With college football recruiting heating up and with the buzz over a proposed expansion of the College Football Playoff, the summer “talking season” is in full swing. When both topics are considered together, several interesting questions arise. We can use data to try to answer these questions.
The first question that is often asked is how much does recruiting matter in which teams make the playoffs and which team ultimately wins the national title? In order to answer this question analytically, one first needs to have a way to quantify recruiting in the first place.
Any method that attempts to quantify the quality and potential of thousands of 17-year-old or 18-year-old high school athletes from every corner of the nation is bound to be a bit fuzzy and subjective. When it comes to individual players, all we can do is to rely on the opinions of recruiting experts and services such as Rivals and 247Sports. When it comes to ranking and comparing the groupings of these player who make up yearly recruiting classes, a bit more analysis can be done.
How to Count the Stars (a refresher)
Over the years, I have developed my own method of “scoring” individual recruits and recruiting classes. Using the Rivals rating system as a basis, and I rely heavily on a correlation that I developed comparing the initial Rivals rating of every high school recruit in its database to the eventual NFL draft rate for players of the same rating.
Figure 1 below gives a version of that correlation, which was updated following the 2021 NFL Draft and includes recruits who graduated from high school between 2007 and 2016.
What Figure 1 shows us is that while stars do matter, talent can be found everywhere. Over a very large set of data, roughly half of all high school five-star recruits wind up getting drafted. About a quarter of four-star recruits get drafted. That rate drops to about eight percent for three-star athletes, and five percent for two-star athletes.
What I realized a few years ago is that this graph can be used to approximate the quantifiable contribution that any individual player is likely (on average) to make on any given recruiting class or roster. The y-axis of Figure 1 defines the “NFL Draft Potential” or NDP of any given player. A collection of players (in a class or on a roster) can now be measured based on the number of likely NFL Draft picks that the group contains.
I used this method extensively in the seven-part recruiting series that I wrote last summer. A lot more detail and examples can be found there.
How to Make the Playoffs
When it comes to the College Football Playoff, then, how well does a program need to recruit in order to make the playoffs and win a national championship? I looked at the recruiting data for all FBS teams back to the 2007 recruiting class and compared it to the teams that actually made the playoffs (or finished in the top-four of the final AP poll) or won a national title since 2007. A summary of that analysis is shown in Figure 2 below.
There is a lot to unpack here. In this case, I used my NDP metric to estimate the average NFL Draft Potential of each team’s roster in each year. Based on this analysis, teams can be “binned” into 10 different groups (shown by the data points) in any given year. These bins can also be sub-divided into four recruiting tiers that more or less correspond to averaging a top-10 class, top-25 class, a top-40 class, and everyone else, as shown on the x-axis.
I have placed team logos onto the graph to give an approximate overall ranking and positioning of some notable teams. Note that each data point represents a collection of different recruiting classes across different years, so this is meant to be a rough guideline and not a precise average.
As for the data itself, the blue line gives the percentage of all teams that fell into that recruiting bin either made the playoffs (from 2014 to 2020) or finished the regular season in the AP top-four (for 2007-2013). The orange line similarly shows the fraction of teams in each bin that went on the win the national title.
Once again, this figure clearly shows that recruiting matters, and it seems to matter quite a bit. It also demonstrates the absolute dominance of Alabama. It should be noted that the right-most data point includes nine total recruiting classes: Alabama’s last seven classes and Georgia’s last two classes. In other words, it is good to be Alabama, and how ever much you think that you want Bama, you definitely don’t want to play Bama. Clearly, the Crimson Tide’s rate of making the playoffs and winning the title is in a class by themselves.
As for the second data point to the right, this represents a collection of about six teams that all have averaged a top-five recruiting class for a significant stretch of time over the last 15 or so years. These teams include Georgia, USC, Ohio State, Florida State, Florida and LSU. Four of those teams have won national titles since 2007, and if we add Alabama’s contribution, these top seven teams have captured almost half of the College Football Playoff/AP top-four places since 2007 (26 of 56, or 45 percent).
Teams that can string together multiple recruiting classes with an NDP score over 5.00 have a clear advantage over the rest of the country. There are the truly elite recruiting teams, and as a result, these teams have better odds to make the College Football Playoff compared to the rest of the FBS (about a one-in-three chance in any given year when a school has this level of roster).
After that, there appears to be a plateau in the data. For teams with a consistent NDP score in the high threes up to 5.0, the odds to make the playoffs are between 15 and 20 percent in any given year. This tier of teams includes Texas, Clemson, Oklahoma, Auburn, Notre Dame, Michigan, and Penn State who generally have classes ranked between No. 6 and No. 15 in the country.
For teams that do not consistently recruit at a top-15 level nationally, there is still hope for postseason glory. About 20 percent of the total College Football Playoff/AP top-four teams have not recruited at this elite level. As the graph shows, Oregon, Washington, Michigan State, Stanford, Wisconsin and TCU have all either made the playoffs or would have in the years prior to 2014. While recruiting is certainly important, coaching and player development is also important, and it has allowed this group of teams to cycle up and go toe-to-toe with the elite recruiting teams from time to time.
That said, while these teams with “weaker rosters” can make the playoffs, they have not shown the ability to actually win the national title recently. Only one team since 2007 has won the national title with an NDP weighted roster score below 4.0 and that was Clemson in 2016.
It should also be noted that outside of the “Alabama Tier” pretty much all of the teams that recruit at the top-10 level have similar odds to win a national championship (six to eight percent) even though the teams at the top end of that group have better odds to make the playoffs.
The final takeaway message from Figure 2 is that although recruiting certainly helps, it is not sufficient to guarantee on-field success. USC by some metrics has been the third best recruiting program in the country since 2007, but since 2010 the Trojans have finished the season unranked (five times) more often than they have finished in the top-10 (only twice).
Texas finishes consistently in the top-10 in recruiting but has finished the season ranked in the top-15 only once since 2010. I hear that there is also a team based in Ann Arbor that is a consistent top-15 recruiting powerhouse, but that has finished the season unranked more times than not since 2008, and has not won a conference title since 2004.
In contrast, teams like Oklahoma and Clemson have just been borderline top-10 recruiting programs over the past 15 years, but that have claimed 11 playoff spots and two national titles (both by Clemson) between them. Clearly, strong recruiting, in combination with player development and coaching is needed to be truly successful in the modern game.
The Rich Get Richer?
The analysis above paints a clear picture of what it takes to make the College Football Playoff and to contend for a national title. What it also suggests is that there is essentially an elite group of teams that tend to dominate recruiting and thus claim most of the playoffs spots. It should be noted that 20 of the total 28 playoff spots since 2014 have been filled by just four teams: Alabama, Clemson, Oklahoma, and Ohio State.
While recruiting certainly impacts the team that makes the College Football Playoff, perhaps a more interesting question is if the existence of the CFP impacts recruiting? I have seen some anecdotal evidence to suggest that the playoff era has increased the gaps between the football “haves” and “have-nots” as the elite players now only flock to those four schools. It is an interesting theory...but is it true?
Based on Figure 2 and the previous analysis, it is already possible to poke a few holes in this theory. After all, I just demonstrated how Oklahoma and Clemson have not historically recruited as well as teams such as USC or Texas (granted, Clemson has had the upper-hand in recruiting the past couple of years, given the Tigers’ recent success on the field). So it does not seem to be the case that the four most frequent playoff teams are sucking up all of talented players.
However, the question remains as to whether the existence of the College Football Playoff itself has caused a change in recruiting patterns. In order to try to answer this questions, I divided the recruiting data that I have into two equal halves: the recruiting classes from 2008 to 2014 (more or less prior to the playoffs) and from 2015 to 2021 (since the CFP was clearly established).
I then used my NFL Draft Potential metric to rank the recruiting classes of all 64 Power Five teams from No. 1 to No. 64 in both the pre-playoff era and current era. In order to smooth out the data, I clustered the teams into groups of five and then tabulated the total number of elite recruits in each bin of five teams. I focused on the top-four Rivals rating categories, which corresponds to all of the five-star and four-star players in a given class, and usually includes around 350 recruits per year. The results of this calculation are shown below in Figure 3.
One thing to note is that the identity of the top-five teams (the farthest left bar in each graph) is not exactly the same is both eras. From 2008-2014, the top-five teams were Alabama, Florida, USC, LSU and Florida State. From 2015 to now, the top-five teams were Alabama, Georgia, Ohio State, LSU and Clemson.
As Figure 3 shows, there is a subtle but clear shift in the distribution of the elite players between the two eras. For the five-star recruits (top left panel), a full 50 percent of them committed to the top-five teams during the College Football Playoff era (the blue bars), compared to just 40 percent in the pre-playoff days (the red bars). That is an increase of 22 five-star players, which corresponds to just over a half of a five-star recruit per class for each of the teams at the very top of the recruiting pecking order. That is a big jump.
For the high four-star recruits (6.0 Rivals rating) the trend is similar, but a bit more subtle. It is more subtle yet with the mid-four star recruits (5.9 rating) and this trend essentially disappears for the low-four star (5.8 rating) recruits. The biggest effect appears to be with the national top-100 or so recruits.
While it is easy to focus on what is happening with the elite teams, this change in the distribution of elite talent is perhaps easier to recognize by looking at the teams in the bottom half of the Power Five (on the right hand side of each panel).
In all four cases, the “tail” of the distribution is longer for the red bars (prior to the College Football Playoff) than after the CFP was instituted. Eight to 10 years ago, a small but noticeable fraction of elite recruits still committed to schools ranked in the bottom half of the Power Five in recruiting prowess. But, since 2015 that fraction has decreased significantly.
This trend is seen more clear if we compare the fraction of elite prospects who committed to teams outside of the top-30 in recruiting rankings before and after the four-team playoff era. That data is shown below in Figure 4.
In this case, each bar is labeled with the total number of recruits in that category. Here the difference is rather striking. After the playoffs were instituted, the number of five-star recruits signing with lower leave Power Five programs dropped from about three players per year to less than one. The number of high four-star recruits dropped by more than half. The decrease for the lower end of the four-star scale was more modest, but it is real and noticeable.
What this analysis seems to show is that over the past seven years, there has, in fact, been a subtle but real concentration of high-end talent to the upper echelon teams in the FBS. Was the introduction of the College Football Playoff the driving force for this shift? That is harder to say. In cases like these, cause and effect is difficult to determine. Correlation does not necessarily imply causation.
The impending expansion of the College Football Playoff from four teams to 12 teams, however, may give an opportunity to test this theory. If more teams will have an opportunity to play for a national title, perhaps the cream of the high school crop will decide to attend like Oregon, Penn State, Miami, Oklahoma State, or even Michigan State instead of Alabama or Ohio State.
As for the Spartans, MSU’s recruiting overall recruiting ranking over the past 14 years is squarely in the middle of the Power Five with an average ranking right around No. 30 in the country. While history has shown that elite evaluation, development and coaching can make up for average recruiting, the odds of a playoff berth are long.
As Coach Mel Tucker gets established in East Lansing and as the playoff format changes, will the Spartans be able to draw enough elite talent to start to climb up the recruiting rankings on a consistent basis? Will the new staff be able to developed that talent and put this into positions on the field to win consistently? While the future is unknown, I believe the analysis above provides a roadmap by which we can judge the staff’s progress. I, for one, am excited to see what the future holds.