How did roster Blue Chip recruit percentage relate to on-field success in 2018 for the college football regular season? A statistical review.

As recruiting season heats up, there is a lot of attention being paid to how teams are able to add blue chip (4-star and 5-star high school players) recruits to their rosters. Just browse the internet and you will find a plethora of correlational analyses touting the importance of blue chip (BC) recruiting. There is no doubt that recruiting is important (after all, the coaches heavily stress recruiting success). To say otherwise is likely naïve and statistically unsupported. What gets missed in many of the conversations (at least in my experience) is the question of how important recruiting is relative to other factors. Where on the scale of ‘not at all’ to ‘all about Jimmys and Joes, not Xs and Os’ does impact of recruiting on winning really lie? Well, that of course varies from team to team, as each has their own situation. But we can explore a sample of talented teams to get a glimpse of the range that matters to most of us; the top 25.

To gain some insight as to the strength of the relationship between recruiting and winning, there are a few ways one could go about it. In this analysis, I look at the relationship between the top 25 most talented teams, as indicated by their Rivals 247 roster composite ratings, and those same teams’ regular season win percentage. Each of these teams’ BC percentage was used to analyze what their 2018 regular season win percentage would be expected to be as compared to the other 24 teams in the sample. To ensure the scope of what is being discussed here is not misunderstood, the purpose is to explore the expected impact on winning if predicted by roster BC%. This is not an exercise in predictive analytics- it is a review of what happened and exploring relative performance on a single, isolated variable that is discussed quite often- BC%. It is not intended to discuss all of the factors that go into winning (coaching, SOS, home field, etc.). It is intended to do just the opposite- flesh out how a top 25 most talented team should perform relative to their peers on the basis of BC roster percentage alone.

Now, to the data. The table below shows each of the teams in the sample. These teams were inclusion was based upon their average roster rating, as previously described. Then, their blue-chip recruit percentage was calculated.

bcp chart

The above list is in alphabetical order of school. In the far-right column, that is how a team performed relative to what their BC% would have predicted. For example, Florida would be expected to win 64% of their 2018 regular season games on the basis of their BC% of 42%. The Gators won 75% of their games, 11% over what would be predicted by BC% alone. The top ‘over performer’ was Notre Dame at +27% with the most ‘under performing’ team being USC at -40%.

bc scatter

The above graph outlines where along a continuum each team would plot. Above the dotted line indicates ‘over performing’, below the line represents the opposite. I included some notable teams for illustration purposes. UCLA, even though their roster wasn’t stacked, still under performed. FSU did terrible.

Here is a scatter plot with each team’s data point entered:

top 50 tabline

Methods:

A simple linear regression was performed to predict overall win percentage of the designated 25 most talented rosters, as determined by an independent scouting service website based on the team’s roster blue-chip percentage (BC%).  A significant regression equation was found (F(1,23) = 10.59, p < .001), with an R² of 0.315. Teams’ predicted win percentage is equal to 0.324 + .756 * X, where X= BC%.

Regular season achievement: The Southeastern Conference.

Now that the regular season is complete, I took a look at how each team fared relative to the talent they have on hand. Taking the composite rating of each SEC team’s roster talent, as calculated by the Rival’s 247 Composite website, I analyzed how each team’s overall win percentage aligned with their talent. Obviously, we would expect teams with more talent to win more than those with less talent. But by isolating the talent as a variable, and then viewing the results of the season, we can see how good of a job each coach has done. To be sure, there is an element of randomness (injuries, etc.). However, this analysis gives you an idea as to how a team performed relative to its talent level.

SEC chart

In the chart above, you can see the regression line through the middle of the chart. This represents the expected value of win percentage (along the vertical, or “Y” axis) is it relates to the 247 Composite Talent average per team (horizontal, or “X” axis). Teams below the line had a lower winning percentage than their roster talent would predict. Teams above the line performed better than their roster talent would predict. And, of course, those teams with scores on or at the line performed as expected. The distance from the line indicates the degree in which a team under-performed or over-performed.

The table below shows the data. Kudos to all the teams that exceeded the expectations. Detention for those that didn’t, especially Tennessee and Auburn.

chart 2 reg season

For our stats friends, the model is significant and has an adjusted R² of .333, which means talent rating accounted for 33.3% of the variance in win percentage, with 66.7% of the results coming from factors other than talent rating. If you care to play around with the numbers (hey, SLR is fun because it is easy), the β = -4.009, ‘Talent’ coefficient (unstandardized) = 0.053. In this model, if you’re Florida for example, with a roster talent of 88.28, you would multiply 88.28 with 0.053 (giving you 4.68) and subtract 4.009 from that, giving you .670. So, you’re expected win percentage with be 67%, which is 8 games. So, Florida out-performed their talent by one game this year. Mullen is good at coaching sports. The table displaying expected win percentage based on talent is below:

win expectancy

Like the Florida example above, you can see how it all plays out for each team in this table. Alabama won one game beyond expectations (so they are above the line in the scatter plot above), while Arkansas lost 4 more games than their roster talent would predict. Of course, a ‘zero’ means they performed at expected levels. Great job this year by Kentucky, winning 3 games over their talent level.

Dan Mullen has improved recruiting at MSU and Florida; No reason to believe he won’t get the players Florida needs to compete for championships

Recruiting is a hot topic for college football fans. And for good reason, the acquisition of talent is one key aspect to winning football games. When it comes to Florida coach Dan Mullen, there is a general notion that he is a great coach but so-so recruiter. While he hasn’t recruited an elite class yet, he also hasn’t been in position to. At Mississippi State, it can be very difficult to recruit top players. While that isn’t the case at Florida, he has only been there for one full recruiting cycle.

As always, I like to test talking points and narratives to see not only if they are true or not, but to what degree is there some truth. So I took a look at MSU and Florida’s recruiting before and after Dan Mullen arrived. In each situation, he improved the school’s recruiting measures significantly by the two metrics that matter most- average class talent rating and blue chip (4 and 5 star recruits) percentage.

mullen recruiting before and after

 

In the above chart, you can see that the five years before Mullen arrived at MSU, the average blue chip percentage (BC%) was 9% overall and the talent rating was 80.41. With Dan, it increased to 17% and 85.84. At Florida, previous coach Jim McElwein had an overall BC% of 34% and rating of 88.36. Using Mullen’s first year and current 2019 class (that will assuredly change), he has a BC% of 61% and rating of 90.21. Of note, McElwein had a good class lined up for 2018, but losing a bunch of games (among other things) got him fired.

Being that Mullen elevated MSU to higher levels of recruiting during his time there, and has already done that at Florida, it appears as if he can recruit after all. We can only wait and see if Mullen can reel in top 5 classes consistently at Florida, but he is clearly trending in the right direction so far, and I would expect that trend to continue.

There is no doubt that UF recruited at an extremely high level under Meyer and Muschamp. So how does Dan compare to these guys? Pretty good actually:

coaching2

Meyer had a talent average of 90.69 compared to 90.21 for Mullen and a BC% of 65% to Mullen’s 61%. Champ had a 92.16 and 56% BC%. This is quite telling… Mullen with an admittedly small sample size is in the same territory as the great recent recruiters at UF. You can see also how highly ranked Meyer and Muchamp’s classes were. This probably indicates a general shift in recruiting rating, which is why standardizing scores may be important for analysis (not done here).  At a glance, however, it seems as if Mullen is doing just fine bring talent to Gainesville. And he should probably get some credit for the great classes Meyer had since he was the Offensive Coordinator at the time.

Achievement Rankings: 2018 ACC vs 2018 SEC. What would FSU look like in the SEC?

In this analysis, I computed the achievement scores for each team in the ACC and SEC and ranked them. The higher the score, the more ‘over achieving’ a team is; they win more games than their talent level would predict compared to other members of their conference. Most of the table is self-explanatory. The talent score is the 4-year moving average of recruiting talent level according to the Rivals 247 Composite ratings.

sec vs acc

As the tables show, Florida is 5th in the SEC in terms of achievement. They are slightly out-performing their talent level within the conference. As we would suspect, the Noles are doing terribly for the amount of roster talent they have compared to their conference peers.

Things get interesting when I switched the teams. How does Florida compare to the ACC teams and vice versa with FSU in the SEC? I took their numbers and transposed them. Of course, every team within each conference would have different records because FSU would have played Georgia and Florida would have played Clemson, etc. But since that would involve speculation, it has no value. We can, however, take the known figures and simply compare them. So, I did:

fsu to sec

As we can see, Florida would drop to underachieving status in the ACC. This is because the ACC, on average, is less-talented than the SEC on average. The mean talent rating for 2018 SEC teams is 88.24. For the ACC it is 86.34. The conference winning percentage for the SEC is 47% (still games to go- will end up at 50%) and 50% for the ACC. When we add Florida to the ACC and take FSU out, the ACC mean talent level drops to 86.15 (makes sense- Florida has a lower talent score than FSU). With FSU in the SEC, the mean talent score rises to 88.42.

What does this tell us? As bad as FSU is in the ACC, they would be just as bad in the SEC, but it would be slightly more acceptable. FSU has the most talented roster in the ACC (edging out Clemson). However, in the SEC they would be the 3rd most talented team and slightly ahead of LSU. Florida in the ACC would be expected to have a better conference record, as the teams are weaker. Florida has the 5th most talented roster in the SEC, but would have the 2nd most talented roster in the ACC.

User note: The achievement scores are based on standardized values. So when you see Clemson, who is undefeated, with a negative value (indicating underachievement), the number is very close to zero, but it is basically saying that with that much talent disparity, Clemson should be undefeated. Since going undefeated obviously cannot be an underachievement, it essentially means Clemson, like Alabama, is doing what it should be- they are performing at expected levels for talent on hand.

Roster Dispersion of Major College Football Teams: From Where Do They Get Their Players?

This study will be an ongoing one, as this is just preliminary. The below displayed maps show where some major college football teams accumulate their players.

18Gators

Florida Gators

18canes

Miami Hurricanes

18nulls

Florida State Seminoles

18buckeyes

Ohio State Buckeyes

18tide

Alabama Crimson Tide

18dwags

Georgia Bulldogs

These maps show the entire roster makeup, so it includes walk-ons, etc. More work is to be done here, but it is at least kinda neat (to me) to see from where teams obtain their players.

WLOCP hangover?Florida, yes. UGA, no.

From Reddit user agage3

“Our performance after a WLOCP game win compared to our performance after a WLOCP loss. I swear when we lose our season just tanks but when we win that game we ride the wave to consecutive late season wins.”

To see if UF truly has a hangover after a loss at the WLOCP, I reviewed the results of the game since from 1998-2017. I looked at how Florida and Georgia each did the following week as well as for the rest of the season. Florida has gone one to win 64% of their games overall after the WLOCP in that time frame with Georgia winning 75%.  While, the game itself tends to go our way, we fair worse afterwards.

fl vs ga

For the statistically-interested, an independent samples t-test was conducted to determine if the mean winning percentage between the two was statistically significant. The results are statistically significant at the a= .10 (p= 0.82). Levene’s test was not significant, thus equal variances were assumed. Effect size was small to medium (r²=0.08).

Takeaway, though not strongly significant (typical a level is 0.05, and medium effect size for r² is 0.09), the dwags have had slightly more success after the WLOCP over the last 20 years, but Florida has the advantage where it counts: head to head and championships.