A Statistical Look at Coaching Success

Given all of the coaching turnover this offseason (a record 11 new coaches), I wanted to see how coaching hires should be expected to fare with their teams.  Are they successful, a failure, or somewhere in between?  By looking at trends from the past, I hope to better understand how many coaches from this new batch of 11 we can expect to be successful, particularly Josh McDaniels.

For my data, I looked at every single new coaching hire since 1995.  Why 1995?  That was Mike Shanahan's and Jeff Fisher's first full season as head coach of the Broncos and the Titans/Oilers.  With Mike Shanahan's firing, Fisher is now the longest-tenured coachin the NFL, so starting in 1995 ensures that every team in the league has made at least one hire in my dataset.

Methodology and Data

Using, I gathered information on every coach hired since 1995, specifically name, team, years, wins, losses, and ties.  Using this information, I calculated a winning percentage and then classified each coach as 'Successful', 'Moderately Successful', or 'Unsuccessful'.  I used some basic guidelines for making these decisions, but there were some cases where I needed some subjectivity to make what I feel was the correct classification.  For instance, Dick Vermeil only coached 3 years in St. Louis and had a winning % of .458, which is relatively lackluster.  However, Vermeil brought the Rams their first Super Bowl trophy before he retired, and I am willing to bet that most people would consider his tenure in St. Louis to be a successful one.

In general, coaches with a tenure of three years or less are considered 'Unsuccessful' as well as four or five year terms with a winning % of .500 or below.  Coaches with 6 or more years or of service with one team are often considered 'Moderately successful', unless they have experienced a relatively large amount of postseason success in their career.  Coaches who have won Super Bowls or have long tenures with a team are considered 'Successful'.  I am sure some discussion could be had on my ratings, but I believe that they effectively classify the coaches, and would be happy to discuss any of the ratings.  Current head coaches also didn't follow these criteria strictly, as some 1-year head coaches, such as Atlanta's Mike Smith, would be considered successful at this point.

To see the table of coaching hires, classifications, and data, please follow this link.


There were 97 coaching hires between 1995 and 2008, an average of about seven new coaches per season.  Of those 97 hires, 21 represent current NFL head coaches and 76 represent past head-coaching stints.  As you can see in the table below, past head coaches were successful about 21% of the time and had a winning % of .567 and an average tenure of 7 seasons.  That's a big difference compared to unsuccessful coaches, who last less than 3 seasons, had a winning % of .356, and comprise 64% of that group.

Past NFL Head Coaches (Hired 1995-2008)

Successful Moderate Unsuccessful
Amount 16 11 49
Win % .567 .540 .356
Avg. Seasons 7 3.64 2.88
Size (%) 21% 14% 64%

Current NFL head coaches, as expected, have a much higher success rate than past coaches.  Most of these coaches still have a job for a good reason, as 90% could be considered at least moderately successful, with only 10% being unsuccessful.

Current NFL Head Coaches

Successful Moderate Unsuccessful
Amount 12 7 2
Win % .620 .520 .469
Avg. Seasons 4.92 3.57 3
Size (%) 57% 33% 10%

Those two datasets are interesting, and can give us some perspective when it comes to looking at current and past head coaches, but if we want to predict future success we will combine those two datasets so that we can have a composite set to work with.  Below are the statistics for all 97 coaching hires:

All NFL Head Coaches (Hired 1995-2008)

Successful Moderate Unsuccessful
Amount 28 18 51
Win % .590 .532 .360
Avg. Seasons 6.11 3.61 2.88
Size (%) 29% 19% 53%


Seeing this, we can expect a little more than one quarter of NFL head coaching hires to be successful, while a little over half are disappointments.

Analysis and Extrapolation

So how is this useful?  How does it apply to 2009?  From a strictly statistical approach, we can attempt to predict how many head coaches from this group will be successful, and how many will crash and burn.  If we take the above percentages and multiply them by the number of new coaching hires (11), we get some rough numbers:  3.19 successful coaches, 2.09 moderately successful coaches, and 5.83 unsuccessful coaches.  Rounding those numbers, we get 3, 2, and 6, which is the approximate number of coaches who will fall into each category if we were to categorize them 10-15 years from now.

It can be argued that the success rate should be expected to be even lower, considering that the teams hiring new head coaches have been relatively unsuccessful in general, and that's due to problems with the organization and talent on the team.  While that is true, I feel these ratios are reasonable and that some of the coaches that are currently considered successful may not be coaching in a few years.  Examples of this may be Wade Phillips, John Harbaugh, or Ken Whisenhunt.  If three such coaches drop from 'Successful' to 'Moderately Successful' or 'Unsuccessful', that will drop the success rate to 43% from 57%, something that should be expected.


[Update:  This last section was just thrown in for fun and discussion, I spent 10 minutes on it]

So knowing how many coaches to expect to succeed, how would I classify the new coaching hires?  See below:


  1. Raheem Morris, TB
  2. Rex Ryan, NYJ
  3. Mike Singletary, SF

Moderately Successful

  1. Josh McDaniels, DEN
  2. Tom Cable, OAK


  1. Todd Haley, KC
  2. Jim Mora Jr., SEA
  3. Steve Spagnuolo, STL
  4. Jim Caldwell, IND
  5. Eric Mangini, CLE
  6. Jim Schwartz, DET

How about you?  Who are your 3 successful, 2 moderately, and 6 unsuccessful coaches from this new batch of 11?

This is a Fan-Created Comment on The opinion here is not necessarily shared by the editorial staff of MHR.