The plan was to take some time away from the Daily Short to welcome a new baby to the family and update the Hot Seat Study database for the first edition of a "Tom Herman and the Hot Seat" column. Upon starting to update the database for the hot seat study, I realized that we would have to wait until after the bowl games are over to update win totals for the latest season.
So, in the interim, @Ketchum suggested I do a smaller study based on the data we currently have (from 2005 through the 2016 seasons) about just how important/unimportant an upcoming bowl win is to Tom Herman's likelihood of finding himself on a future hot seat, based on recent historical precedent set by the 234 tenures examined through the end of the 2016 season.
For those who are new to Orangebloods, or have not seen the highly controversial editions of "Charlie Strong and the Hot Seat" over the previous years (it's fun to go back through and read the comments in these threads calling the studies, which ended up being accurate in their predictions of exactly when Strong and countless others would be fired, "worthless garbage," etc.) here is a small explanation of the total study before we move forward on this small piece of the puzzle. Our focus for this content will be far-less expansive than the new study which will come out at some point in the spring of 2018, and more narrowly focused on a six-versus-seven-win season in Year 1 and the importance of winning that first bowl.
Explanation of the study
We’ve once again examined every Division I coaching hire that has been made from 2005 to, now 2016, in an attempt to better understand - and in some ways quantify - the abstract idea of the hot seat. The goal is to identify the circumstances that lead to both finding and avoiding it.
Here is a link to the data.
On how the data was gathered and organized:
- Each Division I coaching hire during the span was chronicled from the time of its start to its eventual end or the present - whichever was applicable.
- Each hire was categorized as either an offensive or defensive hire based on the head coach’s philosophical football background.
- Each year of each tenure is chronicled by W-L record within the horizontal cells leading to present day or date of tenure's last service. The cells are highlighted to note these six events, as are the coaches' name cells in the “New Coach“ tab.
RED: FIRED - the red cells indicate that the head-coaching tenure was ended due to firing in the corresponding year. If a red cell is not preceded by a pink cell, it means that the coach’s seat did not become hot as detected by the Deep Dig until the year he was fired.
PINK: ON THE HOT SEAT - the pink cells indicate that the given year within the tenure was either spent in, or led to, hot-seat status. This abstract variable was assigned via research on every tenure examined. Typical events that led to “hot seat”-assignment were items such as:
- public responses from athletic directors or other university figureheads asserting the coach in question would indeed be back next season, public ‘votes of confidence,” etc.
- multiple appearances in the annually popular national content pieces and lists identifying the coming season’s hot-seat candidates.
- multiple mentions from local media in news clippings using terms like “hot seat,” “feeling the pressure,” “needs to win this year,” or similar terms within these contexts.
- reports from local media that important program stakeholders had lost faith in the coach.
- other obvious, reasonable indicators of discord within the tenure.
BLUE: SUSTAINING SUCCESS - the blue cells indicate the type of hire every university wants to make, a successful, sustaining presence who has stuck around so far.
GREEN: SUCCESSFUL AND HIRED AWAY - the green cells indicate that the given year of the tenure marked the coach’s last as he was hired away for a better job.
WHITE: GOOD STANDING, SAFE, BUT NOT YET A SUCCESSFUL, SUSTAINING PRESENCE - the uncolored cells featuring coaches’ names in the “TOTAL” tab indicate that a coach is currently “safe” and in good standing but not yet considered to be successful tenures.
* * *
There are a million ways to go with analyzing this data, but those will all be touched on in the updated study for 2018. For our sake, the issues we're trying to get to the bottom of, for now, are:
In the long-term ...
1) what difference is there between coaches who have WINNING (a greater percentage of wins than losses in the initial seasons of given tenure) seasons versus LOSING seasons?
2) what difference is there between coaches who MAKE A BOWL in initial season of given tenure and those who DO NOT MAKE A BOWL?
3) what difference is there, among coaches who attain bowl eligibility in initial season of tenure, between those who WIN IN THE BOWL GAME and those who LOSE IN THE BOWL GAME?
4) how does DIFFERENCE IN WIN TOTAL FROM LAST SEASON OF PREDECESSOR (positive or negative; and total deviation) factor in?
Year One Win Totals and the Hot Seat By Tenure Type
BLUE: SUSTAINING SUCCESS (ALL D1) - (21 tenures)
- 7.33 wins on average in year one
- median win total of 8
BLUE: SUSTAINING SUCCESS (POWER 5) - (15 tenures)
- 7.75 wins on average in year one
- median win total of 7.5
GREEN: SUCCESSFUL + HIRED AWAY (ALL D1) - (57 tenures)
- 6.18 wins on average in year one
- median win total of 6
GREEN: SUCCESSFUL + HIRED AWAY (POWER 5) - (16 tenures)
- 6.47 wins on average in year one
- median win total of 7
WHITE: GOOD STANDING (ALL D1) - (34 tenures)
- 5.26 wins on average in year one
- median win total of 5
WHITE: GOOD STANDING (POWER 5) - (14 tenures)
- 6 wins on average in year one
- median win total of 6.5
PINK: ON THE HOT SEAT (ALL D1) - (18 tenures)
- 6.33 wins on average in year one
- median win total of 7.5
PINK: ON THE HOT SEAT (POWER 5) - (12 tenures)
- 6.83 wins on average in year one
- median win total of 8
RED: FIRED (ALL D1) - (94 tenures)
- 4.51 wins on average in year one
- median win total of 4
RED: FIRED (ALL POWER 5) - (48 tenures)
- 5.15 wins on average in year one
- median win total of 5
Winning vs. Losing Initial Seasons and the Hot Seat by Tenure Type
BLUE: SUSTAINING SUCCESS (ALL D1) - (21 tenures)
71.4% had winning first seasons
28.6% had losing first seasons
BLUE: SUSTAINING SUCCESS (POWER 5) - (15 tenures)
80% had winning first seasons
20% had losing first seasons
GREEN: SUCCESSFUL + HIRED AWAY (ALL D1) - (57 tenures)
52.6% had winning first seasons
47.4% had losing first seasons
GREEN: SUCCESSFUL + HIRED AWAY (POWER 5) - (16 tenures)
62.5% had winning first seasons
37.5% had losing first seasons
WHITE: GOOD STANDING (ALL D1) - (34 tenures)
38.2% had winning first seasons
61.8% had losing first seasons
WHITE: GOOD STANDING (POWER 5) - (14 tenures)
42.9% had winning first seasons
57.1% had losing first seasons
PINK: ON THE HOT SEAT (ALL D1) - (18 tenures)
55.5% had winning first seasons
45.5% had losing first seasons
PINK: ON THE HOT SEAT (POWER 5) - (12 tenures)
66.7% had winning first seasons
33.3% had losing first seasons
RED: FIRED (ALL D1) - (94 tenures)
30% had winning first seasons
70% had losing first seasons
RED: FIRED (ALL POWER 5) - (48 tenures)
40.6% had winning first seasons
60.4% had losing first seasons
First-Year Bowl Game Eligibility and the Hot Seat by Tenure Type
BLUE: SUSTAINING SUCCESS (ALL D1) - (21 tenures)
62% attained a bowl-eligible record in first season (13 of 21)
BLUE: SUSTAINING SUCCESS (POWER 5) - (15 tenures)
80% attained a bowl-eligible record in first season (12 of 15)
GREEN: SUCCESSFUL + HIRED AWAY (ALL D1) - (57 tenures)
57.9% attained a bowl-eligible record in first season (34 of 57)
GREEN: SUCCESSFUL + HIRED AWAY (POWER 5) - (16 tenures)
75% attained a bowl-eligible record in first season (12 of 16)
WHITE: GOOD STANDING (ALL D1) - (34 tenures)
44.1% attained a bowl-eligible record in first season (15 of 34)
WHITE: GOOD STANDING (POWER 5) - (14 tenures)
50% attained a bowl-eligible record in first season (7 of 14)
PINK: ON THE HOT SEAT (ALL D1) - (18 tenures)
55.6% attained a bowl-eligible record in first season (10 of 18)
PINK: ON THE HOT SEAT (POWER 5) - (12 tenures)
67% attained a bowl-eligible record in first season (8 of 12)
RED: FIRED (ALL D1) - (94 tenures)
29.8% attained a bowl-eligible record in first season (28 of 94)
RED: FIRED (ALL POWER 5) - (48 tenures)
43.8% attained a bowl-eligible record in first season (21 of 48)
This is where we'll begin to unpack the data above and see how important winning the first-year bowl game is within its context. First, a few thoughts on the data above:
It's interesting to see the data regarding the coaches who the study pegged in 2017 as having come into the season on the hot seat. As for Year 1 win totals and percentage of tenures within the group who had "winning" first seasons, the only group who had relatively more success than the coaches on the 2017 hot seat in these metrics was the blue, sustaining-success/untouchable group.
This is likely an anomaly and has to do with the hot seat group in 2017 having an unusual number of coaches on the list who experienced above-average first-year success which has waned recently such as Gus Malzahn and Rod Carey (12 wins), Kevin Sumlin (11 wins), Jim Mora Jr. and Matt Wells (9 wins) along with Todd Graham, Kliff Kingsbury, Brian Kelly and Rich Rodriguez (8 wins).
Other than that, the numbers shake out about like you would think: The best group -- the group that all Texas fans hope that Tom Herman goes on to join -- had the most success of all other groups in Year 1. The best coaches, more often than the others, generally make it known in the first year that change is afoot in the win column. Even the ones that take over bad situations. Jim Harbaugh tallied 10 wins in Year 1 versus the 5 that Brady Hoke put up as his predecessor. Urban Meyer won 12 in his first year at Ohio State, a program Luke Fickell led to a 6-7 record one year prior. Jimbo Fisher turned a 7-win FSU team into a 10-win one. Mark D'Antonio took a previously 4-win club to bowl eligibility and seven wins and Brian Harsin broke through the 10-win barrier at Boise taking a previously 8-win group to 12 wins.
With this stated, not all coaches who have reached the blue "sustaining success" category coming into the 2017 season started out with pistols blazing. James Franklin only matched the win total of Bill O'Brien at his first season at Penn State (7) as did Chris Peterson at Washington (8), Ken Niaumatalolo (8) and Terry Bowden (1) .
Other coaches in the successful group who attained a win total similar to Herman relative to previous-year expectation (1-to-2 games better - Herman is currently one game above Strong's 5-game precedent from 2016 with a chance to make it a two-game improvement with a tough bowl win versus Missouri) are: Dana Holgorsen (10 wins, one-game improvement), Bill Snyder (second tenure) (six wins, one-game improvement), Nick Saban (7 wins, one-game improvement), Paul Johnson (9 wins, two-game improvement).
Texas fans can also point to the fact that numerous coaches have gone on to become successful, sustaining presences at their schools despite failing to reach a win total equal-to-or-greater-than their predecessors in Year 1. These include David Shaw (11 wins, one-game regression), Mike Leach (3 wins, one-game regression), Rocky Long (8 wins, one-game regression), Mike Gundy (4 wins, three-game regression), Pat Fitzgerald (4 wins, three-game regression) and Skip Holtz (4 wins, five-game regression). Relative to previous expectations, Herman has already performed better in Year 1 than all of these coaches without even having won a bowl yet. With this being said, the win expectations set by the predecessors of these coaches were all higher than the expectation Herman came in facing at Texas with the exception of Mike Leach at Washington State. In short, the bar for Herman to post improvement from was incredibly low at just 5 wins.
One of the more clear findings from the first portion of the study outlined above comes from the percentage of coaches who experienced "winning" first seasons. Tenures in the "successful" categories (the sustaining success and successful/hired away groups) are overwhelmingly made up of coaches who had better-than-.500 winning percentages in Year 1. Among Power 5 successful, sustaining coaches, Tom Herman's performance against Missouri in the Texas Bowl will determine whether he is on the right or wrong side of an 80%-20% split among winning and losing coaches in their first seasons.
While it can be said that coaches who came into the 2017 season in the "safe; good standing"-category only had winning seasons 43% of the time, we should note that those coaches (currently in good standing) will move into other categories at some point - be they the successful ones or the not-so-successful ones. Of all tenures examined, 112 coaches who were at some point in good standing have been fired or are currently on the hot seat while only 78 have moved from "safe" to "successful/hired away" or "sucessful/sustaining." This means that, of the coaches who are in the currently safe category, only roughly 40% will go on to successful outcomes while 60% will end up being fired. The "safe" category in the terms of this study is not really so safe in a literal sense.
So, we know that it's historically an important aspect of eventually successful coaches -- especially at Power 5 schools -- to start their given tenures with a winning record in Season 1, to finish with a better record than your predecessor posted in his year of firing, to post a win total in Year 1 of at least somewhere around the 7-win mark, and to gain bowl eligibility. In retrospect, all of these factors are overwhelmingly telling over a longitudinal study in predicting outcomes of tenures.
Herman currently has two of these four items in the bag: he's already beaten Charlie Strong's pathetic 2016 win total (albeit tied for the third-lowest mark any currently successful and sustaining head coach in P5 football had to best in his first season), and he's attained bowl eligibility. What's yet to be seen is if he can get into the 7-ish win category which separates him into a grouping much closer to the to the successful groups than the unsuccessful groups. To check those final two boxes, it's going to take winning the bowl game (which might be a box of its own to check) ...
First-Year Bowl Game Outcomes and the Hot Seat
Of the coaches who attained bowl-eligibility in their first seasons of tenure, 52% lost and 48% won. When looking at only the successful group(s) of coaches, we don't really get that much more insight. Of these, 55% won in their initial season's bowl game while 45% lost. Examining the unsuccessful tenures doesn't get us anywhere, either. In fact, of all coaches currently on the hot seat or fired who attained first-year bowl eligibility, 52% of those coaches actually won in their contests. It appears that, based on the data, the actual gaining of bowl eligibility is much more predictive than the win-loss outcome of the contest.
What about how those coaches won and/or lost these games, though?
Among all successful, sustaining tenures, the median margin of victory in the initial bowl game was 8 points while the median loss-margin was -5.5 points. When combining the successful, sustaining tenures with the successful/hired away tenures and examining the same numbers, median margin of victory among winners was 7 points while the median margin of loss was -9.5. This doesn't tell us anything on the surface, either.
A few things do stand out when looking a little deeper, though.
Only four coaches within the entire group of successful/sustaining tenures (19%) lost their Year 1 bowl games to an unranked opponent (Rocky Long, Chris Peterson, Ken Niumatalolo and Todd Johnson). Only two P5 coaches, Peterson and Johnson (13%) lost in Year 1 bowl games to unranked opponents. As we know, Missouri, like Texas, is not a team ranked in the Top 25 although it did receive a tiny three votes in the final AP standings. Losing to an unranked team in the first bowl appearance would put Tom Herman on the wrong side of yet another pretty big split.
But what truly can't happen is to get boat-raced.
Of the 93 total coaches (of all different tenure types), who attained bowl eligibility in Year 1 of the given tenure, only one successful, sustaining head coach was beaten by more than 10 points in the contest. Among P5 coaches, only 2 were beaten by more than a field goal. When filtering by margin-of-loss among all coaches, the landscape becomes littered with red and pink tenures and very little blue or green whereas only 6 of 20 coaches (30%) who started tenures with bowl victories of more than a touchdown saw (or were seeing at 2017's onset) unsuccessful outcomes.
For those who call the Texas Bowl "meaningless," it couldn't be further from the truth.
Not only does it give a largely inexperienced team -- one that will be battling back in 2018 from some unexpected and late-breaking attrition -- 15 critical extra practices and a chance to get real-game action to some new faces for purposes of evaluation. It also gives Tom Herman chance to check off a few very key remaining boxes among indicators of future success based on the 234 division I tenures examined.
On top of securing more wins than his immediate predecessor in the year the predecessor was fired and/or moved on and attain Year 1 bowl eligibility, a Texas bowl victory means:
- Getting to seven wins
(which would, in turn, mean)
- Finishing the season with a winning record
(which would, in turn, mean)
- Winning the bowl game
(which on its own is historically meaningless, but would, in turn, mean)
- Avoiding losing to an unranked opponent in the Year 1 bowl
(and)
- Avoiding getting boat-raced in the Year 1 bowl.
Depending on how you look at it, the Texas Bowl means everything.
So, in the interim, @Ketchum suggested I do a smaller study based on the data we currently have (from 2005 through the 2016 seasons) about just how important/unimportant an upcoming bowl win is to Tom Herman's likelihood of finding himself on a future hot seat, based on recent historical precedent set by the 234 tenures examined through the end of the 2016 season.
For those who are new to Orangebloods, or have not seen the highly controversial editions of "Charlie Strong and the Hot Seat" over the previous years (it's fun to go back through and read the comments in these threads calling the studies, which ended up being accurate in their predictions of exactly when Strong and countless others would be fired, "worthless garbage," etc.) here is a small explanation of the total study before we move forward on this small piece of the puzzle. Our focus for this content will be far-less expansive than the new study which will come out at some point in the spring of 2018, and more narrowly focused on a six-versus-seven-win season in Year 1 and the importance of winning that first bowl.
* * *
Explanation of the study
We’ve once again examined every Division I coaching hire that has been made from 2005 to, now 2016, in an attempt to better understand - and in some ways quantify - the abstract idea of the hot seat. The goal is to identify the circumstances that lead to both finding and avoiding it.
Here is a link to the data.
On how the data was gathered and organized:
- Each Division I coaching hire during the span was chronicled from the time of its start to its eventual end or the present - whichever was applicable.
- Each hire was categorized as either an offensive or defensive hire based on the head coach’s philosophical football background.
- Each year of each tenure is chronicled by W-L record within the horizontal cells leading to present day or date of tenure's last service. The cells are highlighted to note these six events, as are the coaches' name cells in the “New Coach“ tab.
RED: FIRED - the red cells indicate that the head-coaching tenure was ended due to firing in the corresponding year. If a red cell is not preceded by a pink cell, it means that the coach’s seat did not become hot as detected by the Deep Dig until the year he was fired.
PINK: ON THE HOT SEAT - the pink cells indicate that the given year within the tenure was either spent in, or led to, hot-seat status. This abstract variable was assigned via research on every tenure examined. Typical events that led to “hot seat”-assignment were items such as:
- public responses from athletic directors or other university figureheads asserting the coach in question would indeed be back next season, public ‘votes of confidence,” etc.
- multiple appearances in the annually popular national content pieces and lists identifying the coming season’s hot-seat candidates.
- multiple mentions from local media in news clippings using terms like “hot seat,” “feeling the pressure,” “needs to win this year,” or similar terms within these contexts.
- reports from local media that important program stakeholders had lost faith in the coach.
- other obvious, reasonable indicators of discord within the tenure.
BLUE: SUSTAINING SUCCESS - the blue cells indicate the type of hire every university wants to make, a successful, sustaining presence who has stuck around so far.
GREEN: SUCCESSFUL AND HIRED AWAY - the green cells indicate that the given year of the tenure marked the coach’s last as he was hired away for a better job.
WHITE: GOOD STANDING, SAFE, BUT NOT YET A SUCCESSFUL, SUSTAINING PRESENCE - the uncolored cells featuring coaches’ names in the “TOTAL” tab indicate that a coach is currently “safe” and in good standing but not yet considered to be successful tenures.
* * *
In the long-term ...
1) what difference is there between coaches who have WINNING (a greater percentage of wins than losses in the initial seasons of given tenure) seasons versus LOSING seasons?
2) what difference is there between coaches who MAKE A BOWL in initial season of given tenure and those who DO NOT MAKE A BOWL?
3) what difference is there, among coaches who attain bowl eligibility in initial season of tenure, between those who WIN IN THE BOWL GAME and those who LOSE IN THE BOWL GAME?
4) how does DIFFERENCE IN WIN TOTAL FROM LAST SEASON OF PREDECESSOR (positive or negative; and total deviation) factor in?
* * *
Year One Win Totals and the Hot Seat By Tenure Type
BLUE: SUSTAINING SUCCESS (ALL D1) - (21 tenures)
- 7.33 wins on average in year one
- median win total of 8
BLUE: SUSTAINING SUCCESS (POWER 5) - (15 tenures)
- 7.75 wins on average in year one
- median win total of 7.5
GREEN: SUCCESSFUL + HIRED AWAY (ALL D1) - (57 tenures)
- 6.18 wins on average in year one
- median win total of 6
GREEN: SUCCESSFUL + HIRED AWAY (POWER 5) - (16 tenures)
- 6.47 wins on average in year one
- median win total of 7
WHITE: GOOD STANDING (ALL D1) - (34 tenures)
- 5.26 wins on average in year one
- median win total of 5
WHITE: GOOD STANDING (POWER 5) - (14 tenures)
- 6 wins on average in year one
- median win total of 6.5
PINK: ON THE HOT SEAT (ALL D1) - (18 tenures)
- 6.33 wins on average in year one
- median win total of 7.5
PINK: ON THE HOT SEAT (POWER 5) - (12 tenures)
- 6.83 wins on average in year one
- median win total of 8
RED: FIRED (ALL D1) - (94 tenures)
- 4.51 wins on average in year one
- median win total of 4
RED: FIRED (ALL POWER 5) - (48 tenures)
- 5.15 wins on average in year one
- median win total of 5
* * *
Winning vs. Losing Initial Seasons and the Hot Seat by Tenure Type
BLUE: SUSTAINING SUCCESS (ALL D1) - (21 tenures)
71.4% had winning first seasons
28.6% had losing first seasons
BLUE: SUSTAINING SUCCESS (POWER 5) - (15 tenures)
80% had winning first seasons
20% had losing first seasons
GREEN: SUCCESSFUL + HIRED AWAY (ALL D1) - (57 tenures)
52.6% had winning first seasons
47.4% had losing first seasons
GREEN: SUCCESSFUL + HIRED AWAY (POWER 5) - (16 tenures)
62.5% had winning first seasons
37.5% had losing first seasons
WHITE: GOOD STANDING (ALL D1) - (34 tenures)
38.2% had winning first seasons
61.8% had losing first seasons
WHITE: GOOD STANDING (POWER 5) - (14 tenures)
42.9% had winning first seasons
57.1% had losing first seasons
PINK: ON THE HOT SEAT (ALL D1) - (18 tenures)
55.5% had winning first seasons
45.5% had losing first seasons
PINK: ON THE HOT SEAT (POWER 5) - (12 tenures)
66.7% had winning first seasons
33.3% had losing first seasons
RED: FIRED (ALL D1) - (94 tenures)
30% had winning first seasons
70% had losing first seasons
RED: FIRED (ALL POWER 5) - (48 tenures)
40.6% had winning first seasons
60.4% had losing first seasons
* * *
First-Year Bowl Game Eligibility and the Hot Seat by Tenure Type
BLUE: SUSTAINING SUCCESS (ALL D1) - (21 tenures)
62% attained a bowl-eligible record in first season (13 of 21)
BLUE: SUSTAINING SUCCESS (POWER 5) - (15 tenures)
80% attained a bowl-eligible record in first season (12 of 15)
GREEN: SUCCESSFUL + HIRED AWAY (ALL D1) - (57 tenures)
57.9% attained a bowl-eligible record in first season (34 of 57)
GREEN: SUCCESSFUL + HIRED AWAY (POWER 5) - (16 tenures)
75% attained a bowl-eligible record in first season (12 of 16)
WHITE: GOOD STANDING (ALL D1) - (34 tenures)
44.1% attained a bowl-eligible record in first season (15 of 34)
WHITE: GOOD STANDING (POWER 5) - (14 tenures)
50% attained a bowl-eligible record in first season (7 of 14)
PINK: ON THE HOT SEAT (ALL D1) - (18 tenures)
55.6% attained a bowl-eligible record in first season (10 of 18)
PINK: ON THE HOT SEAT (POWER 5) - (12 tenures)
67% attained a bowl-eligible record in first season (8 of 12)
RED: FIRED (ALL D1) - (94 tenures)
29.8% attained a bowl-eligible record in first season (28 of 94)
RED: FIRED (ALL POWER 5) - (48 tenures)
43.8% attained a bowl-eligible record in first season (21 of 48)
* * *
This is where we'll begin to unpack the data above and see how important winning the first-year bowl game is within its context. First, a few thoughts on the data above:
It's interesting to see the data regarding the coaches who the study pegged in 2017 as having come into the season on the hot seat. As for Year 1 win totals and percentage of tenures within the group who had "winning" first seasons, the only group who had relatively more success than the coaches on the 2017 hot seat in these metrics was the blue, sustaining-success/untouchable group.
This is likely an anomaly and has to do with the hot seat group in 2017 having an unusual number of coaches on the list who experienced above-average first-year success which has waned recently such as Gus Malzahn and Rod Carey (12 wins), Kevin Sumlin (11 wins), Jim Mora Jr. and Matt Wells (9 wins) along with Todd Graham, Kliff Kingsbury, Brian Kelly and Rich Rodriguez (8 wins).
Other than that, the numbers shake out about like you would think: The best group -- the group that all Texas fans hope that Tom Herman goes on to join -- had the most success of all other groups in Year 1. The best coaches, more often than the others, generally make it known in the first year that change is afoot in the win column. Even the ones that take over bad situations. Jim Harbaugh tallied 10 wins in Year 1 versus the 5 that Brady Hoke put up as his predecessor. Urban Meyer won 12 in his first year at Ohio State, a program Luke Fickell led to a 6-7 record one year prior. Jimbo Fisher turned a 7-win FSU team into a 10-win one. Mark D'Antonio took a previously 4-win club to bowl eligibility and seven wins and Brian Harsin broke through the 10-win barrier at Boise taking a previously 8-win group to 12 wins.
With this stated, not all coaches who have reached the blue "sustaining success" category coming into the 2017 season started out with pistols blazing. James Franklin only matched the win total of Bill O'Brien at his first season at Penn State (7) as did Chris Peterson at Washington (8), Ken Niaumatalolo (8) and Terry Bowden (1) .
Other coaches in the successful group who attained a win total similar to Herman relative to previous-year expectation (1-to-2 games better - Herman is currently one game above Strong's 5-game precedent from 2016 with a chance to make it a two-game improvement with a tough bowl win versus Missouri) are: Dana Holgorsen (10 wins, one-game improvement), Bill Snyder (second tenure) (six wins, one-game improvement), Nick Saban (7 wins, one-game improvement), Paul Johnson (9 wins, two-game improvement).
Texas fans can also point to the fact that numerous coaches have gone on to become successful, sustaining presences at their schools despite failing to reach a win total equal-to-or-greater-than their predecessors in Year 1. These include David Shaw (11 wins, one-game regression), Mike Leach (3 wins, one-game regression), Rocky Long (8 wins, one-game regression), Mike Gundy (4 wins, three-game regression), Pat Fitzgerald (4 wins, three-game regression) and Skip Holtz (4 wins, five-game regression). Relative to previous expectations, Herman has already performed better in Year 1 than all of these coaches without even having won a bowl yet. With this being said, the win expectations set by the predecessors of these coaches were all higher than the expectation Herman came in facing at Texas with the exception of Mike Leach at Washington State. In short, the bar for Herman to post improvement from was incredibly low at just 5 wins.
One of the more clear findings from the first portion of the study outlined above comes from the percentage of coaches who experienced "winning" first seasons. Tenures in the "successful" categories (the sustaining success and successful/hired away groups) are overwhelmingly made up of coaches who had better-than-.500 winning percentages in Year 1. Among Power 5 successful, sustaining coaches, Tom Herman's performance against Missouri in the Texas Bowl will determine whether he is on the right or wrong side of an 80%-20% split among winning and losing coaches in their first seasons.
While it can be said that coaches who came into the 2017 season in the "safe; good standing"-category only had winning seasons 43% of the time, we should note that those coaches (currently in good standing) will move into other categories at some point - be they the successful ones or the not-so-successful ones. Of all tenures examined, 112 coaches who were at some point in good standing have been fired or are currently on the hot seat while only 78 have moved from "safe" to "successful/hired away" or "sucessful/sustaining." This means that, of the coaches who are in the currently safe category, only roughly 40% will go on to successful outcomes while 60% will end up being fired. The "safe" category in the terms of this study is not really so safe in a literal sense.
So, we know that it's historically an important aspect of eventually successful coaches -- especially at Power 5 schools -- to start their given tenures with a winning record in Season 1, to finish with a better record than your predecessor posted in his year of firing, to post a win total in Year 1 of at least somewhere around the 7-win mark, and to gain bowl eligibility. In retrospect, all of these factors are overwhelmingly telling over a longitudinal study in predicting outcomes of tenures.
Herman currently has two of these four items in the bag: he's already beaten Charlie Strong's pathetic 2016 win total (albeit tied for the third-lowest mark any currently successful and sustaining head coach in P5 football had to best in his first season), and he's attained bowl eligibility. What's yet to be seen is if he can get into the 7-ish win category which separates him into a grouping much closer to the to the successful groups than the unsuccessful groups. To check those final two boxes, it's going to take winning the bowl game (which might be a box of its own to check) ...
First-Year Bowl Game Outcomes and the Hot Seat
Of the coaches who attained bowl-eligibility in their first seasons of tenure, 52% lost and 48% won. When looking at only the successful group(s) of coaches, we don't really get that much more insight. Of these, 55% won in their initial season's bowl game while 45% lost. Examining the unsuccessful tenures doesn't get us anywhere, either. In fact, of all coaches currently on the hot seat or fired who attained first-year bowl eligibility, 52% of those coaches actually won in their contests. It appears that, based on the data, the actual gaining of bowl eligibility is much more predictive than the win-loss outcome of the contest.
What about how those coaches won and/or lost these games, though?
Among all successful, sustaining tenures, the median margin of victory in the initial bowl game was 8 points while the median loss-margin was -5.5 points. When combining the successful, sustaining tenures with the successful/hired away tenures and examining the same numbers, median margin of victory among winners was 7 points while the median margin of loss was -9.5. This doesn't tell us anything on the surface, either.
A few things do stand out when looking a little deeper, though.
Only four coaches within the entire group of successful/sustaining tenures (19%) lost their Year 1 bowl games to an unranked opponent (Rocky Long, Chris Peterson, Ken Niumatalolo and Todd Johnson). Only two P5 coaches, Peterson and Johnson (13%) lost in Year 1 bowl games to unranked opponents. As we know, Missouri, like Texas, is not a team ranked in the Top 25 although it did receive a tiny three votes in the final AP standings. Losing to an unranked team in the first bowl appearance would put Tom Herman on the wrong side of yet another pretty big split.
But what truly can't happen is to get boat-raced.
Of the 93 total coaches (of all different tenure types), who attained bowl eligibility in Year 1 of the given tenure, only one successful, sustaining head coach was beaten by more than 10 points in the contest. Among P5 coaches, only 2 were beaten by more than a field goal. When filtering by margin-of-loss among all coaches, the landscape becomes littered with red and pink tenures and very little blue or green whereas only 6 of 20 coaches (30%) who started tenures with bowl victories of more than a touchdown saw (or were seeing at 2017's onset) unsuccessful outcomes.
* * *
For those who call the Texas Bowl "meaningless," it couldn't be further from the truth.
Not only does it give a largely inexperienced team -- one that will be battling back in 2018 from some unexpected and late-breaking attrition -- 15 critical extra practices and a chance to get real-game action to some new faces for purposes of evaluation. It also gives Tom Herman chance to check off a few very key remaining boxes among indicators of future success based on the 234 division I tenures examined.
On top of securing more wins than his immediate predecessor in the year the predecessor was fired and/or moved on and attain Year 1 bowl eligibility, a Texas bowl victory means:
- Getting to seven wins
(which would, in turn, mean)
- Finishing the season with a winning record
(which would, in turn, mean)
- Winning the bowl game
(which on its own is historically meaningless, but would, in turn, mean)
- Avoiding losing to an unranked opponent in the Year 1 bowl
(and)
- Avoiding getting boat-raced in the Year 1 bowl.
Depending on how you look at it, the Texas Bowl means everything.
