Warning: Cannot modify header information - headers already sent by (output started at /home/content/85/8979285/html/wp-includes/post-thumbnail-template.php:1) in /home/content/85/8979285/html/wp-content/plugins/wp-super-cache/wp-cache-phase2.php on line 62
cowboys game theory | The DC Times

The DC Times

A New Way to Look at the Cowboys, NFL, and Fantasy Football

By

Will the Cowboys be more aggressive moving forward?

Jonathan Bales

I’ve written frequently (specifically in the “Comments” of posts) about how and why the Cowboys should be much more aggressive moving forward in 2010.  At 1-6, there’s really nothing to lose, and Dallas could use the remainder of the season to answer important questions about their squad.  For example. . .

Playing Phil Costa is the aggressive move.  Will he remain the starter once Kyle Kosier returns?

Inserting Akwasi Owusu-Ansah into the starting lineup is the aggressive move.  Will that happen once he overcomes his high ankle sprain?

Sitting veterans such as Marc Colombo, Marion Barber, and Keith Brooking (at least in certain situations) is the aggressive move.  But will it happen?

And going for it on 4th and 4 from your opponent’s 40-yard line is the aggressive move. . .and we know what Coach Phillips has recently elected to do in such situations.

I bring this up because this organization is going to have a major problem if the current coaching staff continues to coach to “save face” instead of doing what is best for the Dallas Cowboys in 2011.  It’s clear that Phillips has recently made in-game decisions in an effort to “soften the blow” so to speak.  Why?  I know many of you think Jerry Jones already has too much power, but he needs to step in right now and make sure the 2010 version of the Cowboys is actually focused on becoming a better team in 2011.

I’m not at all in support of a mid-season coaching change.  Unless JJ thinks the future Dallas Cowboys head coach is already in the organization (which I sure hope isn’t the case), firing your current head coach and defensive coordinator makes no sense. . .UNLESS.

Unless said head coach knows his job is already out the window and is making important decisions based on appearing “respectable” in 2010.

I’m a competitor.  I want the Cowboys to win all the time and anything less than a championship, in my view, is a failure.  So while I want the Cowboys to win every remaining game this year, the future cannot be sacrificed in an effort to do that.  It’s almost irrelevant to me if Dallas ends up 1-15 or 8-8 this year. . .both records will be a failure.  I want the Cowboys to be true winners–the best of the best–and they need to realize that opportunity has passed this season.

Thus, they need to do everything in their power to prepare for a championship run in 2011.  If it means sacrificing the present, then so be it.  But if the goals of the current staff result in punting the ball on 4th and 4 from the opponent’s 40-yard line (or 4th and 3 from the 39-yard line last week), then something needs to change, and it needs to change now.

For those who are frequent visitors to DC Times, you know I try to back up everything I write with statistical proof that is highly relevant to my views.  So here we go. . .

On 4th and 4 from the opponent’s 40-yard line, the decision to punt the ball is incredibly detrimental to the Cowboys.  Statistically, they should go for it on all 4th down plays in that range up until and including 4th and 10.

Evidence of this comes in Advanced NFL Stats Win Probability graphs (which I highly recommend).  They take thousands of results from very specific game situations in the past and determine a team’s chances of winning a game at any particular moment.  What is the probability of a team winning a game when having a 1st and 10 at their own 20-yard line, down four, with three minutes left to play?  I’ve been amazed at the accuracy with which these graphs can provide that sort of information.

If you look at the Win Probability graph for the Cowboys-Jaguars game, you’ll see the Cowboys’ chances of winning decreased from 14 percent to 13 percent after their 4th down punt.

But this alone isn’t evidence that Phillips made a poor decision.  The effectiveness of a choice isn’t determined by how much it increases or decreases a team’s chances of winning, but rather how much it does so in comparison to the alternative.  If the Cowboys went for it on 4th down and gained just the four yards needed for a 1st down, the chances of them winning the game would have actually been closer to 20 percent.  Meanwhile, if they went for it on 4th down and failed, their chances of winning would have decreased to around 12 percent.

I’ll save you the monotonous math, but for the decision to punt to be the correct one, we’d have to assume that the Cowboys had less than a 25 percent chance of converting on that 4th and 4.  That’s clearly not the case.  Actually, the Cowboys have been 7 of 17 (41.2 percent) on 3rd or 4th and 4 dating back to the start of last season.  Of course that sample size isn’t huge, but the pool of data suggesting punting on 4th and 4 from your opponent’s 40-yard line is a horrible decision is statistically significant–meaning Phillips’ decision to do so isn’t just the “conservative” play, it’s also the wrong one.

By

Analyzing Cowboys’ 3rd Down Run Trends Thus Far in 2010



Jonathan Bales

A few months back, I published an article on the Cowboys’ 3rd down run plays from last season.  I did this to determine not only how successful the Cowboys were on these plays as compared to league averages, but also to learn if 3rd down runs held any strategic advantage over passes.

Here is a passage from that post:

In a recent post on why the Cowboys should pass out of “running” formations (and also in one on why teams should attempt a lot more 4th down plays), I spoke briefly about run/pass efficiency on 3rd down.  In short, NFL offenses fair much better when running the ball on 3rd and short (particularly 3rd and 1-3, but up until 3rd and 5).  Incredibly, running the ball is just as effective as passing up through 3rd and 10.  You can click the link above to read more about why this is so and view a graph displaying the conversion rates.

Nonetheless, I wanted to compare the Cowboys’ 2009 results with the league-wide numbers.  How effective was the offense when they ran the ball in “obvious” passing situations?  Note that these results (above) may be (very slightly) off from the numbers of Stats, Inc. or other unofficial stats companies because I did not use the televised ‘down and distance.’  For example, the televised version of a game may have mislabeled a play as ’3rd and 1′ when it was really closer to ’3rd and 2,’ and I have corrected for these mistakes to the best of my ability.

Notice the Cowboys’ yards-per-carry steadily rose (other than on 3rd and 6) as the yards-to-go increased.  This is obviously due to personnel and the game situation.  A defense which has substituted dime personnel on a 3rd and 10 is much more likely to yield a significant gain on the ground.  Of course, the yards-per-carry means nothing if the Cowboys are not achieving first downs.

2009 Third Down Conversion Rate

The chart to the right displays the conversion rate of all Cowboys’ 3rd down plays (of 10 or less yards-to-go) in 2009.  As you can see, the Cowboys were more efficient on 3rd and 1 or 2 when running the ball.  They converted 17/21 (81.0%) plays in these situations, compared to only 7/11 (63.6%) when passing.

As the distance-to-go increased, however, the conversion rate on runs dropped.  The Cowboys converted zero 3rd downs when running the ball with 8+ yards to go (although they attempted just four).

Interestingly, the conversion rate of 3rd down passes remained relatively stable, regardless of the distance-to-go.  You can see a very slight drop in the Cowboys’ 3rd down passing efficiency, but for the most part, the conversion rate was flat.  This is probably due more so to the team’s success in 3rd and long situations rather than an inability to convert on 3rd and short (when passing).

I give offensive coordinator Jason Garrett a lot of flack, but his 3rd down play-calling is generally outstanding.  I’d still love to see him run more on 3rd and medium (the ‘Boys ran just seven times on 3rd and 3-6 all season, compared to 42 passes).  Of course one would expect more passes in this range, but a slight increase in “surprise” runs would be in-tune with league-wide 3rd down conversion rates and could perhaps significantly aid the offense.

2010 3rd Down Runs

You can see above that the Cowboys’ 3rd down play-calling in 2009 fell in line with league averages: running was more successful than passing up until 3rd and 6.  I praised Garrett for his 3rd down play-calling last season, but I still hoped to see more 3rd downs runs this year.

So far, we are seeing those runs.  The chart to the left displays my findings.  Note that the sample size of plays is still small, so my mid-season and final analyses will be more statistically significant.  Still, it is interesting to see what sort of trends we can determine early in the season, if any.

Through three weeks, Garrett has run the ball a lot more on 3rd and 1-5.  In “true” short-yardage situations (one or two yards-to-go), the Cowboys have run on eight of nine plays this season.  That 88.9 percent rate is higher than last year’s 70.0 percent run rate in the same situations.  Further, the Cowboys are converting on a higher percentage of those runs.

On 3rd and 3-5, the Cowboys have run the ball twice this year.  That may not seem like much, but note that the Cowboys ran the ball just five times in that range in all of 2009.

Of course, we can’t be sure whether these numbers mean anything or not yet, but in combination with some of Garrett’s other early-season trends (which I will continue to detail throughout the bye week), it does appear as though the Cowboys’ much-scrutinized offensive coordinator is certainly using advanced statistics and game theory in a much more authoritative manner this season. . .and that is definitely a great thing.



Dallas Cowboys Times is on Twitter.

Subscribe to our free e-mail updates.

By

Should the Cowboys Run More on 3rd Down?

By Jonathan Bales

I’ve spent some time talking about Jason Garrett’s 3rd down play-calls in the past, but only as they relate to the previous play.  I found that, unlike during his 2nd down play-calling, Garrett is actually rather unpredictable on 3rd down (that’s a good thing, of course).

In a recent post on why the Cowboys should pass out of “running” formations (and also in one on why teams should attempt a lot more 4th down plays), I spoke briefly about run/pass efficiency on 3rd down.  In short, NFL offenses fair much better when running the ball on 3rd and short (particularly 3rd and 1-3, but up until 3rd and 5).  Incredibly, running the ball is just as effective as passing up through 3rd and 10.  You can click the link above to read more about why this is so and view a graph displaying the conversion rates.

Nonetheless, I wanted to compare the Cowboys’ 2009 results with the league-wide numbers.  How effective was the offense when they ran the ball in “obvious” passing situations?  Note that these results (left) may be (very slightly) off from the numbers of Stats, Inc. or other unofficial stats companies because I did not use the televised ‘down and distance.’  For example, the televised version of a game may have mislabeled a play as ‘3rd and 1’ when it was really closer to ‘3rd and 2,’ and I have corrected for these mistakes to the best of my ability.

Notice the Cowboys’ yards-per-carry steadily rose (other than on 3rd and 6) as the yards-to-go increased.  This is obviously due to personnel and the game situation.  A defense which has substituted dime personnel on a 3rd and 10 is much more likely to yield a significant gain on the ground.  Of course, the yards-per-carry means nothing if the Cowboys are not achieving first downs.

The chart to the right displays the conversion rate of all Cowboys’ 3rd down plays (of 10 or less yards-to-go) in 2009.  As you can see, the Cowboys were more efficient on 3rd and 1 or 2 when running the ball.  They converted 17/21 (81.0%) plays in these situations, compared to only 7/11 (63.6%) when passing.

As the distance-to-go increased, however, the conversion rate on runs dropped.  The Cowboys converted zero 3rd downs when running the ball with 8+ yards to go (although they attempted just four).

Interestingly, the conversion rate of 3rd down passes remained relatively stable, regardless of the distance-to-go.  You can see a very slight drop in the Cowboys’ 3rd down passing efficiency, but for the most part, the conversion rate was flat.  This is probably due more so to the team’s success in 3rd and long situations rather than an inability to convert on 3rd and short (when passing).

I give offensive coordinator Jason Garrett a lot of flack, but his 3rd down play-calling is generally outstanding.  I’d still love to see him run more on 3rd and medium (the ‘Boys ran just seven times on 3rd and 3-6 all season, compared to 42 passes).  Of course one would expect more passes in this range, but a slight increase in “surprise” runs would be in-tune with league-wide 3rd down conversion rates and could perhaps significantly aid the offense.

And this really has nothing to do with anything I just wrote, but Tashard Choice was at a waterpark today with his family. . .