DallasCowboysTimes

Unrivaled Cowboys Info 24/7

Cowboys Film Study: Counters

We recently studied the Cowboys’ success in 2009 running draw plays.  We discovered that, although Dallas is a superb draw-running team, the frequency with which the team ran the play caused their draw efficiency to decline as the season progressed.

In fact, the Cowboys actually averaged over a full yard less per carry on draws than on all other runs.  To regain the effectiveness of the draw, we concluded that the Cowboys must run less of them in 2010.  In doing so, defenses will be less prepared to defend them and the Cowboys can then reach the Nash equilibrium (the point where the average yards-per-carry will be maximized).

We decided to conduct a similar study on counter plays, with the results shown above.  Counter runs utilize misdirection–a running back either hesitates or starts one way before changing direction and receiving the hand-off.  Offenses will sometimes even pull linemen to the backside of the play to really confuse a defense.

Notice the incredible success the Cowboys had on counters last season, particularly Felix Jones.  Since counters are finesse type runs, it is logical that Jones received the most carries on counter plays and also gained the most yards.  His 10.0 yards-per-carry is absolutely ridiculous, particularly with a sample size as large as 22 runs.

Barber also performed fairly well on counters, perhaps because defenses were less inclined to expect a misdirection play with him in the game as opposed to Jones.  Thus, Barber’s counter average was higher than his yards-per-carry on other runs.

Tashard Choice’s low average means nothing because the sample size of just three runs is much too small to draw meaningful conclusions.

When comparing the overall counter stats with the numbers from the other types of runs (shown to the right), you can see just how effective the Cowboys were running counters in 2009.  They averaged 2.9 yards-per-carry more on counters than other runs, particularly because the opportunity for a big play is so much greater.

Notice the Cowboys had a significantly higher percentage of big plays on counters as well.  In fact, when running counters Dallas was 1.5 times as likely to run for 10+ yards, 3.5 times as likely to run for 20+ yards, and an incredible 6.9 times as likely to run for 40+ yards as compared to all non-counter runs.

It is worth noting that the percentage of negative plays on counters was higher than on non-counters, but this is to be expected from a finesse, misdirection sort of play.  Counters are generally run in situations when an offense is less likely to be debilitated from a negative play (such as 2nd and 5 as compared to, say, 3rd and 2).

Still, the Cowboys were only 1.5 times as likely to lose yardage on a counter as compared to a non-counter, so the risk was well worth the reward.

Ultimately, Dallas would be well-suited to significantly increase the number of counters they run in 2010, especially with Jones.  It may also be smart to replace some of the draw plays with counters, particularly because the two types of run plays are generally called in similar situations.

March 31, 2010 Posted by | Film Study/Stats, Miscellaneous | , , , , , | 4 Comments

Cowboys Film Study: 2nd Down Play-Calling Revisited

A few weeks ago we examined offensive coordinator Jason Garrett’s play-calling trends on 2nd down.  Our results shocked us perhaps more than any we have gathered this offseason.

The graph to the left displays Garrett’s tendencies.  Notice the disproportionate amount of times Garrett called a run play on 2nd down after a pass play, and vice verse.  On 2nd and 3 to 7, for example, Garrett dialed up a run just 29.5 percent of the time following a run on 1st down.  In the same exact situations, though, he called a run 76.5 percent of the time after 1st down passes.

Clearly a coordinator’s play-calling tendencies should not be based solely on the previous play-call (regardless of that play’s result).  We concluded Garrett fell victim to the idea that “alternating creates randomization.”  In his attempt to “mix it up,” Garrett actually became incredibly predictable with his calls.  True randomness has no regard for previous happenings.  As we have shown, however, Garrett allowed previous plays (not simply the result, but whether it was a run or a pass) to affect his current call.

After publishing that study, we received some criticism that our stats were meaningless without knowing the tendencies of other play-callers from around the league.  These criticisms, though, are unjustified.

We are not simply analyzing the percentage of run or pass plays in certain situations.  If that was the case, then yes, we would need to know league-wide tendencies to draw meaningful conclusions about Garrett’s own trends.

Instead, we are analyzing the percentage of runs/passes after a certain type of play.  Let’s look again at the above graph.  On 2nd and 3 to 7, Garrett was 2.95 times more likely to run after a 1st down pass than after a 1st down run.  We are not critiquing how often Garrett called a run in general during those situations–that information is meaningless to us.

Graph provided by AdvancedNFLStats.com

Since the down and distance on 2nd down is exactly the same regardless of the 1st down play-call, we would expect a truly random play-caller to dial up a run after a 1st down pass the same percentage of the time as after a 1st down run, regardless of what that specific percentage may be.  Thus, it is the overall run/pass percentage that would require the knowledge of others’ play-calling tendencies to be meaningful, but not the percentage of runs/passes in a specific down and distance following a specific type of play.

Nonetheless, we were still curious as to the play-calling trends of other coordinators in similar situations.  We had a feeling that, because humans perform so poorly in generating random sequences, we would see that others fall victim to the same fallacy as Garrett, i.e. that “mixing it up” will produce randomness.

Of course, it would be impossible for us to study film on every 2nd down play for every team for the entire 2009 season.  Luckily, we came across similar statistics on AdvancedNFLStats.com (a tremendous site that we highly recommend).  The numbers are listed just above.

The data consists of 14,384 plays, so the sample size is obviously large enough to draw meaningful conclusions.  During those plays, teams ran approximately 50 percent of the time after a 1st down pass, but just 28 percent of the time after a 1st down run.

We contrasted these results with Garrett’s 2009 2nd down play-calls (shown to the left).  Notice that Garrett calls a 2nd down run after a 1st down run at basically the exact same rate as other coordinators around the NFL.  His 2nd down run ratio after a 1st down pass is also incredibly similar to the league-wide average (54.3 percent to approximately 50.0 percent).

So, is this evidence that Garrett is justified in his play-calling?  Not at all.  Remember, opposing offensive coordinators are not involved in a zero-sum game (meaning the success of one does not necessarily cause the failure of the other).  Offensive coordinators around the league can collectively perform well, or collectively do poorly.  In the case of 2nd down play calls, it is the latter.

Further, not all teams suffer from this randomization fallacy at the same rate.  The Pittsburgh Steelers, for example, have done a tremendous job or randomizing their plays over the last few years (graph below).  Notice that their 2nd down run/pass ratio is nearly the same after a 1st down run as it is after a 1st down pass.  The closer these bars come to matching, the closer a team is to reaching the Nash equilibrium, and the more successful they will be on offense.

Graph provided by AdvancedNFLStats.com

Thus, the failures of other coordinators around the NFL do not justify the failures of Garrett.  His success is independent of that of other offensive coordinators.  As we wrote in our initial study of this topic, “If the Cowboys want to maximize the productivity of their potentially explosive offense, Garrett is the first person that needs to change.  Unfortunately, if his play-calling does not become less predictable, neither will the team’s fate in the playoffs.”

We aren’t backing off from that statement.

March 29, 2010 Posted by | Film Study/Stats, Miscellaneous | , , , , , , , , , , | 2 Comments

Cowboys Film Study: Draws

The Cowboys are thought of as one of the best draw-running teams in the NFL.  A lot of their success is due to the footwork of Tony Romo.  His quickness and athleticism allows him to effectively fake slant passes before handing the ball off to either Barber, Jones, or Choice.

As we progressed through the 2009 game film, we noticed that defenses began to become accustomed to this fake and (it seemed) were able to more efficiently defend the Cowboys’ draw plays.  We sorted through our database to uncover the offense’s draw statistics and what we discovered is below.

Note: Two 3rd and long draws were excluded as "give up" plays

Before we tallied the final numbers, we wanted to eliminate any draw plays that could be considered “give up plays”–those draws on 3rd and long that the Cowboys ran simply to gain field position and punt.  There were actually only two times all season that Dallas ran a draw on 3rd and 7 or more and these two plays were discredited.

The Cowboys ran 121 other draws for 547 yards last season (4.51 yards-per-carry).  This average is well below the 5.52 yards-per-carry the Cowboys maintained on non-draw plays.

But why would the Cowboys’ average be so low on a play which they are thought to run better than just about any other team in the league?  One possible explanation is the frequency with which Dallas runs draws out of the formation “Double Tight Right Strong Right.”

Remember in our study on Double Tight Right Strong Right, we noticed the Cowboys ran a strong side dive out of the formation 71.6 percent of all plays and 85.7 percent of the time when motioning into it.  The success of the dive decreased as the season progressed.  Dallas averaged a stout 7.8 yards-per-carry over the first five games but, as defenses became accustomed to the formation, the Cowboys were only able to manage 4.4 yards-per-carry on these dive plays the rest of the season (including just 3.2 against all teams except Oakland).

Of the 116 dive plays they ran out of Double Tight Right Strong Right, 23 of them were in the form of a draw.  The Cowboys gained just 87 yards on these plays for a per-carry average of 3.78 yards.

While this isn’t particularly efficient, the sample size of 23 plays is not enough to significantly alter the overall results of the overall draw plays.  Even if we disregard these Double Tight Right Strong Right draw plays, the Cowboys still averaged only 4.69 yards-per-carry (460 yards on 98 runs) on the remaining draws.

Ultimately, it appears as though the Cowboys’ poor average on draw plays is due more so to dialing up the draw too often than to them simply not being an effective draw team.  There is no doubt that draws can be extremely useful, but perhaps offensive coordinator Jason Garrett could maximize their effectiveness by calling them just a bit less often in 2010.

In the case of the Cowboys’ draw plays, the old euphemism holds true:  you really can have too much of a good thing.

March 26, 2010 Posted by | Film Study/Stats, Miscellaneous | , , , , | 6 Comments

Grading the ‘Boys, Part V: Safeties

In the first four parts of our Grading the ‘Boys Series, we provided in-depth statistical analysis and grades for the offensive linemen, running backs, and cornerbacks.  Today, we take a look at the safeties.

As was the case when grading the cornerbacks, we have to be very careful when interpreting the statistics we gather from our film study.  For example, despite generally being superior tacklers, we might expect the percentage of missed tackles to be higher for safeties than cornerbacks because the latter is forced to attempt less open-field tackles.

For this reason and others, it is also unreasonable to compare statistics between cornerbacks and safeties.  Comparisons can be drawn between players within a position, however, as long as we are aware of the possible limitations to such comparisons.

Below are the results of the 2009 Cowboys’ safety play and the corresponding Dallas Cowboys Times grades.

Notes

  • Chart Key:  TA=Thrown At, Rec=Receptions Yielded, PD=Passes Defended, Yds/Att=Yards Per Attempt Thrown At
  • The best stats are circled in blue, the worst in red.
  • Some of the stats were provided by Pro Football Focus.
  • The final chart details our own custom statistic, the Dallas Cowboys Times Pass Defense Rating.  It incorporates the factors we believe are most valuable in evaluating the success of a safety.  The amount of points a player scores in each category is less important than the difference between his score and the average score.  For example, a point total of 20.0 in a category where the league average is 5.0 helps a player more than a score of 100.0 in a category whose league average is 90.0.
  • The final grade is weighted 2:1 in terms of pass defense versus run defense.

Grades

2009 Safety Pass Defense Totals

  • Ken Hamlin

Pass Defense:  C+

Let’s start off with a grade for which we are sure to receive a lot of flack.  We have been stating from season’s end that Ken Hamlin’s 2009 play was not as poor as people made it seem.

There is no doubting that he is not a ball-hawking safety.  Would you like to have a player like that on your team?  Of course–but only if he doesn’t sacrifice his ability to prevent the big play.  Some safeties, i.e. Antrel Rolle, are considered “playmakers” because they get a lot of picks or have a lot of bone-crushing hits, yet they allow a multitude of big plays.  A true playmaker, though, is able to do these things without conceding long touchdowns.

Hamlin is not an incredible playmaker, but he is also not a liability in the secondary as many fans believe is the case.  He is a cerebral player who has done an admirable job of setting up the defensive coverages and forcing defenses to earn every yard they gain.  Sometimes it can be a good thing to not hear your free safety’s name called too much.  In a way, Hamlin is a bit of a sensei master in the secondary–leading the troops without overexerting himself.  We are only partially joking about that.

While Dallas could certainly benefit from a free safety who is a “true playmaker,” those players are few and far between.  Hamlin isn’t going to solely win the Cowboys football games, but he also won’t lose them.  He’s not an All-Pro sort of safety, but he’s also not one who should be released.

2009 Safety Pass Defense Efficiency

Run Defense:  A-

Let the ridicule begin.  An “A-” in tackling for Ken Hamlin?  Really?

You bet.  Hamlin missed just four tackles (8.0 percent) all season.  In comparison, Terence Newman led all cornerbacks by securing 91.5 percent of his tackles.  Thus, Hamlin was statistically the most consistent tackler in the secondary in 2009, despite playing a position that is arguably the hardest from which to make tackles.

Gerald Sensabaugh

Pass Defense:  C

Sensabaugh has the worst Dallas Cowboys Times Pass Defense Rating of all three safeties, but that is to be expected since he is targeted more frequently at strong safety.  Still, his 67.4 completion percentage against is much too high.

Like Hamlin, Sensabaugh did not make many big plays on the season, securing just one interception.  Unlike Hamlin, however, Sensy conceded a few easy scores.  He allowed five touchdowns (compared to just two for Hamlin), a stat which we do not even factor into our Pass Defense Rating.

Run Defense:  C+

2009 Safety Run Support Statistics

Sensabaugh’s missed tackle percentage of 15.6 percent was nearly twice that of Hamlin’s, despite generally playing closer to the line of scrimmage and thus obtaining more “easy tackle” situations.   He also secured just eight more tackles than Hamlin despite this difference in pre-snap alignment and playing more downs.

  • Alan Ball

Pass Defense:  B

Ball registered a worse score on our Pass Defense Rating than Hamlin, so why are we giving him a better grade?  Well, Ball’s inexperience led team’s to target him frequently.  In fact, he was thrown at on 6.53 percent of all snaps, nearly three times the rate at which opposing quarterbacks tested Hamlin.

Despite this, Ball allowed the lowest completion percentage of any safety at just 45.0 percent.  He also led the safeties in yards-per-attempt against and passes defended percentage.  It is not a stretch at all to label Ball the closest thing Dallas has to a “ball-hawk” at the safety position.

Run Defense:  D

Ball struggled quite a bit against the run.  He missed nearly 1/4 of all tackles, a rate almost triple that of Hamlin.  His tackle-per-play average was also the worst among the three safeties.

Final Safety Rankings

1.  Ken Hamlin:  82.3 (B-)

2.  Alan Ball:  78.3 (C+)

3.  Gerald Sensabaugh:  75.7 (C)

The Cowboys’ safeties are obviously not future Hall-of-Famers.  We believe Hamlin is unfairly ridiculed due to his lack of takeaways (and we realize we are the only ones who view him as underrated), but he is no Ed Reed.

Should the Cowboys address the safety position early in the draft?  If the value is there, yes.  Perhaps Texas safety Earl Thomas will drop down to pick #27.

If the Cowboys do not see good value in the first round, however, there is no reason to panic.  There are a wealth of intriguing second round safety prospects that should present adequate value for Dallas in round two, such as Georgia Tech’s Morgan Burnett.

Further, we believe Hamlin is still a starting quality safety.  He is certainly not irreplaceable, but it is unlikely that a rookie free safety, outside of Thomas or Tennessee’s Eric Berry, could step into the starting role and immediately perform better than Hamlin.

Assuming the team passes on a safety in the first round, expect Hamlin and Sensabaugh to be the Cowboys’ opening day starters and to force more turnovers in 2010.

March 23, 2010 Posted by | --- "Grading the 'Boys", Film Study/Stats, Miscellaneous | , , , , , , , , , , , | 4 Comments

Cowboys Film Study: Initial Drive Statistics

One of the more important aspects of an offensive coordinator’s play-calling ability,we believe, is his success on initial drives–those drives to begin a game and to start the second half.  At these points, a coordinator is generally calling scripted plays.  Thus, he has had all week to plan his attack on the defense (in the case of the initial offensive drive), or the entire halftime (in the case of the opening 2nd half drive).

When given time to prepare, offenses generally outperform defenses.  Year in and year out, the league-wide average yards-per-play and scoring totals for offenses are higher on the first drives of each half than any other drive.

Some teams, of course, are more apt to come out firing on all cylinders.  Unfortunately, the Cowboys have not seemed to be one of them in recent years.

Note: Drives during which the Cowboys simply took a knee were not counted.

First Half

The Cowboys’ first half average yards-per-play (5.78 yards) is significantly lower than the 6.45 yards-per-play the team averaged on what we will call “non-initial drives”– all drives excluding the first drive of the game and the first of the second half.

While point scoring can be fluky and thus susceptible to fluctuations over the course of just one season, Dallas’ points-per-drive was noticeably lower to begin the game (1.69) than in non-initial drive situations (2.30).

But why is the Cowboys’ initial drive success so poor?  Is it fair to place all of the blame on Jason Garrett’s play-calling?

Probably not.  Our guess is that it is a combination of Garrett and the overall mindset of the team.  Remember, Wade Phillips is an excellent “X’s and O’s” coach, but he is not exactly a top-notch pre-game motivator.

2nd Half

A stat that is more indicative of Garrett’s play-calling might be the initial drive of the second half.  At this point, most of the pre-game adrenaline and hype has faded and the mindset of teams has shifted to a more ‘cerebral’ approach.

Unfortunately, the Cowboys’ second half opening drive stats are atrocious.  As the chart details, the team averaged just 4.94 yards per play in these situations.  That is an astounding 23.5 percent drop from the non-initial drive average.

These second half failures are only worse in terms of points, as the Cowboys averaged just 1.06 points-per-drive to open the second half, a 54.0 percent drop from the non-initial drive average of 2.30.

In Garrett’s defense, the team did improve vastly at the end of the season, scoring a touchdown on the first drive of the game over the final three weeks.  They averaged a ridiculous 13.05 yards-per-play on those drives.

Let’s hope he can carry that success into the 2010 season.

March 18, 2010 Posted by | Film Study/Stats, Miscellaneous | , , , , , , | Leave a comment

Cowboys Film Study: Motion Statistics

NFL teams use motion for a variety of reasons:  to uncover defensive coverages, to get defenders out of position, to exploit positive match-ups, and so on.  The frequency of motions also differs greatly among teams.  Some, like the Bengals, like to motion very frequently.  Others, such as Peyton Manning’s Colts, almost never motion.

We analyzed our film database to determine just how many plays the Cowboys motioned in 2010 and exactly how effective those plays turned out.  The results, shown below, were a bit surprising.

As you can see, the Cowboys tend to run the ball at higher rate after motions than on plays where there is no pre-snap movement (46.4 percent runs after motion versus 36.2 percent non-motion).

However, this is not necessarily a knock on Jason Garrett, as the Cowboys frequently remain static pre-snap in situations where the defense knows they are going to pass.

For example, when the Cowboys lined up in “Gun TE Spread” (shown to the right), a formation that they threw out of 83.3 percent of all plays, the offense motioned only 12.5 percent of the time (as compared to a 42.8 percent motion rate on all plays).  Thus, while the run/pass ratio after motions is a bit skewed, it creates no real competitive advantage for the defense.

More significant than the rate at which Dallas runs or passes after motioning is the efficiency (or lack thereof) of the post-motion plays.  As you can see, the Cowboys gained significantly less yards-per-play on both runs and passes after motions (.7 yards less on passes and a full yard less on runs).

Why is this the case?  Are the Cowboys simply less effective on offense when they motion?

It is tough to say, but our initial thought was that the the drop in yards-per-play was caused by a possible tendency to motion on short-yardage plays.  Thus, the upside would be limited and the averages would suffer.

However, on short-yardage plays (three yards to go or less), the Cowboys motioned just 44.8 percent of the time–barely more than the 42.5 percent overall rate.  Thus, while it is good that Garrett effectively spreads out motions among various downs and distances, the low yards-per-play on motions cannot be attributed to an abundance of short-yardage plays.

Another possible explanation is that, because the Cowboys rarely motion in their hurry-up offense, the yards-per-play might be greater on non-motion plays because defenses are more likely to play soft and give up yardage.

The Cowboys poor running average on motion plays is not due to an abundance of short-yardage situations.

However, this could only explain the difference in passing average between motion and non-motion plays, as teams rarely run the ball in hurry-up scenarios.

Further, the Cowboys actually only ran a true hurry-up offense just 80 times in 2009, or just 8.0 percent of all plays.  Thus, it appears that the Cowboys success when not motioning is actually due to something meaningful rather than just down and distance or game situation.

This notion is strengthened by both the rate of big plays garnered and negative plays yielded in motion and non-motion situations.  As the chart shows, the Cowboys had their highest rate of big plays (10+ yards) out of static formations.

While this statistic could be affected by the aforementioned tendency of Dallas to not motion in hurry-up situations, the most surprising and meaningful statistic, in our opinion, is the rate of negative plays given up in each scenario.  The Cowboys actually gave up a sack after motions at nearly 2.4 times the rate they yielded a sack on non-motion plays.  Our sample size of both types of plays makes that number statistically significant.

Further, the rate of overall negative plays (sacks, negative runs, and negative passes) was nearly twice as high on plays where the Cowboys moved a player pre-snap.

But why is this the case?  Why do the Cowboys have a larger downside without an increased upside on plays that they motion, despite not running these plays in “low upside” situations?

For us, the answer is not entirely clear.  Perhaps Tony Romo is just much better at reading defenses than anyone thought.  Maybe motions give him no advantage in reading coverage.  We don’t want to put him on the same level as a Peyton Manning just yet, but perhaps he is approaching a portion of his career where, like Manning, he can effectively read a defense without resorting to pre-snap motions.

The Cowboys may be better off motioning less so defenses cannot decipher their play calls.

Another reason might be that the Cowboys’ motions are giving the defense an idea of where the play is going to be run.  After watching as much film as we do, there are times when we can predict with great precision what play the Cowboys are going to run.  How and where they motion is a big factor in our ability to do this.

This last explanation would explain the significant gap between motion and non-motion run average, as it is easier for a defense to decipher a particular run play from a motion than a pass.

Conclusions

The Cowboys are quite obviously less effective on plays which involve a pre-snap motion.  Further, the reason for this does not seem to be due to particular game situations.

So where should the Cowboys go from here?  Should they scrap motioning completely and resort to an Indianapolis-esque offense?

Like our solution to many of Dallas’ woes, we believe the “Nash equilibrium” should be implemented.  Again, this is the point where the Cowboys’ total yards would be maximized.

Thus, Garrett should steadily decrease the motion rate until the defense compensates enough that the Cowboys’ yards-per-play reaches its peak.  Our guess is that this is around 25.0 percent.  At this point, it is likely the rate of big plays and negative plays will also be maximized and minimized, respectively.

March 15, 2010 Posted by | Film Study/Stats, Miscellaneous | , , , , , | Leave a comment

Cowboys Film Study- Weak Side Runs

Throughout our film study articles, we have chronicled the trends of the Cowboys in certain specific situations, attempting to isolate the cause of their success or failure.  Some statistics are subjective, such as missed tackles, but we strive to obtain statistics that are objective as possible.

In this study, we will analyze the Cowboys’ weak side runs.  Like prior pieces, there is some “gray area” here.  What is a weak side run?  Is the weak side always opposite the tight end?

For this analysis, we have designated the weak side of the formation that which is opposite the tight end and has less than three skill position players.  Thus, in “Twins Left,” the right side is the strong side.  In “Twins Left, Weak Left” (below), however, the left side is strong.

If a formation had no tight end, the strong side is simply the side with the most skill position players.  Also, a multitude of formations have no strong or weak side, such as “Ace” (below).  These formations were not counted toward our results.

The findings we gathered are listed below.  The Cowboys averaged 5.2 yards-per-carry on weak side runs, compared to just 4.7 yards-per-carry on strong side runs.

Why did the Cowboys have more success running weak side than strong?  One possibility is that it surprises the defense.  Dallas ran weak side on just 19.5 percent of all run plays.  Thus, with the defense generally anticipating a strong side run, the success rate of running weak side increases despite a lesser number of blockers.

The lack of blockers to the weak side can also be a good thing because defenses generally line up according to the formation.  Less blockers, then, means less people to block, and less chance for mistake.

Still, if this theory is correct, we might expect the Cowboys percentage of big plays to increase when running weak side.  This is actually not the case.  The Cowboys ran for a play of 10+ yards on 15.3 percent of all weak side run plays in 2009, compared to 14.5 percent on all strong side runs.  This small difference is not statistically significant enough for us to draw meaningful conclusions.

Further, the percentage of negative runs is also approximately the same (9.4 percent on weak side runs versus 11.0 percent on all strong side runs).

With this lack of outliers, it appears as though weak side runs are just slightly more effective for the Cowboys than strong side runs.  The results are not simply skewed by a pair of 80-yard rushes, for example.

How should this information affect the Cowboys and Jason Garrett’s play-calling?  Well, as we detailed in our Witten blocking study, the play-calling should shift until it reaches the “Nash Equilibrium.”  Simply put, this is the point when the overall yards-per-rush will peak.

Jason Garrett will maximize offensive efficiency by always being one step ahead of defensive coordinators.

Note that Garrett cannot simply call all weak side runs because football is a game of opposing minds.  A drastic increase in weak side runs would obviously be met by a large percentage of defensive weak side blitzes.

Instead, game theory suggests Garrett should slowly increase the number of weak side runs until the average yards-per-carry is maximized.

But how will we know when that number is reached?  The answer would be simple if we assumed those people drawing up plays to try to stop the Cowboys offense–the opposing defensive coordinators–were perfectly rational.  In that scenario, the defense would call an increasing number of weak side blitzes until they minimized the overall yards-per-carry.  They would in effect create their own Nash equilibrium.

Of course, defensive coordinators do not always call plays in a rational manner.  Their knowledge is not unlimited, and so sometimes they may call too many weak side blitzes or not enough.  Perhaps sometimes the number of weak side blitzes they dial up has no correlation at all with the offenses’s weak side running rates or successes.

Garrett’s job, then, must be to take into account the thoughts and tendencies of defensive coordinators (perhaps easier said than done), and then adjust his play-calling accordingly.  If Team X calls an inordinate amount of weak side blitzes, for example, then the Cowboys own Nash equilibrium will be shifted to include more strong side runs (and vice versa).

Thus, play-calling (or effective play-calling anyway) is not simply about knowing your own players.  It is about successfully predicting the calls of defensive coordinators by knowing their tendencies.  This may sound extremely difficult (and it is from the standpoint of one individual play), but aberrations tend to flatten out over the course of a game in such a way that, despite not knowing individual play calls, a team can assume a “regression to the mean” of sorts where a team’s overall tendencies will always eventually shine through.

For Garrett, this means being one step ahead of the game.  Instead of simply knowing what you want to do, you have to know what your opponent thinks you are going to do, then adjust accordingly.  When playing against a really stealthy coach, you may have to know what he thinks that you think that he thinks about what play you are going to call.

If you are an offensive coordinator and have called three straight weak side runs in a row, for example, your natural inclination may be to deviate from this tendency.  You might do this in an effort to “mix it up.”  But game theory suggests you should take into account the opposition’s thoughts before making a decision.

Perhaps you know that he is thinking that you are thinking that he will call a weak side blitz to combat your recent success.  Knowing this, you would assume he may call a strong side blitz (or none at all), and you would be correct.  Thus, despite three straight weak side runs, the best play call is yet another weak side run.

Being “unpredictable” isn’t about changing play calls just for the sake of changing the play, but about adjusting your tendencies according to your opposition’s tendencies to create an environment where potential success will be maximized.

That may be the motto for the 2010 Dallas Cowboys– “maximize your potential.”  Should they do that, the team might just be playing in the first ever home Super Bowl.

March 14, 2010 Posted by | Film Study/Stats, Miscellaneous | , , , , , | Leave a comment

Grading the ‘Boys, Part IV: Cornerbacks

In Parts I-III of our “Grading the ‘Boys” Series, we analyzed the production of the offensive line and running backs.  We now swing over to the defense to critique the play of the Cowboys’ top three cornerbacks.

As is the case with every position in football, the success of the defensive backs is very dependent on the play of other positions, particularly those rushing the passer.  Thus, it can become difficult when comparing CB’s from different teams because the efficiency of their respective pass-rushers is directly correlated to the cornerbacks’ own success.

It is easier to compare CB’s on the same team, particularly if they do not match up with specific receivers.  This is the case on the Cowboys, as Terence Newman and Mike Jenkins generally play one side of the field regardless of where the opposition’s receivers line up.

Playing in the slot can be a bit different, and so we must be careful when comparing Orlando Scandrick’s stats with those of Newman and Jenkins.  The percentage of snaps that Scandrick is targeted, for example, will be higher than the starting cornerbacks because he is on the field in all passing situations, but not necessarily on running downs.

Still, we can gather the numbers and effectively isolate a player’s success to the best of our ability.  Below are the results of the Dallas cornerbacks’ 2009 play and the corresponding Dallas Cowboys Times grades.

Notes

  • Chart Key:  TA=Thrown At, Rec=Receptions Yielded, PD=Passes Defended, Yds/Att=Yards Per Attempt Thrown At
  • The best stats are circled in blue, the worst in red.
  • Some of the stats were provided by Pro Football Focus.
  • The final chart details our own custom statistic, the Dallas Cowboys Times Pass Defense Rating.  It incorporates the factors we believe are most valuable in evaluating the success of a cornerback.  The amount of points a player scores in each category is less important than the difference between his score and the average score.  For example, a point total of 20.0 in a category where the league average is 5.0 helps a player more than a score of 100.0 in a category whose league average is 90.0.
  • The final grade is weighted 4:1 in terms of pass defense versus run defense.

Grades

2009 Cornerback Pass Defense Totals

  • Terence Newman

Pass Defense:  B+

So much has been made of Mike Jenkins’ progression in 2009 that people tend to forget how outstanding Terence Newman played.  Newman’s health and ability to perform at his best was undoubtedly one of the primary reasons for the success of the Dallas defense.

Newman was the least targeted Cowboys’ cornerback in ’09, getting thrown at on just 9.49 percent of all snaps.  This statistic is very representative of the way opposing coaches feel about a player.  Newman may be underrated among general fans, but those in the league are very aware of his ability.

Newman recorded an impressive .728 passing yards allowed per snap, surpassed only slightly by Jenkins.  The 7.66 yards-per-attempt against Newman was the worst of all three cornerbacks, but this could be due to the fact that quarterbacks do not generally test him.  When Newman does get thrown at, there is a good bet his receiver is fairly open.

2009 Cornerback Pass Defense Efficiency

A common knock on Terence throughout his entire career has been his inability to make a play on the ball.  It is a valid criticism, as Newman logged just three interceptions last season, and we see it as his biggest weakness.  The largest difference between Newman and Jenkins in ’09 was this ability to make big plays.  Nonetheless, Newman is almost always in position, which surely aids his teammates in their quest to force turnovers.

The statistic which we value most, our own Pass Defense Rating (below), has Newman ranked slightly behind Jenkins in terms of 2009 pass defense efficiency.  Newman checked in with 236.39 points.  In comparison, Darrelle Revis, the most dominant pass defender by far last season, recorded 336.38 points.

Run Defense:  A-

The most underrated component of Newman’s game is his willingness to stop the run.  He recorded the most tackles and missed the least of any cornerback on the team last season.  In fact, his 8.5 percent missed tackle percentage was one of the best in the league.

  • Mike Jenkins

Pass Defense:  A-

No player on Dallas took as big a leap forward in 2009 as Mike Jenkins.  Jenkins, remember, began the season in a rotation with Orlando Scandrick as the Cowboys’ starting cornerback.  His play soon justified his stay in the starting lineup.

Jenkins gave up a completion on just 49.1 percent of passes thrown his way, leading the team.  He also led all three CB’s in yards-per-attempt, yards-per-snap, pass deflections, and, most importantly, interceptions (six).

2009 Cornerback Run Support Statistics

Because interceptions can sometimes be fluky and vary greatly from year to year, we do not put an extreme emphasis on them in our custom Pass Defense Rating.  Despite this, Jenkins led the team with 267.96 points.  Rankings among teammates, more so than among competitors, are very accurate because teammates deal with the same pass rush and game situations.

While we would rate Newman’s ability to purely cover as equivalent or superior to Jenkins’, the former USF cornerback gets the better grade because of his increased play-making ability.

Run Defense:  C+

Jenkins was ridiculed for dodging a tackle against the Giants in his rookie season, and it was obvious he placed emphasis on improving his run support in 2009.  Still, this part of Jenkins’ game needs work.  He recorded less tackles than Scandrick despite playing significantly more snaps.  He also missed 14.6 percent of all tackles he attempted.  This is not horrendous, but it can certainly improve.  Newman has proven that run support is more about “want to” than being physically-imposing.

  • Orlando Scandrick

Pass Defense:  C

Scandrick took a step back in 2009.  The fact that he even had a chance to start this season after being drafted in the fifth round in 2008 is a testament to how well he played in his rookie season.

Our Pass Defense Rating ranks is capable of effectively ranking cornerbacks who play on the same team.

In ’09, however, Scandrick was one of the most targeted defensive backs in the NFL (13.91 percent of all snaps).  Despite this and giving up completions on 62.9 percent of passes his way, Scandrick did a good job of limiting the yards-per-attempt to just 6.83 (Jenkins was only slightly better at 6.71).

Scandrick tallied only 151.90 points in our Pass Defense Rating, though, because of his high target rate and inability to make plays on the ball.

The problem with Scandrick was not that he was out of position or got beat a lot.  As we watched the film, it was apparent Scandrick’s speed and quickness allowed him to cover well, but, for whatever reason, he got outplayed once the ball was in the air.

Thus, his number one offseason priority may be working to get his head turned around in coverage to locate the ball, then subsequently using his athleticism to make a play.

Run Defense:  B-

Scandrick is slight of frame, but he doesn’t get manhandled in the run game.  He actually recorded three more tackles than Jenkins  This number could be inflated, however, because Scandrick lined up closer to the ball-carrier and also gave up a significant number of completions where he was able to immediately make a tackle.

Still, Scandrick had a lower percentage of missed tackles than Jenkins.  Tackling from the nickel position is generally more difficult than it is for a cornerback lined up out wide because a nickel cornerback is in the open field and does not have the ability to utilize the sideline as an extra defender.

Final Cornerback Rankings

1.  Mike Jenkins:  89.8 (A-)

2.  Terence Newman:  88.2 (B+)

Orlando Scandrick may have competition in 2010, perhaps from Javier Arenas.

3.  Orlando Scandrick:  76.6 (C)

So where do the Cowboys go from here concerning the cornerback position?  It is obvious they are highly talented on the outside with Newman and Jenkins, but should they upgrade the nickel spot?

In our opinion, Scandrick has the ability to significantly improve his performance in 2010.  It is quite apparent that he is very close to taking that next step.  The most important aspect of his success will be gaining experience.  With experience comes knowledge, and with knowledge comes success.

Sometimes it appeared as though Scandrick was a bit hesitant on the field in ’09, and the knowledge he will gain from more experience will allow him to “stop thinking” and let his natural ability take over.  There is no doubt that he has the requisite talent to be an incredible cover corner.

We could also see the Cowboys addressing the position during the middle or late rounds of the draft.  One player we are very high on is Alabama CB Javier Arenas.  Arenas primary role in Dallas would be as a return man, but he could also push Scandrick for the nickel spot.  Perhaps a little competition is just what Orlando needs to thrive in 2010.

March 11, 2010 Posted by | --- "Grading the 'Boys", Film Study/Stats, Miscellaneous | , , , , , , , , , , | 7 Comments

Cowboys Film Study- Second Down Play Calls

Jason Garrett's second down play calls were very predictable.

In our first statistical analysis of Jason Garrett’s play-calling, we noticed that he was tipping plays via the formation.

As we studied the Cowboys’ game film this offseason, we also noticed a play-calling trend on 2nd down.  On the majority of 2nd down plays, it appeared as though Jason Garrett called a run if the team passed on 1st down, and vice versa.

We looked into our stat database and the results are shown to the right.

As you can see, our hunches were correct.  Other than on 2nd and 1 or 2, the Cowboys ran significantly more after calling a pass on 1st down.  There were actually only 12 plays called on 2nd down and 1 or 2 after a 1st down pass, so that sample size is probably too small to make significant conclusions.

In the other scenarios, though, the sample size is plenty big enough to conclude that Garrett was tipping plays via the down and distance.  On 2nd and 3 to 7, for example, Garrett dialed up a run on only 23 of the 78 (29.5 percent) plays that followed a 1st down run.  After 1st down passes, though, the Cowboys ran on 2nd down on 26 of 34 plays (76.5 percent).  Thus, Dallas was 2.95 times more likely to run on 2nd and 3 to 7 after a 1st down pass than after a 1st down run.

On 3rd and 8 to 10, that trend, surprisingly, did not get much better.  The team ran on only 10 of 50 plays (20.0 percent) in these scenarios following a 1st down run.  After passes, Garrett called a run on 32 of 58 2nd down plays (55.2 percent), meaning the team was 2.76 times more likely to run on 2nd and 8 to 10 after a pass than a run.

On 2nd and 11 or more, the team was still 2.33 times more likely to run after a 1st down pass than after a run.  Obviously Garrett did some things right in the past few years, but this sort of predictability is unacceptable.  Perhaps the offense’s success in recent years is not because of Garrett, but in spite of him.

Jason Garrett's predictability must change for Dallas to reach their offensive potential.

If we are obtaining these numbers and noticing these trends, you can bet opposing defensive coordinators are aware of them.  Imagine if the Cowboys passed for five yards on 1st down and lined up in Double Tight Right Strong Right on 2nd and 5.  The defensive coordinator could be all but sure that the Cowboys would run up the middle.

So why is Garrett so predictable in these situations?  Our hypothesis is that, ironically, he is trying to be unpredictable.  In an effort to seem “random” in his play-calling, Garrett is “mixing it up.”  True randomness, though, means each event is independent of the previous one.  What is the chance that a flipped coin will come up heads after it came up heads six straight times prior?  Still 50 percent, because the previous events have no bearing on future ones.

While previous football plays do have an impact on future ones (a team is more likely to pass on 2nd down after a sack than a nine yard rush), if Garrett wants to become unpredictable, he must not allow the previous play call affect the current one.  In situations such as 2nd and 3 to 7 when a team is probably about as likely to pass as it is to run, Garrett is particularly predictable.  He thinks he is “mixing it up” by calling a play that is the opposite of the previous one, but in doing so, he is actually dissolving any chance of randomness for which he is shooting.

In fairness to Garrett, he is not the only offensive coordinator that suffers from this delusion.  Studies have proven that it is impossible for human beings to produce a truly random sequence.  We naturally assume that rather long strings of the same occurrence are unlikely in a random sequence, when in fact they are to be expected.

For example, which series of five coin flips is more likely:  “heads, heads, heads, heads, heads” or “tails, heads, heads, tails, tails”?  The answer is that they are both equally likely, yet this doesn’t seem to mesh well with common sense because of our notions of causation.

It is nearly impossible for humans to separate previous events from current ones, a task that is imperative to create randomness.  Thus, almost paradoxically, for play-calling to be as random as is humanly possible, the play-caller would have to try to forget about being random.  He would have to implement an almost “non-aware awareness,” meaning he has to be aware of natural human tendencies concerning randomness without letting this awareness adversely affect his own ability to randomly call plays.  This is evidenced by the fact that it is Garrett’s attempt to call plays randomly that is hindering his ability to do so.

We doubt many offensive coordinators take this approach when calling plays.  Still, the failures of other coordinators do not justify those of Jason Garrett.  If the Cowboys want to maximize the productivity of their potentially explosive offense, Garrett is the first person that needs to change.  Unfortunately, if his play-calling does not become less predictable, neither will the team’s fate in the playoffs.

March 3, 2010 Posted by | Film Study/Stats, Miscellaneous | , , , , , , , , | 1 Comment

Grading the ‘Boys, Part III: Running Backs

In Parts I and II of our “Grading the ‘Boys” Series, we analyzed the efficiency of six Cowboys’ offensive linemen in both run blocking and pass protection.  In doing so, we attempted to isolate one component of the offense as effectively as possible to determine the worth of individual players.

In reality, of course, offenses are holistic systems.  The productivity of each position indirectly affects the ability of players at each other position to properly perform.  In the running game, the success of linemen is affected greatly by the talent level of the running backs, and vice versa.

In Part III of our “Grading the ‘Boys” Series, we will study the productivity of Marion Barber, Felix Jones, and Tashard Choice.  These three players all contributed in different ways and in distinct situations, so we will keep this in mind when analyzing the statistics we gathered from our film study.

Marion Barber was surprisingly ineffective on runs up the middle in 2009.

Notes:

  • In this particular analysis, we will grade each running back on four components:  short-yardage running, overall running, receiving, and pass protection.
  • The four components of the overall grade are not all equal.  They will be weighted 15/45/20/20, respectively.
  • All totals include the playoffs.
  • As always, the best stats are circled in blue, the worst in red.

Grades

  • Marion Barber

Short-Yardage Running:  D-

Barber had by far the lowest average on the team on runs up the middle and in short-yardage situations, averaging just 2.8 yards-per-carry and converting a ridiculously low 56.0 (14 for 25) percent of the time with just one yard to go (for either a first down or a touchdown).  His yards after contact and number of broken tackles were both down significantly from prior seasons.

Would you have guessed Felix Jones led the team in broken tackles?

Overall Running:  C

Barber was more effective on draws and counters than in short-yardage situations.  His 4.2 average was mediocre, but he did carry the ball a lot more in these short-yardage and goal-to-go scenarios.  He still scored seven touchdowns, but only about once every 29 carries.

Receiving: B

Barber was again solid in the passing game, although his receptions decreased due to the presence of both Felix Jones and Tashard Choice.  Barber does an excellent job of disguising his intentions on screen passes.

Pass Protection: B

Marion gave up the most sacks of any running back (three), but he was also on the field during pass plays about twice as much as the other backs.  He does a great job of taking on defenders much larger than him, usually coming out on top.

Overall Grade:  77.2 (C+)

  • Felix Jones

Short-Yardage Running: B+

Despite not receiving a lot of short-yardage carries (five), Felix Jones converted on them 100.0 percent of the time.  His runs up the middle, which may be more indicative of his short-yardage abilities than such a small sample size, is still solid at 4.1 yards-per-carry.  It is hard to relate this number to Barber’s, though, because Barber had more short-yardage runs up the middle which would have decreased his average.

Overall Running:  A

Jones really displayed his value to the Cowboys in 2009.  He averaged a ridiculous 6.2 yards-per-carry, including 220 yards on 22 counters.  He also surprisingly led the team in yards after contact and broken tackles.

Receiving:  B

Jones’ receiving numbers were solid, but with his explosiveness, we would expect them to be a bit higher.  They should increase next season, particularly with teams focusing in on the run when he is in the game.

Pass Protection:  B

We think Jones is underrated in pass protection.  He gave up one sack on the season, but he is rather good at an aspect of his game which most believe is his biggest weakness.

Overall Grade:  89.8 (A-)

  • Tashard Choice

Short-Yardage Running:  B+

It is difficult to grade this aspect of Choice’s game.  He was a respectable 5 for 7 in short-yardage situations and also led the team (by a lot) with a 5.8 yards-per-carry average on runs up the middle.  However, he averaged only 1.9 yards after contact and broke just three tackles all season.

Felix Jones was incredible on counter runs, averaging 10.0 yards per attempt.

Overall Running:  B

Choice averaged 5.0 yards-per-rush on the season, with a large chunk of his rushing yards coming from the Wildcat formation.  It seemed like he was more comfortable taking the direct snap than on other runs, as he was worst on the team in both counter and draw average.  His 5.1 percent touchdown rate led the squad.

Receiving:  B+

Choice led the team in reception average, which you would expect out of your third-down back.  He isn’t incredibly fast or strong, but just solid in all aspects of the game.

Pass Protection:  A-

Choice really has done an excellent job in pass protection since his rookie season.  According to our film study, he didn’t allow a sack all season.

Overall Grade:  87.3 (B+)

Final Running Back Rankings

1.  Felix Jones:  89.8 (A-)

2.  Tashard Choice:  87.3 (B+)

3.  Marion Barber:  77.2 (C+)

A lot of fans are calling for the Cowboys to trade Marion Barber.  Other than the fact that his contract makes this basically impossible, we don’t think it would be a smart move.  Barber still has his place on the team.  If the ‘Boys would trade any of their backs, they would turn a positional strength into a possible weakness.  The team would be just one injury shy of having only one experienced running back on the roster.  Three good running backs is certainly a luxury, but it also is an important component in the success of the Dallas offense.

So how should Jason Garrett alter how he utilizes each of these players in 2010?  First, Barber needs to get fewer carries.  He received 54.3 percent of the regular-season rushes in ’09, with Jones garnering 29.5 percent and Choice 16.2 percent.  In 2010, we would advise the following breakdown:

  • Felix Jones:  50 percent
  • Tashard Choice:  30 percent
  • Marion Barber:  20 percent

Some may argue that we would be giving up on Barber too quickly, but now is not the time to wait on players.  The Cowboys are built to win now, and the most productive players should play.

These percentages could be attained by starting Jones and letting him play two series for each of Tashard Choice’s one.  We would also use Tashard Choice on short-yardage runs, including a bit more Wildcat.  Barber would come in to spell Jones and Choice, particularly on third down, and to finish games out.  In baseball, closers are only successful because they haven’t pitched all game.  The same is true for Marion Barber.  By saving his energy, he could effectively return to the “closer” role, creating the most efficient Dallas Cowboys backfield possible.

In our next “Grading the ‘Boys” segment, we will analyze the productivity of the cornerbacks.

February 27, 2010 Posted by | --- "Grading the 'Boys", Film Study/Stats, Miscellaneous, News, Notes, and Rumblings | , , , , , , , , , | 5 Comments