Illinois Hoops Recruiting Thread (May-June 2017)

[IlliniDent]

3 left is true, been saying with the outstanding offers out there, 3 won't take long to fill. I know you can overload on recruits, give one more than available, but the schollie has to come from somewhere. THT , Finke , and Conditt commit , then Ayo says me too. Staff says come on Ayo, who will leave or how we cover? Little ahead of that issue, but I 'm just curious.

Never has been a problem, there is always a spot for a player like Ayo. There is always some turnover each year

 

[Bubba Eder]

Have to admit I love reading every year people fretting how to come up with scholarships for all the players we will have to turn away. GO ILLINI!!!!

 

[IlliniinIowa]

Have to admit I love reading every year people fretting how to come up with scholarships for all the players we will have to turn away. GO ILLINI!!!!

Thought the exact same thing.

Makes me laugh every time. Hope this time though it comes to fruition. :thumb:

 

[Paigncake]

Don't put all of your trust in me, but from what I'm hearing from someone that has a connection to the staff is that one of our top 2018 recruits will be committing very very soon to the orange and blue..

I'd go with THT on this. Just a guess.

 

[Calvin]

Read the full post

The sheer volume of hs top-10 kids, and therefore drafted players, coming out of Kentucky is impressive, but I don't find that analysis persuasive.

Michigan takes an average recruit of rank 122, and turns that into a 22nd drafted player, but Kentucky takes an average rank 23 and turns that into a 19th drafted player, and that makes Kentucky better?

I'm way more impressed with schools that take lower consensus talent and turn it into gold than Kentucky. Kentucky gets so many top-10 players entering college, it's a given most of them will go to the NBA.

As a study, the sample size is really small. I'll bet if you ran a regression on draft position using h.s. rank and college choice, you'd find Kentucky doesn't fare very well after accounting for talent rank.

 

[HR217]

Have to admit I love reading every year people fretting how to come up with scholarships for all the players we will have to turn away. GO ILLINI!!!!

Every year. It never fails.

 

[Indy Illini Fan]

I'd go with THT on this. Just a guess.

 

[Adnarel]

That emoticon is reserved only for special occasions. I hope you know what you're about.

 

[TriCityBurbs]

 

[DrCS01]

Loyalty after dark

 

[Obelix]

As a study, the sample size is really small.

The study includes all drafted players since 2010, so there is no "sample" in that range, it is the entire population since 2010, the entire Calipari era, which was the original point of the researcher.

The data is actually overwhelming, there is no doubt that based on actual data (all players drafted since 2010), the safest choice for a high-ranked player to the NBA draft is through Kentucky.

Given that the number low-ranked players who did not get drafted in the same period (since 2010) is extremely high, it would actually be very hard to make a case that any school could turn "diamonds in the rough" into NBA players. Kentucky would not care either, their value proposition is not targeted to those players.

 

[SescIllini]

THT could be what Mark Smith was this past year. Not sure how J Goodwin was a "must get" and THT isn't. He seems to have very good hands and a college size body. I think he will be an our first commit.

 

[Calvin]

The study includes all drafted players since 2010, so there is no "sample" in that range, it is the entire population since 2010, the entire Calipari era, which was the original point of the researcher.

The data is actually overwhelming

No offense, O, but your grasp of stats doesn't impress. The data is implying a relationship at the school level, and the confidence level you'll get at that level the report is using is crappy, not overwhelming. If I had the inclination (I don't), I'd run it and show you.

 

[Obelix]

No offense, O, but your grasp of stats doesn't impress. The data is implying a relationship at the school level, and the confidence level you'll get at that level the report is using is crappy, not overwhelming. If I had the inclination (I don't), I'd run it and show you.

LOL... no offense taken, but given that I spend quite a bit of my time on business analytics, I'd say that you are trying to use concepts without really knowing their meaning. Again, there is no "sample," he has taken the entire "population,' every single player that has been drafted since 2010, and he even presented the raw data. So when you say the "sample size is really small," it is totally incorrect.

Furthermore, the study has plenty of statistical analysis and data, pretty overwhelming actually, pointing to the same conclusion. It is a particularly feeble argument to say that IF you did some analysis of your own (which you are not going to do), it would lead to some "magic" and "paradox" and the result will be totally different, yet you know that result already. Pretty weak sauce right there.

 

[OldGreg]

THT could be what Mark Smith was this past year. Not sure how J Goodwin was a "must get" and THT isn't. He seems to have very good hands and a college size body. I think he will be an our first commit.

Mark Smith is a different level prospect than both those guys

 

[Calvin]

LOL... no offense taken, but given that I spend quite a bit of my time on business analytics, I'd say that you are trying to use concepts without really knowing their meaning. Again, there is no "sample," he has taken the entire "population,' every single player that has been drafted since 2010, and he even presented the raw data. So when you say the "sample size is really small," it is totally incorrect.

Furthermore, the study has plenty of statistical analysis and data, pretty overwhelming actually, pointing to the same conclusion. It is a particularly feeble argument to say that IF you did some analysis of your own (which you are not going to do), it would lead to some "magic" and "paradox" and the result will be totally different, yet you know that result already. Pretty weak sauce right there.

You're talking to a guy who works with big data. You're saying you understand regression, correlations, causation, confidence intervals, etc., but your post is...well...lacking in any substance about actual stats supporting the conclusion. No where in that sheet did it show the relevant statistical metrics that would show how much college choice explains deviations from a normalized model for recruit ranking, or what the confidence is for the expected deviation. Point to it or show it, and you'll have some credibility.

 

[Deleted member 8213]

You're talking to a guy who works with big data. You're saying you understand regression, correlations, causation, confidence intervals, etc., but your post is...well...lacking in any substance about actual stats supporting the conclusion. No where in that sheet did it show the relevant statistical metrics that would show how much college choice explains deviations from a normalized model for recruit ranking, or what the confidence is for the expected deviation. Point to it or show it, and you'll have some credibility.

 

[Obelix]

You're talking to a guy who works with big data. You're saying you understand regression, correlations, causation, confidence intervals, etc., but your post is...well...lacking in any substance about actual stats supporting the conclusion. No where in that sheet did it show the relevant statistical metrics that would show how much college choice explains deviations from a normalized model for recruit ranking, or what the confidence is for the expected deviation. Point to it or show it, and you'll have some credibility.

You're not impressing anyone trying to say that you are some master data wiz, yet you fail to provide any analysis but IF you did (which you are not going to do), it would lead to some "magic" and "paradox" and the result will be totally different. I do work extensively and teach business analytics (and R in particular) but that is besides the point. Don't try to use the "I am so good" label on yourself, of some magic result.

On specifics,

1) Obviously, you avoid discussion on totally incorrect statements like "sample size is really small," when the data used actually uses the entire population since 2010 (the Calipari era), every single player drafted, and raw data presented.

2) The analysis is truly overwhelming (and pretty good actually). It does give Kentucky (and others) a huge advantage in metrics such as 'total draft score", which specifically accounts not only for drafted position, but also for the recruits' HS rankings (using 247). Furthermore, it gives Kentucky (and others) a huge advantage in "net score," which also adjusts for "busts", a statistical metric of the top-25 players who went undrafted. So it also accounts for the classic argument (often on this board as well), that some players were busts, high-ranked recruits who did not pan out.

So don't point to a "mythical and fictitious" analysis that would point to a different results (but you are not going to do) or try to impress us with personal statements of how good you are. If you want to present your own statistical analysis to support your "paradox" or contrary result, then do it. For someone as good as your are, and with all the raw data actually at hand (you should thank the person giving you the raw data in spreadsheets), that would be a piece of cake. But my guess is that your big data wizardry will just stop short at yet another post of how good you are, how good your own analysis would have been, and how different your results would be -- IF you did it.

 

[Obelix]

It's interesting here to see the two different recruiting/player development strategies represented here. You've got Kentucky, Kansas, and Duke building teams of one and dones, while North Carolina, Louisville, MSU build teams with guys that stick around in college longer.

Yes, and some other conceptions also confirmed with analysis. It is interesting to note that Michigan ranks 5 (very high) after UK, Duke, Kansas, and Syracuse on "net score," a tribute to Beilein placing players in the draft, often lower ranked (e.g., Burke) while avoiding busts. Much higher than many other higher-considered programs.

 

[Calvin]

You're not impressing anyone trying to say that you are some master data wiz, yet you fail to provide any analysis but IF you did (which you are not going to do), it would lead to some "magic" and "paradox" and the result will be totally different. I do work extensively and teach business analytics (and R in particular) but that is besides the point. Don't try to use the "I am so good" label on yourself, of some magic result.

On specifics,

1) Obviously, you avoid discussion on totally incorrect statements like "sample size is really small," when the data used actually uses the entire population since 2010 (the Calipari era), every single player drafted, and raw data presented.

2) The analysis is truly overwhelming (and pretty good actually). It does give Kentucky (and others) a huge advantage in metrics such as 'total draft score", which specifically accounts not only for drafted position, but also for the recruits' HS rankings (using 247). Furthermore, it gives Kentucky (and others) a huge advantage in "net score," which also adjusts for "busts", a statistical metric of the top-25 players who went undrafted. So it also accounts for the classic argument (often on this board as well), that some players were busts, high-ranked recruits who did not pan out.

So don't point to a "mythical and fictitious" analysis that would point to a different results (but you are not going to do) or try to impress us with personal statements of how good you are. If you want to present your own statistical analysis to support your "paradox" or contrary result, then do it. For someone as good as your are, and with all the raw data actually at hand (you should thank the person giving you the raw data in spreadsheets), that would be a piece of cake. But my guess is that your big data wizardry will just stop short at yet another post of how good you are, how good your own analysis would have been, and how different your results would be -- IF you did it.

Thanks for not pointing evidence. Shows what you know. If you think you're impressing by avoiding the legitimate substance of how small the individual school data is, or how that results in tiny confidence, you're in denial of well known statistical measures.

 

[whovous]

Thanks for not pointing evidence. Shows what you know. If you think you're impressing by avoiding the legitimate substance of how small the individual school data is, or how that results in tiny confidence, you're in denial of well known statistical measures.

I am not a stat guy, but Obelisk most certainly did point to evidence, i.e., the whole Reddit post by the guy from Kentucky. It covers every player drafted from any NCAA school since 2010. That may be a small sample size, considering how few players get drafted each year, but it is also 100% of the potential data pool for the relevant time period.

 

[Obelix]

I am not a stat guy, but Obelisk most certainly did point to evidence, i.e., the whole Reddit post by the guy from Kentucky. It covers every player drafted from any NCAA school since 2010. That may be a small sample size, considering how few players get drafted each year, but it is also 100% of the potential data pool for the relevant time period.

Yes, and actually the analysis is pretty solid, the data and scores do not only account for sheer number of players drafted, but also normalized for HS rankings (using 247), player busts (i.e., top-25 players not getting drafted), etc. It is really well done, especially if people take the time to inspect the analysis in the detailed spreadsheets provided (not just final tables). The best analysis I have seen on a very popular recruiting subject that has often been the center of discussion on this board, and many other, boards. Really deserves another read for people who have not seen it and kudos to Townie Matt for discovering it:

https://www.reddit.com/r/CollegeBas...igh_school_hype_to_nba_stardom_which_schools/

 

[DallasDave]

The question I assume most are asking is "does the school matter in where/whether a top-25 player get drafted?" in which case this analysis includes TOO much data. It has the UK average recruit at 23.32, which means the analysis must include as many non-McD AA's as is it does actual blue chips.

 

[Ga65]

The question I assume most are asking is "does the school matter in where/whether a top-25 player get drafted?" .

This :thumb:

 

[CoalCity]

Now that everyone has impressed us with their statically big weiners can we get back to basketball?