Nowadays analytics are getting more and more popular in college basketball and predictive analytics that used to be used by the most forward-thinking coaching staffs are now commonplace in discussion among fans, such as the NCAA’s NET ranking or KenPom, a metric that gets referenced nearly as much as a team’s record. ESPN’s strength of record, Sagarin, Haslametrics, and Massey ratings have all gotten more and more play recently and there’s a good reason why—they help contextualize just how good a team is in the scheme of over 350 division one teams that would never have the chance to play each other. Predictive metrics can give you much more information than simply a team’s ranking because it accounts for things such as strength of schedule as well as how good or bad a team performed in a win or a loss.
While the metrics referenced in the previous paragraph are some of the most popular it doesn’t include one that is gaining tons of steam among NCAA coaching staffs, media, and Vegas odds makers (and gamblers) and that’s Bart Torvik. That tool has gotten a lot more popular as of late and it’s for a few good reasons. First, it has proven to be one of the most accurate predictors of games over the last few seasons. Secondly, it’s incredibly customizable and you can input variables such as date ranges to evaluate how a team has performed over a certain period of time, something you can’t do in any other metric.
And, as a bonus, it’s totally free. It’s absolutely a website you should check out and play with after reading this.
Other than the fact that it’s an accurate metric that’s laid out intuitively on a free website there is one other thing that makes it special and entirely different than any other popular metric out there.
It has 2020-21 season predictions.
For any of you that follow the other big name predictive metrics discussed in the first paragraph, you’ll know that preseason rankings don’t come out until just before the season starts, something that’s a bit curious given the fact that rosters are usually set long before the season starts. That makes Bart Torvik really stand out and for that reason we can look ahead to what his metric thinks of Florida right now as the roster stands.
Before we take a look at what the metric predicts for Florida, it should be pointed out that 8 of his top 10 preseason teams last year ended up in the top 10 of his metric at the end of the season, so the preseason predictions have been incredibly accurate. Another note—in the current algorithm he has Keyontae Johnson and Andrew Nembhard returning, so the metric is currently accounting for the Gators essentially being at the fullest strength they could be entering next season.
With that being said… Bart Torvik’s preseason ranking for Florida entering the 2020-21 season is…drum roll please…
5th.
5th!
This is equal parts interesting and exciting, because his preseason rankings have been accurate in the past and while anything can happen, it’s encouraging to know that predictive metrics are looking favorably at Florida. Of course, the assumption that Keyontae Johnson and Andrew Nembhard are returning play a big role in this, but considering where they are currently projecting even if one of them didn’t come back it’s looking like the team is in good shape.
Another thing that makes Bart Torvik’s metric so stellar is that it has an algorithm to predict player stats, including for JUCO transfers and top-50 recruits (sorry Niels Lane, didn’t quite make the cut). Predicting each player’s personal production helps make for the most accurate assessment of a team’s standing going into the season, and also makes for a fascinating look at what can be expected from each athlete. Here are the statistical predictions for each player, including the basic counting stats as well as what percentage of minutes that player will be on the floor:
Keyontae Johnson
15 Points
6.9 Rebounds
1.8 Assists
80% Min
Andrew Nembhard
12.7 Points
3.5 Rebounds
4.4 Assists
79% Min
Noah Locke
12.7 Points
2.8 Rebounds
1.2 Assists
71% Min
Scottie Lewis
10.1 Points
3.8 Rebounds
1.1 Assists
67% Min
Anthony Duruji
6.4 Points
4.5 Rebounds
1 Assist
59% Min
Tre Mann
5.1 Points
2.3 Rebounds
1 Assist
47% Min
Omar Payne
5.8 Points
4.8 Rebounds
0.6 Assists
45% Min
Tyree Appleby
4.7 Points
1.7 Rebounds
2.2 Assists
36% Min
Osayi Osifo
3.4 Points
2.3 Rebounds
0.4 Assists
31% Min
Samson Ruzhentsev
2.2 Points
0.6 Rebounds
0.5 Assists
19% Min
Jason Jitoboh
2 Points
1.2 Rebounds
0 Assists
17% Min
Ques Glover
2.3 Points
0 Rebounds
0.7 Assists
12% Min
Those numbers will all generated by a formula based on past performances of transfers and recruits from similar leagues as well as what could be expected from a returning player so they can be taken with a grain of salt, but computer generated predictive numbers from the Gators were too interesting to not be included from a metric that suggests that Florida is the 5th best team in the country.
The first reaction seeing these numbers is that Tyree Appleby and Anthony Duruji aren’t expected to be huge impact players, something that’s curious considering they’ll be veteran players who were both extremely effective at their previous spots and now have a redshirt year under their belts. It’s also not like they came from awful leagues, both played in solid mid-major conferences where good basketball was played. However, for them to be projected to have more production those numbers need to be taken from somewhere else and it’s tough to say who might drop. Bart Torvik isn’t predicting any huge jumps in production from any particular player, but if everyone does end up returning for the Gators their strength will be in their depth and the fact that they won’t need one player to shoulder the weight offensively.
As mentioned earlier the 5th place ranking that Florida is projected to have is assuming the return of both Keyontae Johnson and Andrew Nembhard. Fortunately, as also mentioned earlier, Bart Torvik is a customizable wonderland that allows you to also see what the projection would be if you subtracted a player. So, this is what it would look like if Keyontae Johnson or Andrew Nembhard leave (unfortunately, it’s not customizable enough to see what the Gators’ prediction would be if they both left):
With Keyontae Johnson and Andrew Nembhard returning: 5th
With Keyontae Johnson leaving: 17th
With Andrew Nembhard leaving: 9th
And, just so you know, if Scottie Lewis left: 7th
So, as you probably predicted, Keyontae Johnson’s decisions have huge implications to Florida’s season next year, with Andrew Nembhard’s choice also holding a ton of weight. Since the tool doesn’t allow for it we don’t know what would happen if they both left, but if they’d drop to 17th without Johnson they’d almost certainly drop into the 20’s without both of them.
Keep that in mind when waiting for Johnson’s decision, as Bart Torvik shows how important he is to the team. If he were to leave it’s still looking like the Gators are in good shape, but with him as well as Nembhard returning, Bart Torvik would suggest they are title contenders.
Does that get you fired up many, many months from next season starting? Thought so.