Recruiting Nerds

[hoosierpap]

I had someone (who all know, but I didn't clear this with him) PM me regarding some questions about recruiting in general and then, more specifically, Anthony Leal's situation. The general questions were about offering scholarships. Below is my experience/knowledge of the situation. It's unnecessarily long, and gets unnecessarily techy at the end but I figured it may generate some discussion while the on the floor product has been less than desirable. 

 

 

I don't think you're going to love this initial answer- but I'll explain- each coach/program has their own style. There are no hard and fast rules to the approach of scholarship offering. The factors are all there; the when, the why, the who, how many, etc. You have some coaches/programs who take the (infamous to us) shotgun approach which starts early in classes. You have coaches/programs who take the opposite approach and don't offer anyone until a certain criteria is met. And then you have, in my estimation, the programs that operate in between those two extremes and are malleable to the individual recruitment and class composition/quality. 

In the first scenario (ex. Crean), the idea is to get the offer out when they're younger, or showing some potential. The idea is you show "we're serious" initially. By the end, the approach is "we knew it before others, we were sure, we were right, you can trust my judgement, we've been around longer, we're family at this point." The problem with this approach is it's simply a numbers game which most can see it (these kids all talk, all know each others offers) and this is where you get the "cheapened offer" talk. "Oh, he believes in 35 of the top 100 just like me?" The overall idea is you hit on some, you miss on others both for commitments and impact on the floor. 

Now, sometimes this stuff really pays off. Some kids/families buy into that. To some kids, it's truly important. That probably has to do more with biological and socioeconomic factors than anything else- but can be effective. 

In the other extreme, there are coaches/programs where the approach is much more selective and patient. For example, notoriously, John Beilein, does not offer a kid until 3 things have happened: the recruit has been on campus, John Beilein has seen the recruit play in person, and the recruit has completed his sophomore year of high school. Now, that's about as strict/selective and patient as it comes currently. It instantly takes UM out of recruitments especially in a time where kids are stacking offers for no other reason than to measure up to their buds. Think what you will of that mindset, but it's a reality, and plays into things. He doesn't waiver from the approach though as it allows him to evaluate the prospect with the idea of offering the right scholarship to the right prospect for the right results on the floor. 

This approach obviously strengthens the weight of the offer. Some kids/family's value the heck out of that. They value that the decision has been thought out on the programs end and there is substance and meaning to it.

Another example, Stanford doesn't offer until you get accepted to their school. That certainly limits them, but can have some serious appeal.

Most coaches/programs operate in the middle of that spectrum. If a prospect is elite- why wait? MOST depends on the family/camp/prospect and what you gather is important to them. Is this kid a bit of an attention seeker or someone in his camp feels disrespected and value an offer even if it isn't actionable? Offer. Is this kid not focused on recruiting and an offer would have more weight when family/camp/prospect is getting more serious? Hold off. Does a coach really see potential which may be exploitable with an early offer? Offer. And so on and so on, but this approach is most common.

With regards to slow playing a kid...it's tough but they know what they're doing. Generally, you have SOME insight into where a kid is in a recruitment before you offer. So, if there is no chance of a commit coaches/programs can make offers that aren't actionable- it's a mutual thing. For coaches/programs it means- "we're serious, we're going to be around, consider the idea of putting on this uniform."

Most times that all goes unsaid. In certain situations, say a blue blood is offering a fringe 4 star w some unseen potential it's more "we're serious, we're watching, here's what you need to work on, work hard enough and you can succeed here." In more rare situations, like Anthony Leal, it's well known he will commit to IU at anytime. IU is genuinely not sold on Leal. It's not that they know he's in their pocket and they're just going to keep trying until he's the last girl at the bar. You have to do it with respect and I can tell you this is the approach I have heard to this fairly unique situation. The pitch to Leal is "we like you as a player, as a person and we care about you and your family. We want the best for you and for this program. Here's what we'd like to see develop, we want like hell to have you here but we want to make sure you're in the best spot to succeed on the floor." Now some of that is because it's not only in IU's best interests to secure top talent to maximize success, but it's also a pretty bad look around the state if the hometown boy with parents employed by the school can't get in a game and has to transfer. I think Archie operates on his own accord. Some coaches pay attention to other offers, maybe the upcoming trip to MSU for Leal DOES impact the offer/not offer- I just tend to doubt it. It would certainly in some coaches.

You may have no interest in this next part but it's been I've been thinking about since I was back in college. I should probably just try it and track it on my own. All recruiting at this point is subjective. The shotgun approach is less informed subjectivity. The Beilein approach is more informed- but nonetheless- subjective.

I'm fairly convinced that there is a science to it. I work in a field where computers and people merge to predict behavior and determine outcomes. I'm fairly certain that an algorithm could be programmed to more accurately yield desired results based on personality traits, socioeconomic factors, brand influence vs accolades, performance, growth of performance, quality of opponent, and a myriad of other factors. A "Moneyball" approach, if you will. I think there would be resistance to this but using empirically data-based analytics to accompany subjective feelings/assessments would be a useful tool.

Just today, I saw Dakich on Twitter defending his take that Porzingis would be a bust because of "his eyes" (meaning Porzingis') and endlessly, laughably arguing that anyone who wasn't a coach "wouldn't get it." It's that type of subjective nonsense that leads to a lot of swings and misses, imo, and analytics could be used as a tool guide coaches in recruiting.

[RatpigHoosier]

Dropping some serious knowledge on us. Thank you. This is obviously well researched and long observed.


Sent from my iPhone using BtownBanners

[mdn82]

It’s not a bad idea. AI or BI goes a long way in the business world. Predictive models based off personality traits aren’t new. But finding a formula that closely resembles your wants/needs and predictability could be fantastic. There are few fanbases in America than this one that would be pissed at that approach. That’s enough for me to try. The people that use their eye test would hate it on another level. Fantastic

 

 

Sent from my iPhone using BtownBanners

 

[Stromboli]

Appreciate the insight.  USPS likely already has the algorithm for recruits, programmed into a supercomputer in his basement CBBHQ. 

 

I can tell you that seeing the guys play has enormous value for me. At EYBL the talents emerges clearly. And it's hard, because these kids are 15, 16, 17 years old, so the kid you see in April isn't the same kid as in December. 

 

Nice to hear the professional way Archie handles this stuff. 

[HoosierTrav]

10 minutes ago, mdn82 said:

It’s not a bad idea. AI or BI goes a long way in the business world. Predictive models based off personality traits aren’t new. But finding a formula that closely resembles your wants/needs and predictability could be fantastic. There are few fanbases in America than this one that would be pissed at that approach. That’s enough it for me to try. The people that use their eye test would hate it on another level. Fantastic


Sent from my iPhone using BtownBanners

I rely in AI and machine learning in my business, to deliver a much higher quality prospect. It gives me a competitive advantage. I’d love to see this develop. 

[Rico]

And now "metrics" have become involved in the recruiting process!  Can we skip that and just go to some DNA test?  LOL

[mdn82]

And now "metrics" have become involved in the recruiting process!  Can we skip that and just go to some DNA test?  LOL

I would love to see various coaches and their recruiting boards to see how they currently rank and rate their wants and needs. Most of the AI data is already gathered. It’s more about building it out really. They could use previous years as a test to see how well the model could work. Honestly it’s a fantastic idea and one that I am sure a few coaches probably utilize. Only schools like Duke that don’t have to be as selective aren’t as concerned with finding the right player. They can get whoever they want.


Sent from my iPhone using BtownBanners mobile app

[hoosierpap]

46 minutes ago, Rico said:

And now "metrics" have become involved in the recruiting process!  Can we skip that and just go to some DNA test?  LOL

 

[Walking Boot of Doom]

Pap -

Do it. You could likely partner with 247 or such as they have a vast database and the resources to keep the data refreshed. Pitch the concept of D&A (not DNA...) as a service to programs - could be a differentiator especially for mid-majors needing every advantage to succeed in recruiting.

Ready or not, humans and how things work are becoming more and more predictable using analytics and AI powered models. However, Dakich and his eye tests will continue to reign supreme on the Twitter scoreboard.

Joking aside, as Trav mentioned above, powerful analytics and deeper knowledge will only empower those with the ability to identify the right talent even more. Improvements in programs recruiting the right players would only improve the quality of college basketball as a whole.

[hoosierpap]

5 hours ago, Walking Boot of Doom said:

Pap -

Do it. You could likely partner with 247 or such as they have a vast database and the resources to keep the data refreshed. Pitch the concept of D&A (not DNA...) as a service to programs - could be a differentiator especially for mid-majors needing every advantage to succeed in recruiting.

Ready or not, humans and how things work are becoming more and more predictable using analytics and AI powered models. However, Dakich and his eye tests will continue to reign supreme on the Twitter scoreboard.

Joking aside, as Trav mentioned above, powerful analytics and deeper knowledge will only empower those with the ability to identify the right talent even more. Improvements in programs recruiting the right players would only improve the quality of college basketball as a whole.

Yeah, I'm by no means saying it should entirely replace the coach in recruiting. Hell, the coach is a huge source of data collection in the model. It would just be a collaborative tool as it is in professional sports- operating on a spectrum- some coaches/programs value it in differing degrees. I get the backlash on a site with a user base with such diversity but analytics, big data, AI, is undeniable.

Ultimately, this thread was about recruiting approach and I don't think the original user who PM'd me even cared about the last bit.

[BGleas]

When I worked for the Celtics, we’re talking about 14 years ago, I shared a cubicle with one of our analytics/tech guys and he would share with me their dashboards for college prospects. At the time Daryl Morey (current Rockets GM and the first “analytics GM” in the NBA) was our EVP of Technology, and he led the team to develop a platform for predicting how college players would do using all these formulas, statistical techniques, etc. 

They created a software for inputting player data, including class, position, # of college games played, height, all of the players stats, and probably a ton more stuff and it would determine for them how successful the player would be based on historical data on similar players.  

It was all pretty fascinating. 

[Class of '66 Old Fart]

11 hours ago, hoosierpap said:

Yeah, I'm by no means saying it should entirely replace the coach in recruiting. Hell, the coach is a huge source of data collection in the model. It would just be a collaborative tool as it is in professional sports- operating on a spectrum- some coaches/programs value it in differing degrees. I get the backlash on a site with a user base with such diversity but analytics, big data, AI, is undeniable.

Ultimately, this thread was about recruiting approach and I don't think the original user who PM'd me even cared about the last bit.

Wasn't that he didn't care, just so far over his head he was clueless.