clock menu more-arrow no yes mobile

Filed under:

Data viz–based analysis: Performance of sophomore football coaches

Using Win Pct. and SRS I chart the 2nd Year Performances of College HC

NCAA Football: California at UCLA Kirby Lee-USA TODAY Sports

On January 8th, 2017, the AD Mike Williams announced the firing of the fourth-year HC Sonny Dykes, ending the era of high flying Bear Raid football for the California Golden Bears. With a roster filled with talent (Demetris Robertson, Devante Downs, James Looney, Darius Allensworth, Tre Watson, and Vic Enwere) and more questions (None of the projected 2017 QBs took meaningful gametime snaps, the defense as a whole continued to appear on the wrong side of the NCAA record books, poor attendance), the Cal Athletics Department went out to search for a new head coach who would usher in a new era of Cal football.

Less than a week later, Justin Wilcox was tapped by the California athletic department as the 20th Head Coach of the California Golden Bears. In his first year, he improved the defense from a historically poor one to an average one. His team was able to find a role for the highly touted Devante Downs and was able to turn a walk-on Haas Business Major Patrick Laird (MVP) into a legitimate RB who is a threat in the running and passing game. A team that was expected to win 3–4 games (per 2017 S&P+ projections) managed to exceed expectations with 5 wins (including a hallmark win against a #8 WSU).

Now we’re eagerly expecting what will happen to the team in 2018. Due to the low turnover rate of the players and coaches (HC, OC, and DC remain the same) we are expecting a much better performance in 2018. The ESPN FPI has Cal ranked 27th in the nation with a projected 7.7–4.4 record.

I took a look at the performance of coaches in their second seasons in the BCS era 1998–2016. This will take a simplistic view of the matter by simply looking at win percentages and SRS (Simple Ranking System) values for each team with a HC’s 1st and 2nd years.

All of the FBS by Division

In the Stanfurd Cardinal, we can see teams that have won less in their second year than first. Overall, we can see that there are quite a few teams that in their sophomore seasons have gotten worse whether it is due to their scheme being figured out, their starters transferring out due to the new regime, or the variance of football.

What we can see from this chart is the fact that only a handful of team make huge leaps in SRS from Year 1 to 2; most regress to the mean of 0 or are within reasonable variance of their Year 1 performance. In 2017 Cal recorded a SRS of -0.16 which corresponds to Rich Brooks’ first season at Kentucky in 2003 (in 2004 the Wildcats went 2–9). It is also close to Jim McElwain’s first year at Colorado State in 2013 (in 2014, CSU went 10–2 under McElwain after which he was hired at Florida), and Les Miles’ 2001 season at Oklahoma State (in 2002, OkState went 8–4).

-0.16 is usually a ranking of an average team in its Year 2 of a new coach; in Cal’s situation of being both a team with a first-time starting QB and a major change in defensive front-7 scheme, this is a more than admirable result. However, it can also mean that the potential for improvement can also be limited since less than half of the teams go above 0 SRS in Year 2.

Density Curve

We can see here that the distribution of the wins in Year 1 and Year 2. We can see that statistically speaking these teams improve, smoothing out the curve where teams are roughly even change to go 1 to 7 wins in Year 2.

However, from a performance standpoint, it is much harder to improve the team from an SRS perspective. Most of the teams remain below average despite the improvement. Which is reasonable due to the fact that teams that were sub .500 have a lot of space to improve and will need more time to improve. Cal was -0.16 which is already ahead of schedule since that is slightly above the mean for a team in Year 2.


In the Pac-12/Pac-10, Arizona has had the worst luck out of the Pac-12 teams. USC had four coaching changes; however, all of the new HCs (Carroll, Kiffin, Sarkesian, Helton) were .500 or above in their first seasons. Likewise, Stanfurd doesn’t have David Shaw’s data. We can see that most of the time, coaches improve their teams; however, only three teams went over .500 in Year 2: Dirk Koetter’s 2002 ASU, Steve Sarkesian’s 2009 Washington, and Rick Neuheisel’s 2009 UCLA team.

We can see that most of the teams improved their on-field performance despite the lack of > .500 Year 2 performance. This is most likely the combination of a though Pac-12 as well as the Pac-12 teams scheduling tough OOC teams.


Note : click on the picture to zoom in, Cal is 4th from the left in row 2.

Overall, Cal’s two HCs performed above board improving both ways. With Sonny Dykes beginning 1–11, it was easy to improve since Cal always faced FCS opponents. Across the board, Cal stands out.

We can also see that the absolute ceiling of a five-win team is ~11 wins (Darrell Hazel at Kent State in 2013) with a floor of 1 win (One of whom was Stanfurd’s Walt Harris in 2006).

Note : click on the picture to zoom in, Cal is 4th from the left in row 2.

What happens is that Cal never got above average in the SRS rankings. The best improvement was Arkansas’s to an SRS of 15. Overall the pattern we can see is either improvement or stagnation.