Tags
With recent news that the Thunder have focused their search on Florida coach Billy Donovan, it is worth reviewing his NCAA performance to assess his upside as an NBA coach.
Donovan peaked when his Gators won back to back championships in 2006 and 2007 with the elite core of Joakim Noah, Al Horford, and Corey Brewer. Of course any coach would excel with that trio staying until their junior seasons, so it is more interesting to assess how he fared following their departure:
| Season | final kenpom rank | Recruiting class rank |
| 2015 | 45 | 13 |
| 2014 | 3 | 6 |
| 2013 | 2 | 24 |
| 2012 | 11 | 24 |
| 2011 | 16 | 9 |
| 2010 | 48 | 20 |
| 2009 | 42 | 2 |
| 2008 | 44 | 3 |
It is fair to expect a bit of a down year following a max exodus of talent, but given the quality of talent he brought in with perennial top 25 recruiting classes four seasons outside of the top 40 is rather alarming.
His loaded 07-08 recruiting class included current NBA players Nick Calathes and Chandler Parsons and he also returned sophomore Marreese Speights. Succeeding with youth can be difficult, but it is not like he was working with a bunch of talentless duds. Meanwhile, Florida failed to reach the NCAA tournament with three future NBA players.
The following season Florida lost Speights to the NBA but returned Calathes and Parsons for their sophomore seasons (along with #53 RSCI player Alex Tyus from the same class) and with another loaded recruiting class still missed the tournament.
In 2009-10 Florida finally returned to the tournament as a 10 seed in spite of Calathes departing for Europe, but still could not crack the top 40. It was not until Parsons’s senior season that Florida finally returned to the top 25 making an Elite 8 run before losing to Brad Stevens’s Butler squad.
The whole Parsons era looks bad for Donovan’s record– in four full seasons with the player and plenty of access to other elite talent Florida only cracked the top 40 once and never made the top 15. While the #16 team that went Elite 8 may seem like it saves some face, that team was loaded with experienced talent. The starting lineup:
| Player | RSCI rank | Season | PER | WS/40 |
| Vernon Macklin | 16 | Senior | 22.2 | 0.16 |
| Erving Walker | 79 | Junior | 20 | 0.177 |
| Chandler Parsons | 39 | Senior | 19.7 | 0.163 |
| Alex Tyus | 53 | Senior | 18.7 | 0.145 |
| Kenny Boynton | 9 | Sophomore | 17.5 | 0.155 |
This is the type of talent that a good coach should be able to convert into a top 5 contender. Forget that Parsons is the only NBA player on the roster– in spite of being a senior he completely blends in with the rest of the starters statistically. Not only did Donovan underwhelm with a collection of upperclass 4 and 5 star talent, but he also failed to maximize Parsons’s talent. His value as a second round steal was evident from his rookie season where he finished 9th in ROY voting and 5th in rookie win shares. It is not difficult to envision a superior coach making Parsons an obvious first round selection after four seasons.
The counterpoint this is that coaches improve, and the Gators certainly took a leap forward after Parsons’ departure. In 11-12, Florida replaced their three departing senior starters with freshman Bradley Beal, sophomore Patric Young, and junior Erik Murphy and finished 11th and once again lost in the Elite 8, this time to Louisville.
The appealing part of Donovan’s post-repeat resume comes in the 12-13 and 13-14 seasons, as Florida finished with back to back top 3 kenpom teams without a single NBA draft pick on the roster. He did it with defense: After ranking 123rd, 113th, 91st, 41st, and 90th in schedule adjusted defensive rating in the prior five seasons, Florida leapt forward to the 4th and 2nd defenses. They did this with a middling block rate as well, as the defense delivered a strong balance of forcing opponents into difficult mid-range shots, forcing turnovers, rebounding, and limiting free throw attempts.
The timing of the spike in defense is no coincidence either, as it coincided with Donovan becoming analytically enlightened and placing an increased emphasis on defense. That said, the article does not cite any example of Donovan using analytics to improve the defense. It only states that analytics were employed to bring Donovan to the obvious realization that defense is important, so there is little evidence to suggest that he developed into a defensive mastermind. If anything the writeup implies that Donovan takes an overwhelmingly simple minded approach to analytics, focusing on the result rather than the process.
The common link to the great defensive teams was the elite 2014 graduating class of Patric Young, Will Yeguette, Casey Prather, and Scottie Wilbekin. While none of them were noteworthy NBA prospects by the time they graduated (Young had lottery hype as a sophomore), they were all high quality NCAA players who made strong contributions on both ends of the floor.
Give Donovan credit for maximizing returns on the talent of his class of unheralded college stars, but he was unable to sustain any of their success once they left. This was not for lack of talent either. In 2014-15, Florida was ranked 7th in both the AP and coaches pre-season polls, projected to finish 7th by kenpom and team rankings projection models and 6th by Dan Hanner’s model. The team suffered their share of bad luck, especially with injuries. #31 RSCI freshman Brandone Francis missed the entire season due to academic ineligibility, rotation guard Eli Carter missed 5 games, and Florida’s two best players Michael Frazier and Dorian Finney-Smith missed 7 and 5 games respectively. But injuries were not all that went wrong: 5 star sophomores Chris Walker and Kasey Hill both had vastly disappointing seasons. Hill regressed statistically from his freshman season and Walker fell from a possible lottery pick to likely undrafted after finally getting consistent playing time. Further neither #20 RSCI freshman Devin Robinson, #49 freshman Chris Chiozza, or Duke transfer Alex Murphy were able to play well enough to atone for their deficiencies.
Donovan’s newfound defensive obsession enabled the Gators to maintain the #11 kenpom defense, but their offense plunged to #151 after ranking #12 and #18 in the prior two seasons. This supports the narrative that Donovan merely gives most of his attention to defense rather than having concocted a strong defensive scheme. There is no reason for the offense to be so woefully bad other than punting on that end to meet arbitrary defensive benchmarks that Donovan deems necessary to contend. A good coach should aim to balance both ends for the optimal bottom line, not go all in on one side of the ball no matter the cost to the other.
Florida had the depth and talent to sustain a bit of injury misfortune and still have a top 25 team, but they finished as the #45 kenpom team and missed the tournament with a below .500 record. It is as if the Parsons era was starting to repeat itself all over again– Donovan gets acclaim for thriving off an exceptional class and then fails to maximize his incoming talent and disappoints. This is why he is a middle of the road NCAA coach: when he has the right collection of players he can thrive, but he is also prone to big disappointment in other scenarios where it is easy to envision a great coach excelling.
It is somewhat baffling that any NBA team is interested in Donovan after his most recent season, let alone the team with the best collection of young talent in the league. Dipping into NCAA ranks for a coach with elite recruiting pedigree and average coaching acumen is a tried and true formula for NBA coaching failure, and there is no reason to expect Donovan to buck the trend. This is especially true with the recent influx of strong NBA coaches, as the bar for good NBA coaching gets higher every season. While there are some legitimately good NBA coaching prospects in the NCAA (namely Fred Holberg), Donovan does not even sniff the radar as a coach with upside. Perhaps he can learn to use analytics to a more useful degree and avoid being a colossal failure, especially with a lackluster predecessor in Scott Brooks. But there is no way that of all of the coaching candidates in the basketball world, he is the candidate most likely to succeed at one of the most desirable NBA coaching jobs.
Dean On Draft 
I will point out that Noah, Brewer, and Horford would underrated coming out of high school as none were top 100 recruits. Recruits tend to go through the same guesswork that drafting does. So I can’t blame if his teams underperformed because they made more hype one year than the next. Young players are unpredictable so rating them out of high school is just guesswork. Selection and outcome bias are apt here. Presti gets so much credit for being lucky. Seattle would have drafted Oden if they had the number 1 pick. They lucked into Ibaka, Westbrook, and even Harden, who was good in college. The bad process is finally starting to catch up to them. Durant might be warming up to the thought of Donovan as coach.
http://basketball.realgm.com/wiretap/237610/Billy-Donovan-Interested-In-Meeting-With-Durant-Westbrook-Ibaka-Before-Accepting-Job
It doesn’t matter that they weren’t top 100 recruits, those guys gave him lifetime access to top 100 guys and he has not parlayed this talent into many great teams. That class reeks of a blind dog finding multiple bones.
IMO you gotta give Thunder credit for getting Harden, Westbrook, and especially Ibaka picks right but Presti has been pretty bad at things outside of drafting.
But does acquiring Perkins instead of using that money to resign Harden, Trading for horrible players in Dion Waiters or Enes Kanter(who has been great for them but that suffers from outcome bias), or not pulling the trigger on getting Tyson Chandler(who had injury issues so that was valid) give you any confidence in his player evaluating ability? I’m not so sure he is a great as he is given credit for. They wouldn’t have fired Brooks if Durant was healthy because they would be contending so even he is falling for a bit of the bias. I agree that there is much of a difference between the two coaches because I think most coaches are the same for the most part in regards to impact on wins. I would have loved for them to get Houston’s D-league coach, Nevada Smith or Mike D’Antoni.
I agree. I am a big doubter of Presti as a GM. But he has made a number of other good picks: Reggie Jackson, Mitch McGary, and Steven Adams were all value. It is likely at least partially due to luck but who knows, maybe he has a great scouting staff or something. It could be 100% luck but it is more likely that the Thunder are at least doing something right with respect to drafting and development.
Alvin Gentry? OKC need an offense since 2012.
Not a fan of hiring retreads but I would definitely take Gentry over Donovan.
Your argument about Donovan underachieving or failing to maximize talent isn’t persuasive in part because it’s clear you’d come to a conclusion about Donovan prior to actually writing this article. This is obvious because you choose to ignore information that cuts against your argument.
Take 538’s analysis where SRS and seeding are used to estimate tournament wins.* They essentially chart positive residuals from the model to see who outperforms expectations, and Donovan looks pretty good by this metric. Of course this analysis is limited only to tournament performance, but it’s still much wider in scope than the lazy analysis posted here.
Analysis of the NBA is obviously data driven, and nobody is going to take an argument that begins with “well let’s discount influential and important data points because they fail to support this argument I’m positing about” seriously. FWIW, I’m skeptical of the Donovan hire, but this sort of poorly constructed argument against it isn’t effective. People will read and respect your work significantly more if you carefully consider all the available data before reaching a conclusion.
*http://fivethirtyeight.com/datalab/tom-izzo-is-the-best-coach-in-modern-ncaa-tournament-history-by-far/
Who is to say that I did not carefully consider all of the available data?
Yeah I came to a conclusion before writing the article because I bet college basketball for a living and I have fairly confident opinions about what is what in that world. But it isn’t because I dislike Florida or Donovan at all (FWIW I bet them quite a bit in 13-14), it is because he’s just an average NCAA coach based on my assessment of available data.
I am not 538. They are data people, I am an intuitive guy. You can love that article all you want but it also ranks Pitino and Calipari above Donovan and they were colossal failures as NBA coaches. I would not expect it to be predictive at all.
Nice replacement for Florida. Plays at a fast pace and with a school like Florida, he should be able to attract better players, in theory. I wonder if going to a bigger school is going to make him more risk averse thus playing slower. Would love to see the coach at VMI get a big time coaching job just to see his high pace playing style, a style that players love saying that they want, in college.
We need a new big board.