Kentucky Football: The Numbers Predict The Season: 2016 Version

The previous two years saw reviews of preseason prognostications for the upcoming football season. Projecting outcomes in football using statistics is notoriously wonky given the reality of a small sample size relative to something like baseball (12 games versus 180 games).

Nonetheless, recent advancements have helped close the gap between a sea of numbers and comfortably reliable narratives. While they haven't predicted UK's final ranking, they have gotten close to nailing the eventual record.

This year's version will focus on S&P+, FPI, and Pythagorean projection. Phil Steele was cut for time.

S&P+

This system is one half of the F+ advanced statistical rankings used during the season for game previews. In the 2014 preview, S&P+ projected UK to have the 75th best team in the country. The team ended up finishing ranked 58th overall at 5-7. The system undervalued the team overall, but got certain aspects correct such as the offense's efficiency relative to the defense's efficiency.

Last preseason, the model predicted UK would finish ranked 52nd which roughly translated to the five-six win range. Again, the model got the win total correct, but UK ended up finishing 95th. Thanks, bilious offense.

This preseason, S&P+ projects UK in 83rd place seemingly due to returning production (more on that below). UK is projected to finish last in the SEC, and is 14 spots below Vanderbilt (which coincidentally finished last season ranked 83rd). Going off 2015's final rankings, an 83rd place finish projects between 4-6 wins.

FPI

In 2015, ESPN's FPI projected UK to be the 48th best team in the country which is a step up from 2014 when it ranked UK 54th overall. To review, both S&P+ and FPI both thought UK would be slightly better in 2015 than in 2014.

Specifically in 2015, FPI projected that against a perfectly average defense, UK's offense will add 2.5 points per game, while the defense will add 3.1 points per game.  FPI's year-end rankings are not available, but we know UK greatly underplayed both of those projections; however, UK's offense was actually expected to do worse (3.6 points per game in 2014), while the defense was expected to improve (0.8 points per game last season). FPI got the team dynamics correct even if the scale appears to be off.

This preseason, FPI projects UK to finish 56th and win between five to seven games. This is a rosier picture than S&P+, consistent with how FPI has viewed UK relative to S&P+ the last two seasons, but FPI also projects UK to finish last in the SEC. Team efficiencies have yet to be released.

Pythagorean Projection

The Pythagorean projection formula suggested last summer UK was poised to win between five to six games in 2015 given the fact UK played well enough to win 5.5 games in 2014. Teams that win one less game per season than their projection tend to improve the following year in the NFL -- especially if their record was near .500. UK didn't lose one less game, but was close by this model's standards and was essentially a .500 team. It was a small sign of optimism for 2016: UK was nearly good enough for six wins in 2014, and so could be on the upswing in 2015.

In 2016, the projection is bleaker. This model suggests UK played well enough for 5.3 wins last season. So basically, UK won as many games as it should have last season. That's not encouraging news for 2016, according to Pythagoras.

Oddness

The models project UK to be worse this season in terms of rankings, but UK does remain within the bounds of P5 teams that won between 4-7 games in 2015. That wide range brings as much hand-wringing as it does promise.  These rankings surprise few.

Last preseason the models projected UK would probably win six games and make a bowl game. They had UK pretty well pegged. Recall last off-season: UK had momentum and the Stoops regime appeared one season ahead of schedule. UK returned 16 starters, good recruits were a year older, the 'Cats made a splashy offensive coordinator hire, and finished 1-2 in one-score games

Going by those same standards it's not entirely unreasonable, but still a tad odd, to rank UK last in the SEC this preseason. Why is it odd? Because the environment is similar to last preseason when the models thought more highly of the 'Cats. Just as last preseason, UK returns 16 starters, good recruits are another year older, and the 'Cats have made another splashy offensive coordinator hire. UK was 4-2 in one-score games in 2015 (Stoops is 5-6 overall at UK), which one could argue shows increased resilience, but even points scored for and against wasn't exactly a great predictor for how the SEC East fared in 2015.

Conclusion

So what gives? There is likely a perceived lack of momentum. The 2014 successes were probably overhyped, and now 2016 expectations are a victim as a result.

[Another argument is the SEC East will be better this year, and UK will get left behind. Maybe, maybe, for a specific mid-tier SEC East program, but uniformly and across-the-board true? That's difficult to believe for reasons ranging between relative talent accumulated over the previous four signing classes, to an assumption UK remains static while others improve, to UK hosting pivotal opponents in 2016. We can expand further this summer. For now let's, er, punt.]

Specifics: the 2014  defense scored six defensive touchdowns which artificially boosted the scoring numbers making Neal Brown's final offense appear better than it was; meanwhile, the 2014 team also had an strong +8 turnover margin which helped to erase margins of loss. Looking at key metrics like yards allowed or third down conversions illustrates the 2014 team caught some breaks to get to five wins.

The 2015 team didn't have some of the same advantages. The turnover margin last season was -2 due to an offense committing seven more turnovers than the previous year. That's not to mention the defense allowed four fewer points per game despite a boom-or-bust offense keeping them on the field for long periods. UK still won more close games than any season under Stoops. There's a valid argument the 2015 team dispenses the 2014 team despite identical records.

Unlike the last preseason, the expectations pendulum has gone too far in the opposite direction. The most athletic players ever signed to UK will reach upperclassmen status this fall. This should be a time for excitement, and yet the projections seem to agree with the zeitgeist of a wide swath of BBN despite the commonalities between this preseason and the last.