2017 Recruiting Class Could Be The Best Yet

[stever20]

Here's why using team rankings to judge a program's or conference's true recruiting health is so tainted, especially at the top. Looking at recruiting years 2013-2016, there were 4 programs that were each in the Top 10 team rankings every year.

1. UK
ESPN Top 100 players signed - 20
ESPN Top 100 on current roster - 7
Avg Team Recruiting Ranking - 1.5

2. Duke
Top 100 Signed - 16
Current Roster - 9
Team Avg. Rank - 2.75

3. Kansas
Signed - 11
Roster - 4
Rank - 6.0

4. Arizona
Signed - 12
Roster - 5
Rank - 6.0

Nova
Signed - 6
Roster - 6
Rank - 32.8

I will concede that Duke, UK, KU and Ariz. all have the better individual classes each year. But most of those players they recruit leave early. So by using team rankings you are providing greater statistical value to those programs that recruit one-and-dones, which are more prevalent in the P5 schools. So in your analysis, you show that UK has the statistical value of picking up 20 ESPN 100 players, when their current roster only shows 7. That is what we call bad data.

Also there is nothing factored in for experience. In other words, who is more valuable to their program, Kris Jenkins and Josh Hart who helped make up the 37th best class for Nova or Rawle Alkins and Kobi Simmons, frosh at Ariz who made up the 6th best class? I guess time will tell but I think I know where I would put my $ for this year. Next year doesn't matter b/c all 4 of those Wildcats might well be gone.

Where I disagree with you on that is that 1st off you sometimes have guys that do stick around for 2-3 years. 1 of the reasons that Duke is projected to be so good this year is that Grayson Allen is going to be returning. Duke has 9 guys like you showed in the top 100 this seasons roster. Kentucky has 7.

Also, a lot of times, the one and done guys just are that much more physically talented than the 3 and 4 year players. Folks here want to bring up last year's season. There's a reason why Gassaway asked if last year was a blip or a trend. Because the trend since the NBA went to the 1 and done rule has been the freshmen dominating college basketball.

[GumbyDamnit!]

Where I disagree with you on that is that 1st off you sometimes have guys that do stick around for 2-3 years. 1 of the reasons that Duke is projected to be so good this year is that Grayson Allen is going to be returning. Duke has 9 guys like you showed in the top 100 this seasons roster. Kentucky has 7.

Also, a lot of times, the one and done guys just are that much more physically talented than the 3 and 4 year players. Folks here want to bring up last year's season. There's a reason why Gassaway asked if last year was a blip or a trend. Because the trend since the NBA went to the 1 and done rule has been the freshmen dominating college basketball.

Agree that they are more physically talented in a NBA draft projection kind of way. But I am never going to agree that means you are a better college player than a wiley 4 year senior. Ryan Arch, Shabazz Napier, Kemba Walker, Kaminsky, Valentine, Draymond Green, Tyler Hansbrough, J. Noah, Russ Smith, Sharron Collins, etc. all represent players that stuck around and were every bit as impactful as any freshman could hope to be. The last 10 years there have been 106 players recognized as Consensus 1st or 2nd team AA. 17 of those were freshmen and only 3 of them--Anthony Davis & Kidd-Gilchirst and J. Okafor have won national championships. I would not call that "dominance." I think we are fed all this hype about the Diaper Dandies and more often then not they have so-so years before they bolt for the NBA $. There are exceptions... Melo, Anthony Davis. But for every one of those guys I'll raise you a Labissiere and Shabazz Muhammad.

[stever20]

just looking at the numbers a few things are clear....

from 2006-11- so these would have been the start of the NBA 1 and done rule....
1st and 2nd team aa's
26 sr
19 jr
14 so
8 fr

2012-16- 1 fewer year....
23 sr
10 jr
10 so
10 fr
gotten a lot less with the jr and so especially.

Also look at the total breakdowns by year...
2016 7 sr 1 jr 2 so 1 fr
2015 4 sr 3 jr 1 so 3 fr
2014 6 sr 1 jr 2 so 2 fr
2013 2 sr 3 jr 3 so 2 fr
2012 4 sr 2 jr 2 so 2 fr
2011 5 sr 3 jr 3 so 1 fr
2010 5 sr 3 jr 1 so 2 fr
2009 1 sr 6 jr 4 so 0 fr
2008 4 sr 2 jr 3 so 2 fr
2007 4 sr 3 jr 1 so 2 fr
2006 7 sr 2 jr 1 so 1 fr

last year was the most senior AA's in a decade. and one of only 4 years where 2 of the top 10-11 players in the country weren't freshmen. You can see why Gassaway asked what he did. From 2007-15, there were 35 senior 1st/2nd AA's in 9 years- so not even averaging 4 per year. Last year there were 7.

From your list of players- some of those guys never won a championship. Kaminsky was beat by Duke in the title game. Who led Duke? Freshmen....

I was a bit too strong I'll admit saying freshman dominate college basketball, but I don't think seniors necessarily do either.

[XtoDC]

I agree with the people on here that have pointed out how inconsistent the team rankings can be for a team from year to year. If a team has a really good roster coming back with only two scholarships open in an offseason and they bring in a high level transfer and the 110th best recruit in the country, they are going to have a pretty low team recruiting ranking for that year. The team's fans and coaches are probably thrilled with that result because they filled both scholarships (and presumably positions of need) without messing up the chemistry of the returning group since the players brought in are probably sitting out for the most part.

What is the best way to capture that though? Right now how most people are doing it is using the class rankings for each year, but that really penalizes schools when they have full rosters. Even if they get the one recruit they want and need this methodology says that school stunk at recruiting in that year. I would argue recruiting when you have a full roster is more difficult than when you have plenty of spots to fill and playing time to offer recruits. If athletes couldn't leave for the NBA early would UK still get every single one of the guys they get? When Jay Wright decides he is only going to bring in two recruits in a class and has two top 75 guys locked up he gets penalized in the recruiting rankings because he turned down the in state kid rated 115 that would absolutely love to play for Villanova.

I think to judge recruiting it might make more sense to look at a rolling three year average of the caliber of players being brought in. I'm thinking a system that assigns a point value for each range that a recruit falls into, then add it all up and divide by the number of recruits they took in those three years. It probably would be really difficult to do for all the schools to see where everyone compares, but maybe it would be possible just for the Big East schools? There is probably another even better way that someone smarter and/or more dedicated than me could come up with, so I would love to see other people tinker with the idea or tell me why it sucks.

5 points: 1 - 20
4 points: 21 - 60
3 points: 61 - 115
2 points: 116 – 200
1 point: 201 - up

Looking at Xavier using 247 Sports composite (since it was easy and they use multiple recruiting rankings to get the scores)

2017: 3, 2, 2
2016: 3, 3, 2
2015: 2
2014: 4, 3, 3, 3, 2, 2
2013: 3, 1
2012: 4, 3
2011: 4, 3, 3
2010: 3, 3, 3, 1
2009: 3
2008: 4, 2, 2, 2, 2

The best 3 year rolling period averaged 3.00 points from 2010 - 2012 and 2011 - 2013. The worst was actually 2013 – 2015 (and if everything stays status quo it will be surpassed by 2015 – 2017). It is interesting that the stats come out like this because the anecdotal evidence that we have would say otherwise. Xavier would have likely kept the commitment of DSR if they had been in the Big East. Trevon Bluiett would not have come to Xavier without the Big East.

I set up this methodology with no real agenda behind it, so I am sure the numbers could be tweaked with the ranges and the recruiting services changed to get a different result. I do believe something like this though would be a more fair way of looking at recruiting trends for a school. Thanks to anyone who actually bothered to read all this and have any thoughts or even want to take it further for other schools. Here are the final Xavier results if anyone is still interested.

2008 - 2010 2.50
2009 - 2011 2.88
2010 - 2012 3.00
2011 - 2013 3.00
2012 - 2014 2.80
2013 - 2015 2.56
2014 - 2016 2.70
2015 - 2017 2.43

[gtmoBlue]

I agree with the people on here that have pointed out how inconsistent the team rankings can be for a team from year to year. If a team has a really good roster coming back with only two scholarships open in an offseason and they bring in a high level transfer and the 110th best recruit in the country, they are going to have a pretty low team recruiting ranking for that year. The team's fans and coaches are probably thrilled with that result because they filled both scholarships (and presumably positions of need) without messing up the chemistry of the returning group since the players brought in are probably sitting out for the most part.


I think to judge recruiting it might make more sense to look at a rolling three year average of the caliber of players being brought in. I'm thinking a system that assigns a point value for each range that a recruit falls into, then add it all up and divide by the number of recruits they took in those three years. It probably would be really difficult to do for all the schools to see where everyone compares, but maybe it would be possible just for the Big East schools? There is probably another even better way that someone smarter and/or more dedicated than me could come up with, so I would love to see other people tinker with the idea or tell me why it sucks.

5 points: 1 - 20
4 points: 21 - 60
3 points: 61 - 115
2 points: 116 – 200
1 point: 201 - up

I set up this methodology with no real agenda behind it, so I am sure the numbers could be tweaked with the ranges and the recruiting services changed to get a different result. I do believe something like this though would be a more fair way of looking at recruiting trends for a school. Thanks to anyone who actually bothered to read all this and have any thoughts or even want to take it further for other schools.

Excellent take on the situation XtoDC.

For some reason I think a more even distribution would be fair...

5 points: 1 - 25
4 points: 26 - 50
3 points: 51 - 75
2 points: 76 -100
1 point: 101 – 125 (150)
0 point: 150 -200

[stever20]

the chart reinforces what I was saying about Xavier having recruited so well prior to joining the Big East.

2014 17 6 players 2.83 avg
2008 12 5 players 2.40 avg
2011 10 3 players 3.33 avg
2010 10 4 players 2.50 avg
2016 8 3 players 2.67 avg
2017 7 3 players 2.33 avg
2012 7 2 players 3.5 avg
2013 4 2 players 2.0 avg
2009 3 1 player 3.0 avg
2015 2 1 player 2.0 avg

at least so far- of the 10 classes raw, the A10 version had 3 of the top 5. With only 2009 as a bad one(and I'm guessing that's when Miller left for Arizona).

to your point on the raw avg....
1 2012 3.5 avg
2 2011 3.33 avg
3 2009 3.0 avg
4 2014 2.83 avg
5 2016 2.67 avg
6 2010 2.50 avg
7 2008 2.40 avg
8 2017 2.33 avg
9 2013 2.00 avg
10 2016 2.00 avg

so 3 best years by what you are saying was 12,11, and 09. With at least right now, 2 of the 3 lowest averages being 2016 and 2017.

The thing is, looks like Mack was recruiting pretty darn well in the A10 even. The numbers don't really support the anecdotal evidence at all.

[ChelseaFriar]

We are still the favorites to land Wabissa Bede and Christian David, both Top 100 guys on 247. We are just one of the ten teams in the conference. Stever is gonna have a conniption by the end of this recruiting season.

I hope you guys land Bede. I think he's going to be a great 4 year player. I wish Cooley went after him harder but he seemed to really focus on Makai Ashton-Langford, who is down to UConn and PC but appears to be a heavy UConn lean at this point.

Ashton-Langford to UConn. UConn could have 3 ESPN Top 35 PGs on their roster next season if they all return.

[GumbyDamnit!]

I think to judge recruiting it might make more sense to look at a rolling three year average of the caliber of players being brought in. I'm thinking a system that assigns a point value for each range that a recruit falls into, then add it all up and divide by the number of recruits they took in those three years. It probably would be really difficult to do for all the schools to see where everyone compares, but maybe it would be possible just for the Big East schools? There is probably another even better way that someone smarter and/or more dedicated than me could come up with, so I would love to see other people tinker with the idea or tell me why it sucks.

5 points: 1 - 20
4 points: 21 - 60
3 points: 61 - 115
2 points: 116 – 200
1 point: 201 - up

Looking at Xavier using 247 Sports composite (since it was easy and they use multiple recruiting rankings to get the scores)

2017: 3, 2, 2
2016: 3, 3, 2
2015: 2
2014: 4, 3, 3, 3, 2, 2
2013: 3, 1
2012: 4, 3
2011: 4, 3, 3
2010: 3, 3, 3, 1
2009: 3
2008: 4, 2, 2, 2, 2

The best 3 year rolling period averaged 3.00 points from 2010 - 2012 and 2011 - 2013. The worst was actually 2013 – 2015 (and if everything stays status quo it will be surpassed by 2015 – 2017). It is interesting that the stats come out like this because the anecdotal evidence that we have would say otherwise. Xavier would have likely kept the commitment of DSR if they had been in the Big East. Trevon Bluiett would not have come to Xavier without the Big East.

I set up this methodology with no real agenda behind it, so I am sure the numbers could be tweaked with the ranges and the recruiting services changed to get a different result. I do believe something like this though would be a more fair way of looking at recruiting trends for a school. Thanks to anyone who actually bothered to read all this and have any thoughts or even want to take it further for other schools. Here are the final Xavier results if anyone is still interested.

2008 - 2010 2.50
2009 - 2011 2.88
2010 - 2012 3.00
2011 - 2013 3.00
2012 - 2014 2.80
2013 - 2015 2.56
2014 - 2016 2.70
2015 - 2017 2.43

I am confused about a couple of things...

1. Why such an arbitrary point system? 1-20 (20), then 21-60 (40), then 61-115 (55) and then 116-200 (85). Why not provide a uniform value as gtmo suggested? Or possibly a value range that uniformly increases (20, 25, 30, 40 or 20, 40, 60, 80). But with the randomness of your value ranges, signals to me that the specific value assignments are not random at all and there is some reason why you would group them as you did.

2. Why a 3 year rolling average when you can just look at the 4 years before and 4 years after the league was formed?

So taking the two points above into effect, I did some math of my own using the 247 site.

Value ranges:
1-25 = 5
26-75 = 4
75-125 = 3
125-200 = 2
201+ = 1

X in A10 (2009-2012) = 12 recruits for an average rank of 2.75
X in BE (2014-2017) = 13 recruits for an average rank of 2.76

I left out 2013 because Stever and I can't agree on which class that year belongs (to the A10 or the BE), as it was not yet announced where X would be until March of '13. Interestingly 2013 seemed to be a complete anomaly for X. They recruited 3 kids but two of them were 200+. In fact since 2013, according to 247 they have not brought in anyone ranked 200+ since joining the BE. Not counting 2013 while int he A10 they brought in 3 players 200+.

So indeed X recruited well prior to joining the BE and continues to recruit well after. What may tip the #'s in the BE's favor is what happens with the rest of the 2017 class. For instance if X lands Scruggs, as an example, then their average rank for recruits jumps to 2.86, which would be a slight uptick from where they were before.

[XtoDC]

I think to judge recruiting it might make more sense to look at a rolling three year average of the caliber of players being brought in. I'm thinking a system that assigns a point value for each range that a recruit falls into, then add it all up and divide by the number of recruits they took in those three years. It probably would be really difficult to do for all the schools to see where everyone compares, but maybe it would be possible just for the Big East schools? There is probably another even better way that someone smarter and/or more dedicated than me could come up with, so I would love to see other people tinker with the idea or tell me why it sucks.

5 points: 1 - 20
4 points: 21 - 60
3 points: 61 - 115
2 points: 116 – 200
1 point: 201 - up

Looking at Xavier using 247 Sports composite (since it was easy and they use multiple recruiting rankings to get the scores)

2017: 3, 2, 2
2016: 3, 3, 2
2015: 2
2014: 4, 3, 3, 3, 2, 2
2013: 3, 1
2012: 4, 3
2011: 4, 3, 3
2010: 3, 3, 3, 1
2009: 3
2008: 4, 2, 2, 2, 2

The best 3 year rolling period averaged 3.00 points from 2010 - 2012 and 2011 - 2013. The worst was actually 2013 – 2015 (and if everything stays status quo it will be surpassed by 2015 – 2017). It is interesting that the stats come out like this because the anecdotal evidence that we have would say otherwise. Xavier would have likely kept the commitment of DSR if they had been in the Big East. Trevon Bluiett would not have come to Xavier without the Big East.

I set up this methodology with no real agenda behind it, so I am sure the numbers could be tweaked with the ranges and the recruiting services changed to get a different result. I do believe something like this though would be a more fair way of looking at recruiting trends for a school. Thanks to anyone who actually bothered to read all this and have any thoughts or even want to take it further for other schools. Here are the final Xavier results if anyone is still interested.

2008 - 2010 2.50
2009 - 2011 2.88
2010 - 2012 3.00
2011 - 2013 3.00
2012 - 2014 2.80
2013 - 2015 2.56
2014 - 2016 2.70
2015 - 2017 2.43

I am confused about a couple of things...

1. Why such an arbitrary point system? 1-20 (20), then 21-60 (40), then 61-115 (55) and then 116-200 (85). Why not provide a uniform value as gtmo suggested? Or possibly a value range that uniformly increases (20, 25, 30, 40 or 20, 40, 60, 80). But with the randomness of your value ranges, signals to me that the specific value assignments are not random at all and there is some reason why you would group them as you did.

2. Why a 3 year rolling average when you can just look at the 4 years before and 4 years after the league was formed?

So taking the two points above into effect, I did some math of my own using the 247 site.

Value ranges:
1-25 = 5
26-75 = 4
75-125 = 3
125-200 = 2
201+ = 1

X in A10 (2009-2012) = 12 recruits for an average rank of 2.75
X in BE (2014-2017) = 13 recruits for an average rank of 2.76

I left out 2013 because Stever and I can't agree on which class that year belongs (to the A10 or the BE), as it was not yet announced where X would be until March of '13. Interestingly 2013 seemed to be a complete anomaly for X. They recruited 3 kids but two of them were 200+. In fact since 2013, according to 247 they have not brought in anyone ranked 200+ since joining the BE. Not counting 2013 while int he A10 they brought in 3 players 200+.

So indeed X recruited well prior to joining the BE and continues to recruit well after. What may tip the #'s in the BE's favor is what happens with the rest of the 2017 class. For instance if X lands Scruggs, as an example, then their average rank for recruits jumps to 2.86, which would be a slight uptick from where they were before.

Gumby/Gtmo - Thank you for the thoughtful criticism. No system would be totally perfect and I typed this up at work this morning so mine has plenty of holes.

Wouldn’t any point system be pretty arbitrary? And like I said, I had no agenda. I really had no inkling how everything would eventually shake out. This has also only been applied to Xavier so far, it may show a surprising positive uptick for another school when applied.

Using different ranges is definitely something about the system that could be refined, I just came up with them on a whim because I didn’t want to spend a lot of time on it and do a lot of research. I do think it is very important though for the range to increase for each subsequent point rather than a uniform value. The difference between the 25th ranked recruit and the 1st ranked recruit is far greater than the difference between the 50th and 26th (and so on and so on). Having been subscribed to the recruiting websites we have been told there isn’t that much of a difference between the 75th guy and the 110th guy (or whatever example they gave). Where those actual tiers should be though and how much the range should increase each time I have no clue, I just went with my gut. ¯\_(ツ)_/¯

I went with a three year rolling average because to me that seemed like the fewest amount of classes possible to get a decent sample size of recruits. Four years obviously includes more data and would be even better.

I really didn’t even want to include 2017 in there since it was still in flux, but I figured it was already pretty highly rated in terms of class rankings. I was surprised to see where the recruits actually were on an individual basis. But as you mention, if Xavier can get Scruggs, or if Naji Marshall improves his spot just a little bit, and DEFINITELY if they can land Kris Wilkes the average jumps up very quickly.

[JohnW22]

I don't care what numbers are brought up. Xavier does not have Sumner, Blueitt, and Macura if they're still in the A-10. Xavier's recruiting has been so much better since joining the Big East and they now have the ability to jump in late on some recruits and still land them, if they were in the A-10 they'd never be able to jump in late on 4 stars and still land them.