How the Dolphins Can Turn the 2013 Season Around

[Shouright]

One of the most strongly correlated statistics with winning percentage in the NFL is the difference between a team's adjusted net yards per pass attempt (ANY/A) on offense and its adjusted net yards per pass attempt (ANY/A) on defense. Historically this difference in ANY/A has had a 0.84 correlation with win percentage in the NFL, which is extremely strong.

Here's an article that says more about ANY/A, including its relationship with Super Bowl teams over the years:

http://nflminds.blogspot.com/2011/06/adjusted-net-yards-per-passing-attempt_20.html

Here is the 2013 league data with regard to ANY/A:

TEAM	D ANY/A	O ANY/A	ANY/A DIFF	WIN %
Houston Texans	5.8	4.8	-1	0.286
Pittsburgh Steelers	5.5	5.6	0.1	0.286
Seattle Seahawks	3.4	7	3.6	0.875
Tennessee Titans	5.6	5.8	0.2	0.429
New Orleans Saints	4.9	8.1	3.2	0.857
Carolina Panthers	4.6	6.3	1.7	0.571
Indianapolis Colts	5.5	6.5	1	0.714
Kansas City Chiefs	4.2	5.4	1.2	1.000
Baltimore Ravens	6.7	5.5	-1.2	0.429
Oakland Raiders	6.1	4.6	-1.5	0.429
Tampa Bay Buccaneers	6.4	4.5	-1.9	0.000
New England Patriots	4.6	4.8	0.2	0.750
Green Bay Packers	6.9	8.3	1.4	0.714
Cleveland Browns	5.1	4.8	-0.3	0.375
Atlanta Falcons	7.4	6.5	-0.9	0.286
San Francisco 49ers	5.1	6.9	1.8	0.750
St. Louis Rams	7.2	5.5	-1.7	0.375
Jacksonville Jaguars	8	3.9	-4.1	0.000
New York Jets	6.4	4.3	-2.1	0.500
San Diego Chargers	7.5	8.2	0.7	0.571
Washington Redskins	7.2	5.3	-1.9	0.286
Chicago Bears	7.1	6.8	-0.3	0.571
New York Giants	5.7	4.7	-1	0.250
Cincinnati Bengals	5.2	6.6	1.4	0.667
Minnesota Vikings	6.8	4.4	-2.4	0.143
Miami Dolphins	5.1	4.8	-0.3	0.500
Arizona Cardinals	5.2	4.4	-0.8	0.500
Buffalo Bills	5.9	4.8	-1.1	0.375
Detroit Lions	6.3	7.5	1.2	0.625
Denver Broncos	6	9.2	3.2	0.875
Philadelphia Eagles	6.4	6.3	-0.1	0.375
Dallas Cowboys	7.2	6.3	-0.9	0.500
AVERAGE	5.97	5.89	-0.08	0.50
STANDARD DEVIATION	1.08	1.36	1.74	0.25
DOLPHINS Z-SCORE	-0.81	-0.80	-0.13	0.02
CORRELATION WIN %	-0.57	0.64	0.85

As we can see above, the Dolphins' offensive ANY/A (4.8) through the first half of the season has been almost a standard deviation below the league average (5.89). Its defensive ANY/A (5.1) has been almost a standard deviation better than the league average (5.97), meaning of course that they have surrendered fewer ANY/A than the average team in the league. Its ANY/A differential (-0.3) is only slightly better than the league average (-0.08), which stands to reason given the strong correlation between ANY/A and win percentage. The average win percentage in the NFL is about .500, and the average ANY/A differential is about zero.

Let's take a look at how the Dolphins have fared so far this year in this regard:

	DOLPHINS ANY/A	OPPONENT ANY/A	ANY/A DIFF	W/L
BROWNS	5.48	2.19	3.29	W
COLTS	8.56	6.30	2.26	W
FALCONS	4.75	5.95	-1.20	W
SAINTS	2.59	11.63	-9.04	L
RAVENS	6.35	6.06	0.29	L
BILLS	4.15	3.69	0.46	L
PATRIOTS	1.98	3.00	-1.02	L
BENGALS	6.06	2.88	3.18	W

Reading between the lines in the above data, I think we find there are two major ways that the Dolphins can turn their season around this year:

1) The league-wide correlation between pass attempts and ANY/A is nil (-0.02). However, the correlation between pass attempts and ANY/A for Ryan Tannehill is much greater (-0.40). What that means of course is that the fewer passes Ryan Tannehill throws, the better the Dolphins' offensive ANY/A, whereas for NFL quarterbacks league-wide, there is no such relationship between pass attempts and ANY/A.

This stands to reason, given that Ryan Tannehill is currently a developmental quarterback. Prior to the Cincinnati game, Ryan Tannehill attempted passes on more than 50% of the Dolphins' offensive plays. Finally, against Cincinnati last Thursday, as many people here had been suggesting, the Dolphins scaled back Tannehill's passing load, and he attempted passes on only 46% of the Dolphins' offensive plays. Instead, the run game was emphasized more than usual.

2) In the table above, notice the correlation between Cameron Wake's presence and the Dolphins' opponents' ANY/A. Wake was hurt during the Indy game, missed the following three games, played sparingly against Buffalo, and played increasingly more against New England and Cincinnati, culminating in the sack for a safety to win the Bengals game Thursday.

The correlation between the Dolphins' sacks of the opponent and their ANY/A in the games above is -0.61, meaning that the more sacks the Dolphins get, the lower their opponents' ANY/A. Likewise, the correlation between INTs thrown by the opponents above and their ANY/A is -0.75, meaning again that the more INTs the Dolphins get, the lower their opponents' ANY/A. Both of these figures (sacks and INTs) have been appreciably higher in the games this year in which Cameron Wake has not been sidelined or limited by injury, which of course stands to reason, given that Wake is one of the league's best at sacking the quarterback, and quarterback pressures often cause interceptions.

So, the take-home message here in my opinion is that the Dolphins have a recipe for winning the rest of the season, and it consists of:

1) Continuing to emphasize the running game on offense,
2) Continuing to limit Ryan Tannehill's overall passing load, and
3) Continuing to make plays defensively, spearheaded by Cameron Wake.

 

[bryan]

Translation.....teams with their quarterback playing well are likely to win.

 

[Shouright]

Translation.....teams with their quarterback playing well are likely to win.

:confused2:

 

[kcbrown]

1) Continuing to emphasize the running game on offense,
2) Continuing to limit Ryan Tannehill's overall passing load, and
3) Continuing to make plays defensively, spearheaded by Cameron Wake.
4) SCORE 24+ points per game.

 

[Shouright]

1) Continuing to emphasize the running game on offense,
2) Continuing to limit Ryan Tannehill's overall passing load, and
3) Continuing to make plays defensively, spearheaded by Cameron Wake.
4) SCORE 24+ points per game.

Interestingly, points scored is more weakly correlated with wins (0.73) than ANY/A differential (0.84).

 

[Awsi Dooger]

Note that Carolina has a nice net stats, same as last year. They continue to underachieve in the won/loss column but it's not nearly as dramatic as last year, when they were the only team deep into the season that had very good passing stats combined with a terrible record. Consequently I won lots of bets on them late in the season, after mentioning here that they had very good relevant stats. There really isn't a team this season that looks like an obvious team to back late in the year, based on underachieving early.

I'm not as loyal to adjusted stats as base numbers because there's too much weight to interceptions, which can be somewhat random. Fortunately I found some complimentary categories, like points per pass attempt.

For a team on the Dolphins' level it still should be, "run the ball often, pass the ball well." But I don't believe Philbin differentiates between a moderate quarterback and an elite one, in terms of how the offense needs to vary depending on which type you have. Heck, many posters here are still in denial. One guy posted "sigh" not long ago, when more rushes were suggested.

 

[Shouright]

Note that Carolina has a nice net stats, same as last year. They continue to underachieve in the won/loss column but it's not nearly as dramatic as last year, when they were the only team deep into the season that had very good passing stats combined with a terrible record. Consequently I won lots of bets on them late in the season, after mentioning here that they had very good relevant stats. There really isn't a team this season that looks like an obvious team to back late in the year, based on underachieving early.

I think you could probably do the converse, though, and bet against the Chiefs.

I'm not as loyal to adjusted stats as base numbers because there's too much weight to interceptions, which can be somewhat random. Fortunately I found some complimentary categories, like points per pass attempt.

Net yards per pass attempt is more stable, and its differential is also correlated strongly with wins (0.79).

For a team on the Dolphins' level it still should be, "run the ball often, pass the ball well." But I don't believe Philbin differentiates between a moderate quarterback and an elite one, in terms of how the offense needs to vary depending on which type you have. Heck, many posters here are still in denial. One guy posted "sigh" not long ago, when more rushes were suggested.

No doubt. And we blew two games (New England and Buffalo) not realizing it sooner.

 

[Spiff]

[...]
1) Continuing to emphasize the running game on offense,
2) Continuing to limit Ryan Tannehill's overall passing load, and
3) Continuing to make plays defensively, spearheaded by Cameron Wake.

Without looking at the data I could have told you that teams that run the ball more and throw less and teams that make plays on defense stand a great chance of winning football games.

I am not saying that running the ball more equals more wins, rather it's the other way round: when you have a lead you tend to run more than you pass. That's why the stats tell you that the more you run the more you win, but that's not really the case. When you lead you run more than you pass and when you're behind you pass more than you run. On average.

So your conclusion is not exactly rocket science. We all know that if we continue to make plays on defense chances are greater we win the ballgame.

 

[Shouright]

Without looking at the data I could have told you that teams that run the ball more and throw less and teams that make plays on defense stand a great chance of winning football games.

I am not saying that running the ball more equals more wins, rather it's the other way round: when you have a lead you tend to run more than you pass. That's why the stats tell you that the more you run the more you win, but that's not really the case. When you lead you run more than you pass and when you're behind you pass more than you run. On average.

True, but the finding with regard to the relationship between rushing attempts and wins is also true for first half data, when it can't be argued that one team is running more because they're comfortably ahead. In fact, two of the strongest predictors of wins at halftime are YPA differential and rushing attempts differential, independent of the effectiveness of those rushing attempts.

So your conclusion is not exactly rocket science. We all know that if we continue to make plays on defense chances are greater we win the ballgame.

It appears to have been far too much "rocket science" for the Miami Dolphins' coaches over the first seven games of the season, who were using Ryan Tannehill during that period as though they thought he was more like Aaron Rodgers, rather than using the approach outlined above that was effective against a very good Cincinnati Bengals team.

 

[tcdrover]

The Dolphins can turn their season around & make this year a success by doing ONE simple thing...

SWEEPING THE JESTS!!!!

Seriously, I still have not recovered from them blowing the home game against the Bills with their probowl QB,
but we are only at the halfway point. Sweeping the jests is what would do it for me, and finishing at least at .500.

 

[Shouright]

The Jets just beat the Saints in New Orleans today using the same recipe outlined above, and even with a rookie QB.

Believe it or not, Geno Smith and company's ANY/A (6.36) was higher than Drew Brees's (5.96), and the Jets outrushed the Saints to the tune of 198 yards on 36 attempts (5.5 YPC) to 41 yards on 13 attempts (3.2 YPC). The Jets also intercepted Brees twice.

In the first half, the Jets ran the ball on 17 of 32 plays (53%). Similarly, in our game against the Saints, we ran the ball in the first half on 14 of 28 plays (50%) and were still in the game. The difference from then on was that the Jets continued to run the ball in the second half (53%), whereas we came out and passed the ball on 5 of the first 6 plays in the second half, with the game still in reach.

 

[Awsi Dooger]

Yeah, the Jets strategy today was very typical of the old NFL but you seldom see it anymore. I was planning to mention it somewhere tonight but this is a convenient venue.

Too many dunce teams these days are run by young offensive minded coaches. They encounter a game against Brees and brainstorm that they need to wing it 45 times or more with their third or fourth tier quarterback. It's comical suicide.

The Jets succeeded via the old contrast approach, maximizing what you do best and making life miserable for the cutesy passing team. These days so few teams legitimately emphasize defense that the elite quarterbacks are shell shocked when they aren't allowed to stand back there and be protected by the referees and the dainty strategy of the opposition.

When I used to bet against Marino in the '80s and '90s these are the games I looked for, ones in which the opponent would maul us physically and rack up the rushing attempts while we ran it 20 times or fewer. It seldom failed. I still have no idea how those Marino years are viewed fondly but I realize it's an argument I can't win around here.

 

[Shouright]

Yeah, the Jets strategy today was very typical of the old NFL but you seldom see it anymore. I was planning to mention it somewhere tonight but this is a convenient venue.

Too many dunce teams these days are run by young offensive minded coaches. They encounter a game against Brees and brainstorm that they need to wing it 45 times or more with their third or fourth tier quarterback. It's comical suicide.

The Jets succeeded via the old contrast approach, maximizing what you do best and making life miserable for the cutesy passing team. These days so few teams legitimately emphasize defense that the elite quarterbacks are shell shocked when they aren't allowed to stand back there and be protected by the referees and the dainty strategy of the opposition.

When I used to bet against Marino in the '80s and '90s these are the games I looked for, ones in which the opponent would maul us physically and rack up the rushing attempts while we ran it 20 times or fewer. It seldom failed. I still have no idea how those Marino years are viewed fondly but I realize it's an argument I can't win around here.

When does anyone ever "win" an argument around here? :unsure:

 

[Shouright]

Reading between the lines in the above data, I think we find there are two major ways that the Dolphins can turn their season around this year:

1) The league-wide correlation between pass attempts and ANY/A is nil (-0.02). However, the correlation between pass attempts and ANY/A for Ryan Tannehill is much greater (-0.40). What that means of course is that the fewer passes Ryan Tannehill throws, the better the Dolphins' offensive ANY/A, whereas for NFL quarterbacks league-wide, there is no such relationship between pass attempts and ANY/A.

This stands to reason, given that Ryan Tannehill is currently a developmental quarterback. Prior to the Cincinnati game, Ryan Tannehill attempted passes on more than 50% of the Dolphins' offensive plays. Finally, against Cincinnati last Thursday, as many people here had been suggesting, the Dolphins scaled back Tannehill's passing load, and he attempted passes on only 46% of the Dolphins' offensive plays. Instead, the run game was emphasized more than usual.

2) In the table above, notice the correlation between Cameron Wake's presence and the Dolphins' opponents' ANY/A. Wake was hurt during the Indy game, missed the following three games, played sparingly against Buffalo, and played increasingly more against New England and Cincinnati, culminating in the sack for a safety to win the Bengals game Thursday.

The correlation between the Dolphins' sacks of the opponent and their ANY/A in the games above is -0.61, meaning that the more sacks the Dolphins get, the lower their opponents' ANY/A. Likewise, the correlation between INTs thrown by the opponents above and their ANY/A is -0.75, meaning again that the more INTs the Dolphins get, the lower their opponents' ANY/A. Both of these figures (sacks and INTs) have been appreciably higher in the games this year in which Cameron Wake has not been sidelined or limited by injury, which of course stands to reason, given that Wake is one of the league's best at sacking the quarterback, and quarterback pressures often cause interceptions.

So, the take-home message here in my opinion is that the Dolphins have a recipe for winning the rest of the season, and it consists of:

1) Continuing to emphasize the running game on offense,
2) Continuing to limit Ryan Tannehill's overall passing load, and
3) Continuing to make plays defensively, spearheaded by Cameron Wake.

Tonight's game against Tampa Bay:

Dolphins' ANY/A = 4.27
Buccaneers' ANY/A = 3.65

Dolphins' pass attempts = 42
Dolphins' rush attempts = 14

Dolphins' defensive plays made = 3 (2 sacks, 1 INT, 0 fumbles recovered)

Too much imbalance in favor of the pass in this one IMO. Of course the running game wasn't working, but IMO this team needs the running game to work, and needs to feature it, to win in this league right now, given its other personnel on offense. Having Ryan Tannehill pass the ball on 72% of the team's plays from scrimmage isn't part of its recipe for winning right now IMO.

 

[Goonies]

You start by firing Ireland right now, so we finally get rid of that rotten tuna stench.