Mike Durnin - Analytics - Article

How To Revolutionize Defensive Game Planning With Analytics

In the game of football there are a multitude of attributes related to success on the scoreboard. Analytics is one of those attributes. However, to believe there are one or two attributes more integral to success than others is foolish.  Allow me to use the analogy of a bicycle wheel (Diagram 1). Think of players as the outer rim of the wheel and the hub as success on the scoreboard. The spokes are the various attributes such as nutrition, training, understanding of offensive/defensive concepts (game education), psychological/motivational aspects, statistical analysis, team dynamics, etc.

Analytics; Game Planning; Defense

As any professional cyclist will tell you, if any spoke is weakened, the efficiency of the wheel is diminished. As such, we believe it is important to make sure you are strengthening each and every attribute (or spoke in the wheel) in order for it to be as strong as possible to provide the greatest opportunity for success.

RELATED ARTICLE: Inside The Headset: Bronco Mendenhall & Michael Christensen [PODCAST]

In the area of statistical analysis – or as some call it, analytics – each coach or segment within the program is looking to answer different questions. The most important aspect is to effectively research the questions coaches want answered and, even more importantly, be able to effectively report the findings in such a way that coaches and players are able to understand and utilize the information. We will look at ourselves (self-scout), our opponents, and league tendencies, along with other leagues or levels for factors associated with success on the scoreboard. This information is then used to “teach” the game of football to our players.

Sam Kash Kachigan in his book, Statistical Analysis, An Interdisciplinary Introduction to Univariate and Multivariate Methods, used a basic diagram to simply describe how analytics are utilized (Diagram 2). He began with the world before analysis; then data is collected. This data is then organized and manipulated, after which the data is interpreted. Then you have a new world which begins the process again.

Analytics; Game Planning; Defense

Football analytics functions in a similar manner (Diagram 3). One can look at it in terms of each individual play, each individual game, a season, etc.

Analytics; Game Planning; Defense

Now, what are the questions that need to be answered? While the exact questions will differ for every coach, there are basic questions all coaches should be looking to answer. Here is a list of questions one would ask if they were looking at an opponent’s defense or performing a self-scout of their offense.

Who is our Opponent Offensively?

  1. Favorite Runs
  2. Favorite Passes
  3. Run-Action Plays
  4. Special Plays
  5. Favorite Receiver
  6. Favorite Sets/Formations
  7. Favorite Hit Chart (Run/Pass)

The astute student of the game will understand that answers to the basic questions will tell a more complete story in specific situations. Yes, while the game is played sequentially, it is best studied thematically. The same thought process would be utilized when studying a defense.

What are the answers to those questions in the following situations?

  1. Start of a series
  2. Normal Down and Distance
  3. 2nd and Long
  4. 3rd and Short (we use the parameters of 3 and 1 or 2 yards)
  5. 3rd and Medium
  6. 3rd and Long
  7. 3rd and X-Long
  8. 4th and Short
  9. 4th and Medium
  10. 4th and Long

Furthermore what happens in each of those situations when you add other parameters?

  1. Personnel Grouping or Formation Family
  2. Field Position
  3. Score Differential (Discussed later in the article)
  4. Time in the Game
  5. Result of previous play or series
    1. After an explosive offensive run play
    2. After an explosive offensive pass play
    3. After a negative offensive play
    4. After a turnover

Each of these answers will tell a story about the offense. (You would study the same types of questions and situations when studying a defense.) At this time, one must realize it is highly unlikely any statistic will be 100 percent for one play or target. While I just indicated 100 percent is unrealistic, there can be situations when even the best teams can become predictable. In 2012, the New England Patriots ran over the left side on the first 3-and-1 situation in each of the first 13 games of the regular season. They again did it in game 14 and the 49ers blitz the game for a TFL effectively stalling the first drive of the Patriots.

While the questions within each of these situations will provide a wealth of information, one of the keys again is the ability to organize the data so you are able to answer the questions coaches and players want answered and to have it be understandable at all. The process does require time, especially in terms of labeling each play – the accuracy of which is of upmost importance. However, once the labeling is completed, the time element to provide simple reports occurs rather quickly.

RELATED ARTICLE: Understanding the Allure Of Football Analytics

Side note, it is important the individual(s) labeling each play must take an attitude that they are in the notation business and not in analysis mode. This will allow them to quickly go through all the plays. We actually split the responsibilities throughout our defensive staff so each coach has at least a couple of attributes they are charting. We will collect information on myriad categories or aspects of each and every play.

The following chart (Diagram 4) outlines many of the categories we will chart on each play. Note, it is important that charting plays off of video is different from studying an opponent on video. For the sake of time, one must dedicate himself to charting (or looking for specific input information) rather than studying an opponent’s play, etc. We will also chart aspects of the defense the opponent is facing for each play, such as type of front, pressure or not, basic structure of the coverage.

Analytics; Game Planning; Defense

One aspect to keep in mind as you see the various attributes labeled is that it is important to be able to give yourself groupings with meaning. As such, we will have a name for the play, but we will also indicate a concept. So while we may label a play 24 counter, we will also label it as a “gap” play. In the same way, we will label a formation as Trace, but also label it in a formation family of “1 X 2”. All of this will help you organize and give you more meaningful information.

In addition, we are constantly collecting data on ourselves during the course of the season for self-scouting purposes. This self-scout is important as it is the information our opponent is using to study us. The following chart (Diagram 5) is an example of the some of the information we chart on ourselves on a play-to-play basis.

Analytics; Game Planning; Defense

We use the “UD_D Coach” attribute when the defensive coordinator indicates there is another aspect specifically investigated later in the week for a certain opponent.

As Kachigan alluded to, once the information is collected, organized and manipulated, it is important to effectively interpret and report to coaches and players. One of the keys we have found is to create one-page reports. Each page is a synopsis of the important information. While someone will need to analyze the pages upon pages from various reports, it is much easier and time-saving to create one-page reports and provide playlists (clips) to other coaches and players.

In studying an opponent’s offense, we believe it is imperative to thematically study them as we develop the one-page reports. In order to accomplish this one must arrange a series of questions. Think in terms of a matrix.

The matrix (Diagram 6) will give an overview of the offensive philosophy of your opponent within situation.

Analytics; Game Planning; Defense

One of the areas we study is the pass offense of our opponent. While not earth shattering to coaches, let me explain the various aspects we study our opponent’s pass offense.

We group plays from the opponent’s pass offense into the following general concepts:

  1. Dropback Pass (DB)
  2. Quick Pass (QK)
  3. Move (MV) (includes sprint- or dash-type plays)
  4. Play-Action Quick (PA-QK) (includes RPOs)
  5. Play-Action Dropback (PA-DB)
  6. Play-Action Move (PA-MV) (includes boot, naked)
  7. Screen (SCR)
  8. Trick (TR)

We believe all passes fall into one of these groups and gives us a great overview of our opponent’s pass offense. We also chart each throw by the quarterback in reference to where the pass is being thrown. The Pass Zone Analysis Chart (Diagram 7) gives you the reference points we use on a weekly basis looking at not only where the ball is being thrown, but to whom it is being thrown. The same format is also looked at in terms of where the ball is being thrown by type of pass play (DB, QK, etc.).

Analytics; Game Planning; Defense

We also study pass charts examining following situations:

  1. 3rd Down situations
  2. Red Zone situations

One aspect we will always go back and study is how philosophy changes in reference to score differential. We use the following categories to classify each play in a breakdown:

  1. Zero (Tie game)
  2. Positive 1 or Negative 1 (score differential if the game is within 8 points)
  3. Positive 2 or Negative 2 (score differential if the game is within 9 to 16 points)
  4. Positive 3 or Negative 3 (score differential if the game is more than 16 points)

All of this information lends itself to practice emphasis and game planning for the defensive staff and facilitate the “game education” of the players.

Self-Scout

One of the most important aspects of football analytics deals with an exhaustive study of what you do! In simple terms, the self-scout. Knowing what you do when and how well you are doing it is essential for improvement. The essence of the self-scout provides insight as to what your opponent is seeing when they are studying you.

As we do with opponent scout, we study the game thematically.

No matter which situation we study, we always look at our success percentage. This is determined by percent of time we “win the down”. For us to win the down defensively, we determine the situations for the following:

1st and 10Hold the offense under 4 yards (if 1st down is a different distance, hold under 40 percent of the yardage)
2nd DownHold the offense under ½ the distance for a first down (if 2 and 6, hold under 3 yards)
3rd DownHold the offense less than distance for conversion
4th DownHold the offense less than distance for conversion

We will also study the percentage of explosive plays in the various situations. We determine explosive plays as:

  1. Runs of 12 yards or greater
  2. Passes of 16 yards or greater

The information on the success of winning the down and limiting explosive plays is critical in game planning. You may have a defense call that is good at winning the down (60 percent of the time or more of winning the down), but 25 percent of the plays are explosive, in essence you have a high-reward/high-risk call. If you have a defensive call that is average at winning the down (45 to 59 percent), but only allows 5 percent explosive plays, you in essence have a “safe” call. While the actual call to be used in a game is up to the coordinator and coaches, having this type of information strengthens the ability of the signal caller.

An example of this type of information can be found in the following end of the season chart (Diagram 8). This chart studies “where” each pass has been thrown versus our defense. We can also use this format to then study what type of passing play was used in each situation. The chart can be and should be used on a weekly basis. Once you look at the numbers, it is then important to study the corresponding video to determine the reasons why!

Another aspect we study is “how” we have aligned to various formation families and formations. Have we shown an open coverage look (2 high safety look), a closed coverage look (1 high safety look) or a zero look (blitz look).

Again, the list of questions is endless. Your own mind and the minds of your staff have only to be open to the realm of possibilities.

On behalf of the entire coaching staff at the University of Dubuque, we sincerely hope you were able to find a few pieces of information you will be able to use within your program and specifically in your analytics. While we have touched on a number of areas within the world of analytics as it pertains to football, there is much more to this spoke on the wheel towards to success. A special thanks to Mike Eayrs (my mentor) for all your guidance through the years!

Please don’t hesitate to contact us if you have any questions. Best of wishes to all you and use continue to lead young people – teaching life lessons through our great game of football.

____________________________________

Mike Durnin joined the University of Dubuque in January of 2013. He enters his 30th year in the collegiate coaching ranks and serves as the assistant head football coach along with coaching the defensive line at UD. Prior to that, Durnin had been the head football coach at Luther College in Decorah, Iowa, from 2008-2012. Durnin began his carrer at Minnesota State coaching wide receivers and defensive backs. From there he spent two seasons as defensive coordinator at Yankton College and seven years as a coordinator at the University of Puget Sound (4 years defensive, 3 years offensive). He also served as the backfield coach and recruiting coordinator at Illinois State and as the defensive line coach and defensive coordinator at UW-La Crosse .


For more information about the AFCA, visit www.AFCA.com. For more interesting articles, check out AFCAInsider.com and subscribe to our weekly email.

If you are interested in more in-depth articles and videos, please become an AFCA member. You can find out more information about membership and specific member benefits on the AFCA Membership Overview page. If you are ready to join, please fill out the AFCA Membership Request Form.

Comments 1

Leave a Reply

Your email address will not be published. Required fields are marked *