combos or experienced all the sight-read adjustments more than once under game conditions.

WRT, Danny ... I suppose the closest in 2018 in terms of the types/depths of routes he's been running so far this season is Dorsett. But that's a pretty small sample size, and, again, pretty much their only point of overlap. I mean, folks can criticize Dorsett for not having Danny's "clutchness" (as if anyone ever will), but has Dorsett really even been tested yet? How can we know until he has been? Meanwhile, Dorsett has been catching 83% of everything thrown his way this season - with 14 1st-downs and 2 TDs in 25 catches - so it seems to me that he might have the potential for some degree of "clutchness". We just need to wait and see. But then, patience has never been considered a virtue on this forum, has it?

 

[Tony2046]

Huh?
How do you offset a 97 yard ko return with one good play with regard to average per kick return?
Aside from 3 poor returns the kr average is 22. That’s not great but decent. So they have been decent aside from a couple of breakdowns. Doesn’t mean the breakdowns aren’t important, but it provides context.

IF you stifle an offense all day and win 40-7 and they ran 19 times for 35 yards before breaking off a garbage time 95 yard td run, you end allowing 6.5 yards per rush. To say he run defense sucked all day because of that stat is wrong. Context.

Because the average is relative to the league or all 32 teams and not just the Patriots. So you would have to say that no other team had a few bad plays in order to say the Patriots are better than their average dictates or subtract the three poor returns from every other team and see where the Pats rank.

 

[Tony2046]

@AndyJohnson

Question for you or anyone. Which year would you use as a comparison when comparing special team's performance? 2016? I figure if the goal is SB championship 2016 would be a good comparison.

I'm not finished putting things together but my argument is based on how the offense and special teams have put the defense in a tough position this season. Squib kicks, long KR, muffed punts, Fumbles, Interceptions etc..

So from a defensive prospective:

1. They have defended 117 drives.
2. 68 of those drives started at the opponents 25 yard line or less and they gave up a total of 94 points or 1.4 points per drive.
3. 23 of those drives started at the opponents 40 yard line or more and they gave up a total of 109 points or 4.7 points per drive.

Of those 23 drives the ball was obtained by the opponent via:

4 Kickoffs = 24 points
7 Punts = 34 points
3 Missed Field Goals = 7 points
1 Muffed Punt = 7 points
3 Failed 4th down attempts = 13 points
4 Fumbles = 21
1 Interception = 3

 

[Ring 6]

Because the average is relative to the league or all 32 teams and not just the Patriots. So you would have to say that no other team had a few bad plays in order to say the Patriots are better than their average dictates or subtract the three poor returns from every other team and see where the Pats rank.

I am not arguing good or bad I am arguing that context matters.
A team that allows 28 yards per return with almost all being between 25 and 30 is very different than a team that allows 22 per return with almost all being 20-25 but had one long return skew the results.
It’s the difference between starting in bad shape once over the course of 10 games and starting 6 yards in the hole every single time.

Saying “it’s the average” without considering context is short handing the thinking.

 

[Ring 6]

@AndyJohnson

Question for you or anyone. Which year would you use as a comparison when comparing special team's performance? 2016? I figure if the goal is SB championship 2016 would be a good comparison.

I'm not finished putting things together but my argument is based on how the offense and special teams have put the defense in a tough position this season. Squib kicks, long KR, muffed punts, Fumbles, Interceptions etc..

So from a defensive prospective:

1. They have defended 117 drives.
2. 68 of those drives started at the opponents 25 yard line or less and they gave up a total of 94 points or 1.4 points per drive.
3. 23 of those drives started at the opponents 40 yard line or more and they gave up a total of 109 points or 4.7 points per drive.

Of those 23 drives the ball was obtained by the opponent via:

4 Kickoffs = 24 points
7 Punts = 34 points
3 Missed Field Goals = 7 points
1 Muffed Punt = 7 points
3 Failed 4th down attempts = 13 points
4 Fumbles = 21
1 Interception = 3

Or:

The 2018 team SO FAR has gone on defense on average at the 30.7 yard line, 27th in the league. They have allowed 2:34 top (10) 5.8 plays (19) 32.2 yards (18) 2.0 points (17) per drive.

In 2016, when we had the #1 defense in the nfl
It was start at 25 (1st) top 2:34 (8) plays 5.7 (16) yards 29 (10) points 1.41 (2)

So we are starring ON AVERAGE 5 yards worse allowing 3.2 more yards and .59 more points per drive.

The problem is it’s an average so way more things than field position come into play. 8.2 yards per drive and .59 points per drive is basically a td a game. Is that because of 5 yard of field position on average? Is it 5 each drive which seems like something tha should be easy to overcome or is it 60 on one drive which creates the extra TD? Is the 3.2 yards allowed, about 40 a game the difference between 2 red zone stops and fgs and 2 red zone scores? That has nothing to do with special teams.
Bottom line stats may make you think but to use cumulative football statistics over the course of a season without context or situational awareness is just bad analysis and any conclusion drawn from it is specious at best.

 

[Hyped]

What is the MEDIAN return allowed??? That is the question here, IMO.

 

[QuantumMechanic]

What is the MEDIAN return allowed??? That is the question here, IMO.

There are so, so many NFL stats that would be better and more informative if the median were used instead of the mean. RB YPC is the primary one. And this is another good one.

 

[maineman209]

There are so, so many NFL stats that would be better and more informative if the median were used instead of the mean. RB YPC is the primary one. And this is another good one.

A measure of how far above or below the median a player or a team is in a particular statistical category would be much more informative than a superficial 1-32 ranking.

Stuff like that doesn't sell newspapers, though.

 

[Ring 6]

There are so, so many NFL stats that would be better and more informative if the median were used instead of the mean. RB YPC is the primary one. And this is another good one.

Actually there are pretty much no good nfl stats if you expect them to equal a complete judgment.
Meduan is no better than mean because it also leaves out too much info.
If 2 RBs have the same median but one does substantially better on the carries above median, which involves an important skill set, median tells you half the story.

Baseball is a great sport to use cumulative stats to judge players and teams because there are 162 games and situational differences normalize.
People keep trying to analyze football with statistics and they keep creating new statistics. Thats just because there are no statistics that tell the story the way people who use them want them to.

 

[Tony2046]

I am not arguing good or bad I am arguing that context matters.
A team that allows 28 yards per return with almost all being between 25 and 30 is very different than a team that allows 22 per return with almost all being 20-25 but had one long return skew the results.
It’s the difference between starting in bad shape once over the course of 10 games and starting 6 yards in the hole every single time.

Saying “it’s the average” without considering context is short handing the thinking.

Ok sure if we're talking about the ranking but to be ranked last or almost last on PR and KR is not because of two or three bad plays. It's definitely a red flag. That would be one hell of a "context" to say we should be ranked in the top ten because of these three plays.

I've rewatched those plays because I've been interested in the effects of the ST and the offensive play on the defense in a complementary sense for about the last four weeks or so. Back two or three weeks ago I argued that the defense would be good enough to win a championship if the "elite" offense and the ST's could stop putting the defense in such ridiculous field positions.

Imo, the Special teams have hurt this team as a whole this season. Some of those bad plays were because of lack of execution and some because of game planning. I think they should be much more conservative on the Special team play calling. Kick it out of the end zone and stop squib kicking the ball hoping for a turnover. Focus on hang time rather than trying to pin the kick returner behind the five yard line.

All the defensive numbers improve dramatically when the opposing team starts on their 25 or less. Turnovers, points, 3 and outs, etc...

The statistical charts I posted simply support what I'm seeing on film. I'm not using statistics as an argument rather to support it.

 

[Ring 6]

Ok sure if we're talking about the ranking but to be ranked last or almost last on PR and KR is not because of two or three bad plays. It's definitely a red flag. That would be one hell of a "context" to say we should be ranked in the top ten because of these three plays.

I've rewatched those plays because I've been interested in the effects of the ST and the offensive play on the defense in a complementary sense for about the last four weeks or so. Back two or three weeks ago I argued that the defense would be good enough to win a championship if the "elite" offense and the ST's could stop putting the defense in such ridiculous field positions.

Imo, the Special teams have hurt this team as a whole this season. Some of those bad plays were because of lack of execution and some because of game planning. I think they should be much more conservative on the Special team play calling. Kick it out of the end zone and stop squib kicking the ball hoping for a turnover. Focus on hang time rather than trying to pin the kick returner behind the five yard line.

All the defensive numbers improve dramatically when the opposing team starts on their 25 or less. Turnovers, points, 3 and outs, etc...

The statistical charts I posted simply support what I'm seeing on film. I'm not using statistics as an argument rather to support it.

But if you are giving up 22 yards a kick return other than 2 screw ups that puts them at about the 25 anyway. The question is how many screw ups will you have vs how many times you pin them deep (including due to block in the back penalty) or get a turnover?
That’s a different argument than assuming they start at the 30 every time you kickoff. That’s not what the stat means.
The issue is to fix what caused 2 long returns not to give up on doing something that has created a big advantage. That’s part of the issue with stats. Bill Belichick isn’t going to coach his team by saying we allow 28 yards per kick when we kick it to about the 2 so that’s giving them 5 more yards than a touchback so give up and kick it through the end zone.
He is looking at something that 25 out of 27 times worked at least marginally well, so fix what happened the 2 times it didn’t.
Coaching, unlike this board would lead you to believe, isn’t about putting together a plan, trying it then reading a stat sheet to decide if it was a good idea. It’s about having a plan and working consistently to improve execution of it.
When we get to the post season and after a game tying it go ahead TD we kick it to the 2, and force a fumble or get a penalty and they start at the 9 and it changes the game, a couple of regular season long returns were an acceptable price to pay.
The issue is coaching it up not scrapping it, and with the long running history of quality special teams play here it would be silly to doubt it will get fixed.

 

[jah]

Actually there are pretty much no good nfl stats if you expect them to equal a complete judgment.
Meduan is no better than mean because it also leaves out too much info.
If 2 RBs have the same median but one does substantially better on the carries above median, which involves an important skill set, median tells you half the story.

This is a good point even though I agree with the others that median would be good to see. Nick Chubb is the perfect example of the kind of RB whose skillset is sold short by median yards.

 

[Tony2046]

But if you are giving up 22 yards a kick return other than 2 screw ups that puts them at about the 25 anyway. The question is how many screw ups will you have vs how many times you pin them deep (including due to block in the back penalty) or get a turnover?
That’s a different argument than assuming they start at the 30 every time you kickoff. That’s not what the stat means.
The issue is to fix what caused 2 long returns not to give up on doing something that has created a big advantage. That’s part of the issue with stats. Bill Belichick isn’t going to coach his team by saying we allow 28 yards per kick when we kick it to about the 2 so that’s giving them 5 more yards than a touchback so give up and kick it through the end zone.
He is looking at something that 25 out of 27 times worked at least marginally well, so fix what happened the 2 times it didn’t.
Coaching, unlike this board would lead you to believe, isn’t about putting together a plan, trying it then reading a stat sheet to decide if it was a good idea. It’s about having a plan and working consistently to improve execution of it.
When we get to the post season and after a game tying it go ahead TD we kick it to the 2, and force a fumble or get a penalty and they start at the 9 and it changes the game, a couple of regular season long returns were an acceptable price to pay.
The issue is coaching it up not scrapping it, and with the long running history of quality special teams play here it would be silly to doubt it will get fixed.

Good points Andy. I'll break it down further later on.

 

[Ring 6]

Good points Andy. I'll break it down further later on.

I think a good breakdown if you are inclined to do so would be starting position after kickoffs this year be last.
In other words
# of kickoffs
# of touchbacks
# inside 10
# 11-15 and so forth.

 

[Hyped]

This is a good point even though I agree with the others that median would be good to see. Nick Chubb is the perfect example of the kind of RB whose skillset is sold short by median yards.

Like @AndyJohnson said...median only gives half of the picture...

I am also wary of all the stats in football and think most of them are useless...but I'd still like to see that half of the picture. It does give some context even if we ignore above/below median return yardage.

 

[Ring 6]

Like @AndyJohnson said...median only gives half of the picture...

I am also wary of all the stats in football and think most of them are useless...but I'd still like to see that half of the picture. It does give some context even if we ignore above/below median return yardage.

We are programmed to think statistics are the only objective data, but in football statistics just aren’t very useful, and their usefulness is very narrow but unfortunately they are used broadly all the time.

 

[IllegalContact]

$hitty talent down the end of the bench..........can't fix stupid

 

[DaBronxPats14]

Maybe its kick off new rule formation. Before KO Team had a head start to run down field. Ghost could send it to a corner where Returner was boxed in for short ydg at or before 25 yd line. Looks like Returners are already at 20 yd line before any Pat is near him.

 

[Hyped]

We are programmed to think statistics are the only objective data, but in football statistics just aren’t very useful, and their usefulness is very narrow but unfortunately they are used broadly all the time.

And football has many variables that other sports don't (e.g. a QB is evaluated on how well his OL blocks, his receivers catch the ball, and etc...whereas baseball...it's only the hitter/pitcher and in the NBA it's PER).

 

[maineman209]

Like @AndyJohnson said...median only gives half of the picture...

I am also wary of all the stats in football and think most of them are useless...but I'd still like to see that half of the picture. It does give some context even if we ignore above/below median return yardage.

Comparing mean and median can be useful. If the mean is significantly above or significantly below the median, it's obvious that further context and analysis is required. It's the "Bill Gates walks into a blue-collar tavern" thing wrt to annual income.

With certain rankings (e.g., 3rd-down conversion percentage), calculating the median value can provide some perspective quickly, as can calculating standard deviations.