Hey guys, I forgot to share this here...
As I mentioned recently, I intend to move on from writing about CFB recruiting, at least in the short term. I still have a few pieces of key information to share however, that I think are important to future fan discussion.
With that said, have you ever noticed how often we reference specific national recruit ranking thresholds like “top-100” or “top-250” when offering our opinions on the state of things? It seems like people throw them out there all the time, but let’s be honest… nobody has really provided information to back up their actual significance. This bugged me for a long time, so this fall I dove into the numbers looking for factual clarity.
As part of my All-SEC project series (which I don’t plan to finish right now), I pulled detailed recruiting information on the 549 All-SEC players from the 2010-2019 seasons. Then, I narrowed things down slightly, to players from that dataset who were recruits from the 2007-2017 recruiting cycles. This eliminated prospects who hadn’t had sufficient development time to be considered (i.e.: 1st and 2nd year players in 2019), along with a few 5th year seniors from the 2016 recruiting class (which had minimal statistical impact).
From there, I divided the data set up by national composite recruit rankings at intervals of 25 (prospects 1-25 nationally, 26-50, 51-75, etc., all the way down to unranked). Once I divided the All-SEC players up by their recruit ranking, I analyzed the data in 2 distinct ways:
1) I assigned a simple pass/fail test, to discover a “success rate percentage” for each rankings group. So, if 10 out of 100 total SEC school signees from a particular band earned All-SEC recognition during the period, then the success rate for that band was 10%.
2) I factored in all the multi-year award winners to establish an “All-SEC seasons per signee” for each of the same national ranking intervals as in part 1. So, if there were 25 total All-SEC seasons from a particular band that included 100 signees during the period, then the rate was .25 All-SEC seasons per signee. I think this is an important 2nd view of the issue, since players from a particular ranking-band might become repeat winners more than those in another.
So, whether you’re evaluating a recruit’s chances to hit it big time, the elite-player potential within an individual class, or the top-to-bottom elite potential of a roster, these numbers should provide an important new reference baseline. FINALLY, we have facts to back up people’s claims about the importance of recruits at each national ranking level.
Moreover, this information does more than merely help us look forward more accurately. It also provides us a fun new tool with which to evaluate the performance of recruiting classes as they leave the roster. All you have to do is review a class and identify the All-SEC seasons score for each signee in the class (.71 for a #4 national recruit, .46 for a #32 overall recruit, etc.). Once you assign these scores, then add each player up into a class total. The sum of the individual scores should provide a baseline expectation of All-SEC seasons for that class over their 4-5 years on the roster. Let’s say that number ends up at 4. If the group produces 6, then they’ve overachieved according to the 10-year league average. If they produce only 3, then they’ve slightly underperformed. This obviously has some weaknesses (top-recruiting schools like Bama / Georgia could suffer from a diminishing rate of return, since lots of top recruits compete for a limited number of on-field spots), but hey, this is more than we’ve ever had to explore this issue. If nothing else, it’s a start that I hope others will build upon.
Ok, after that long-winded explanation, here’s the data…
Primary Takeaway - It’s all about 5-stars and top-50 players. The success rate and seasons per signee rates are simply staggering for players 1-25, and to a lesser degree those ranked 26-50. Things drop off quickly from there. Remember, as I laid out in the All-SEC project series, it’s all about assembling 6-8+ All-SEC performances during the same season. Now, consider the following scenario, which will compare the recruiting classes of 2 teams, based on the All-SEC seasons per signee metric:
Team 1 signs the following 23-man class:
1 Top-25 recruits (5-star)
2 recruits ranked 51-75
19 Additional four stars, all ranked 126-150 nationally, giving the class a 100% blue chip ratio
If this class performed to the SEC’s average, we could expect it to produce 4.4 All-SEC seasons during their time at the school.
Team 2 instead signs the following 23-man class:
3 Top-25 players
2 recruits ranked 26-50
4 recruits ranked 51-75
13 3-stars, ranked 401-450
This much more top-heavy Team 2 class has far less depth, but projects to produce 6.18 All-SEC seasons (nearly 50% more than the class of team 1!).
As it pertains to producing championship-level group of All-Conference talent, the data is clear here: It’s all about landing 5-stars and truly elite prospects. Unfortunately, “top-250” and “top-100” are overrated thresholds these days, at least, if we’re talking about prospects at the lower end of those bands. But let’s be honest, nobody refers to players like Gervon Dexter as either (even though he’s technically both, as well as a 5-star), because we like to celebrate prospects by their loftiest recognition. While I have some other ideas about this data, I look forward to reading your takes… and honestly, I just don’t have time to explore it fully.
A few limitations to consider:
1) I can’t effectively evaluate individual coaches against this info, because the sample size shrinks so dramatically. Remember, we have a fairly good data set to work with overall, but that group of 549 All-Sec players played for 14 different teams. When you further divide the data set by an additional two-dozen ranking intervals, the data ends up all over the place.
2) This is a measure of top-end talent and not meant to suggest that quality, sub-elite performances aren’t an important component of championship teams. But again, history says you need at least 6-8+ simultaneous All-SEC performances to win the conference.
3) Yes, coaching matters too. A lot actually. But the broader-scoped star-based data I have on this issue doesn’t appear to show any SEC teams significantly escaping the realities of historic recruiting trends.
What I have left…
While this is the biggest piece of info I had left unshared, I still have some data on the All-SEC success rates of each on-field position by recruiting ranking. I’ll try to get that to you as well. There may be a few other things as well. For now though, I’ll answer questions about this as best I can.
As I mentioned recently, I intend to move on from writing about CFB recruiting, at least in the short term. I still have a few pieces of key information to share however, that I think are important to future fan discussion.
With that said, have you ever noticed how often we reference specific national recruit ranking thresholds like “top-100” or “top-250” when offering our opinions on the state of things? It seems like people throw them out there all the time, but let’s be honest… nobody has really provided information to back up their actual significance. This bugged me for a long time, so this fall I dove into the numbers looking for factual clarity.
As part of my All-SEC project series (which I don’t plan to finish right now), I pulled detailed recruiting information on the 549 All-SEC players from the 2010-2019 seasons. Then, I narrowed things down slightly, to players from that dataset who were recruits from the 2007-2017 recruiting cycles. This eliminated prospects who hadn’t had sufficient development time to be considered (i.e.: 1st and 2nd year players in 2019), along with a few 5th year seniors from the 2016 recruiting class (which had minimal statistical impact).
From there, I divided the data set up by national composite recruit rankings at intervals of 25 (prospects 1-25 nationally, 26-50, 51-75, etc., all the way down to unranked). Once I divided the All-SEC players up by their recruit ranking, I analyzed the data in 2 distinct ways:
1) I assigned a simple pass/fail test, to discover a “success rate percentage” for each rankings group. So, if 10 out of 100 total SEC school signees from a particular band earned All-SEC recognition during the period, then the success rate for that band was 10%.
2) I factored in all the multi-year award winners to establish an “All-SEC seasons per signee” for each of the same national ranking intervals as in part 1. So, if there were 25 total All-SEC seasons from a particular band that included 100 signees during the period, then the rate was .25 All-SEC seasons per signee. I think this is an important 2nd view of the issue, since players from a particular ranking-band might become repeat winners more than those in another.
So, whether you’re evaluating a recruit’s chances to hit it big time, the elite-player potential within an individual class, or the top-to-bottom elite potential of a roster, these numbers should provide an important new reference baseline. FINALLY, we have facts to back up people’s claims about the importance of recruits at each national ranking level.
Moreover, this information does more than merely help us look forward more accurately. It also provides us a fun new tool with which to evaluate the performance of recruiting classes as they leave the roster. All you have to do is review a class and identify the All-SEC seasons score for each signee in the class (.71 for a #4 national recruit, .46 for a #32 overall recruit, etc.). Once you assign these scores, then add each player up into a class total. The sum of the individual scores should provide a baseline expectation of All-SEC seasons for that class over their 4-5 years on the roster. Let’s say that number ends up at 4. If the group produces 6, then they’ve overachieved according to the 10-year league average. If they produce only 3, then they’ve slightly underperformed. This obviously has some weaknesses (top-recruiting schools like Bama / Georgia could suffer from a diminishing rate of return, since lots of top recruits compete for a limited number of on-field spots), but hey, this is more than we’ve ever had to explore this issue. If nothing else, it’s a start that I hope others will build upon.
Ok, after that long-winded explanation, here’s the data…
Primary Takeaway - It’s all about 5-stars and top-50 players. The success rate and seasons per signee rates are simply staggering for players 1-25, and to a lesser degree those ranked 26-50. Things drop off quickly from there. Remember, as I laid out in the All-SEC project series, it’s all about assembling 6-8+ All-SEC performances during the same season. Now, consider the following scenario, which will compare the recruiting classes of 2 teams, based on the All-SEC seasons per signee metric:
Team 1 signs the following 23-man class:
1 Top-25 recruits (5-star)
2 recruits ranked 51-75
19 Additional four stars, all ranked 126-150 nationally, giving the class a 100% blue chip ratio
If this class performed to the SEC’s average, we could expect it to produce 4.4 All-SEC seasons during their time at the school.
Team 2 instead signs the following 23-man class:
3 Top-25 players
2 recruits ranked 26-50
4 recruits ranked 51-75
13 3-stars, ranked 401-450
This much more top-heavy Team 2 class has far less depth, but projects to produce 6.18 All-SEC seasons (nearly 50% more than the class of team 1!).
As it pertains to producing championship-level group of All-Conference talent, the data is clear here: It’s all about landing 5-stars and truly elite prospects. Unfortunately, “top-250” and “top-100” are overrated thresholds these days, at least, if we’re talking about prospects at the lower end of those bands. But let’s be honest, nobody refers to players like Gervon Dexter as either (even though he’s technically both, as well as a 5-star), because we like to celebrate prospects by their loftiest recognition. While I have some other ideas about this data, I look forward to reading your takes… and honestly, I just don’t have time to explore it fully.
A few limitations to consider:
1) I can’t effectively evaluate individual coaches against this info, because the sample size shrinks so dramatically. Remember, we have a fairly good data set to work with overall, but that group of 549 All-Sec players played for 14 different teams. When you further divide the data set by an additional two-dozen ranking intervals, the data ends up all over the place.
2) This is a measure of top-end talent and not meant to suggest that quality, sub-elite performances aren’t an important component of championship teams. But again, history says you need at least 6-8+ simultaneous All-SEC performances to win the conference.
3) Yes, coaching matters too. A lot actually. But the broader-scoped star-based data I have on this issue doesn’t appear to show any SEC teams significantly escaping the realities of historic recruiting trends.
What I have left…
While this is the biggest piece of info I had left unshared, I still have some data on the All-SEC success rates of each on-field position by recruiting ranking. I’ll try to get that to you as well. There may be a few other things as well. For now though, I’ll answer questions about this as best I can.
