Recruiting Talent vs. Conference Standings

#51      
I'm curious of how this would turn out. I do think it's a slightly different question, but it's one that kind of is a natural next step from here. We now have an idea of what talent was at each program each year, so now we want to know how that talent was used.

I'll get to work on this. Will take some time obviously but I think I'm most curious in how it might change things.

Do you (or anybody else) know of a reliable source for the conference data? sports-reference.com only has the conference stats broken out from the 2010-11 season forward. I could of course just use that season as a starting point, but then we would lose S&C's favorite season in 08-09 ;)

The best source for conference data is the stats publishing on the websites of the different programs and also sports-reference.com. Not sure how far back data is available though, I would imagine there is great correlation or being even identical. Part of the problem is data entry, getting info manually on each player is a pain (I do not envy the amount of time you spent), there must be some database that can be read, but not sure if those are available for free.

PS. Another problem that I found is that the "roster" section on 247 does not often have all individual players linked (or their correct ranking), so it becomes even more tedious having to search each player on each team individually).
 
Last edited:
#52      
The best source for conference data is the stats publishing on the websites of the different programs and also sports-reference.com. Not sure how far back data is available though, I would imagine there is great correlation or being even identical. Part of the problem is data entry, getting info manually on each player is a pain (I do not envy the amount of time you spent), there must be some database that can be read, but not sure if those are available for free.

PS. Another problem that I found is that the "roster" section on 247 does not often have all individual players linked (or their correct ranking), so it becomes even more tedious having to search each player on each team individually).

The 247 part was a bit annoying when getting this data for that very reason. Once I had all the rosters laid out, I basically just went through their historical recruiting classes and took their rating information from there, then did the player search to fill in the gaps for transfers, converted walk-ons, etc. It was kind of a pain but it did give me something to do during some slow hours at work.
 
#53      
It all kinda depends on precisely the question you're trying to ask, right? You weight it in the way you've suggested and the question becomes "to what extent did highly ranked players contribute to the various teams?".

We now have an idea of what talent was at each program each year, so now we want to know how that talent was used.

Personally, I do not believe that HS ranking is the sole determinant of "talent." Talent extends into many other things as individual accolades and performance proofpoints (please see my previous post on golden eras). But even in the case that we use HS rankings as the sole surrogate (which I disagree), the question that we are analyzing is not how ranked players were used. That can be answered pretty deterministically, we know. What we try to do is use Effective Ranking as a predictor (independent variable) to analyze and investigate correlation with B1G performance (dependent variable, which we also know deterministically for each year).

The reason it would be more interesting is that Effective Ranking is closer (although not the same) with people's view of the loaded term "talent" (in past seasons). I do not think that when someone says (correctly or incorrectly) that Missouri was very talented last year, it has much to do with Michael Porter. Or someone can say, we were so talented in those years (with Groce) because of Tracy (in years he did not play). Given that injuries, suspensions, on and off-court problems, player/coach issues, etc. etc. are so prominent, it helps towards the analysis in that direction. Yet, since this measure does embed HS rankings, the same problems and bias will still exist IMO. If I had to take a guess, as with individual rankings (which is the basis for all this), the correlation will be greater at the high end rather than lower end. Correlation also does not mean perfect correlation either, even at higher end.

To relate all this back to recruiting, it does not mean that you will find "talent" only high up in the rankings. There are many very talented players who were not as highly ranked. But chances are higher, that you will find "talent" at the high end of the rankings. That's the notion of correlation (stronger at the high end). Anyway, every interesting discussion so far.
 
#54      
The 247 part was a bit annoying when getting this data for that very reason. Once I had all the rosters laid out, I basically just went through their historical recruiting classes and took their rating information from there, then did the player search to fill in the gaps for transfers, converted walk-ons, etc. It was kind of a pain but it did give me something to do during some slow hours at work.

I hear you, I did Effective ranking for just Wisconsin and Illinois last night and I was surprised that some players on the 247 roster showed with no ranking, yet when you searched them individually in 247, they did have a ranking.

For example, you at Wisconsin roster last year on 247,
https://247sports.com/Team/Wisconsin-Badgers-Basketball-86/Roster

Charles Thomas IV has no ranking. But if you search individually (still on 247)... Voila! (he has a ranking)
https://247sports.com/player/charlie-thomas-76194

There are others too.
 
#55      
That makes me think that a conference record between 8-10 and 11-9 is what should be expected.
Anything outside of those bands will say something about coaching or injuries, etc. etc.

IMO everyone here should be paying close attention to Obelix's posts and understand why we shouldn't make statements like this. I will try to add on and say it in my own words (probably not as effectively). There are 3 huge factors that these graphs don't address.

1) Positional gaps. More specifically, positional gaps AT THE 2 MOST IMPORTANT POSITIONS. A real killer. If Trent Frazier is in the same class as Malcolm and Kendrick then Groce is still here. Let's go one further. If Trent and Tilmon are in the same class as Malcolm and Kendrick then we would've been a B1G contender.

2) Not all rankings are created equal. JCL and Ayo were both ranked around the same (35ish IIRC). Any coach would take Ayo over JCL if they had to pick. Other rankings are based more on potential rather than readiness to compete at the collegiate level (Meyers).

3) Age. Experience is important.

If BU continues recruiting like the last 2 then he will also be gone in 3 or 4 years.
 
#56      
2) Not all rankings are created equal. JCL and Ayo were both ranked around the same (35ish IIRC). Any coach would take Ayo over JCL if they had to pick. Other rankings are based more on potential rather than readiness to compete at the collegiate level (Meyers).

Ayo 32, JCL 37 in the RSCI, so there's a little gap there, but it is just flatly, categorically untrue that every highly ranked recruit we've had over the past decade was knowably predestined to be a huge disappointment and a lesser, more limited player than their rankings indicated. That is the big lie, take the orange pill and see that it is not so.

We've had a bit of bad luck, and this stuff about positional balance is certainly true and meaningful, but that is just another thing that folds into the real explanation, the thing that stares us dead in the face: shockingly, uniquely incompetent coaching staffs. Weber & Co just found ways to do everything wrong, Groce & Co were utterly in over their heads at this level, and Underwood has gotten off on the wrong foot.

The good news is what we have seen is a special, extreme kind of bad. No reason to think it's going to continue.
 
#57      

SKane

Tennessee
I wouldn't say that Weber & Co found ways to do "everything" wrong. Lots of Weber recruits developed and improved - McCamey, Paul and Leonard, for instance. Several lower-ranked recruits improved - Davis, Tisdale, and Frazier. Even Brock. Richardson did not progress much offensively but did progress defensively. There were some very bad misses too (Head, Shaw, Legion). I just think that "everything" is overstating it.
 
#59      
Ayo 32, JCL 37 in the RSCI, so there's a little gap there, but it is just flatly, categorically untrue that every highly ranked recruit we've had over the past decade was knowably predestined to be a huge disappointment and a lesser, more limited player than their rankings indicated. That is the big lie, take the orange pill and see that it is not so.

We've had a bit of bad luck, and this stuff about positional balance is certainly true and meaningful, but that is just another thing that folds into the real explanation, the thing that stares us dead in the face: shockingly, uniquely incompetent coaching staffs. Weber & Co just found ways to do everything wrong, Groce & Co were utterly in over their heads at this level, and Underwood has gotten off on the wrong foot.

The good news is what we have seen is a special, extreme kind of bad. No reason to think it's going to continue.

My point in comparing JCL to Ayo is not to say JCL underperformed. I would argue JCL did exactly what he was supposed to do. My point was players are ranked highly for different reasons. JCL was ranked highly for his lights out shooting, and that's what he did and I wish he didn't transfer. But Ayo is ranked highly for his ability to take over a game and make all his teammates better. The rankings are similar but Ayo has a much better chance to influence the team because of his position and skillset.
 
#60      
I would argue JCL did exactly what he was supposed to do.

That's a pretty weak argument IMO.

JCL should have been a much, much better player at Illinois than he was. Nothing is a guarantee, plenty of guys even at the very top of the rankings end up as busts, but any reasonable expectation for JCL would have been for much more than we got.

Here's guessing his DePaul performances will make that even clearer.
 
#62      

Illwinsagain

Cary, IL
Really good stuff from Bartovik. He has us at 16-13 and 10-10 with a 11th place B1G finish.


http://barttorvik.com/team.php?team=Illinois&year=2019
Missing the 2nd and 3rd game in Maui, since he can't know who that will be. Splitting those 2, beat Mizzou, take 1 more in conference (PSU?), I'll take the rest.

19-12, 11-9.

Edit: He actually has us in a 6 way tie for 5th in conference, just the worst ranking of the others. NW, one spot ahead in the rankings has a 9-11 record in conference. I was wondering how 10-10 could be 11th.
 
Last edited:
#63      

Tevo

Wilmette, IL
What all of these charts show is pretty clear: The rating services are intentionally "over-ranking" the players that Illinois will eventually sign, in a concerted effort to keep us from our rightful place among the Blue Bloods.
 
#64      
There are so many variables in recruiting, talent, and team record it is ridiculous. Try all you might it is hard to predict how each recruit will perform for any given team. Teams with unranked players can beat a team of ranked players. Cinderella teams can make it to the final four. North Carolina can win 3 championships in the last 15 years. Sure, a team with highly ranked recruits (aka the term most people use to determine talent) usually has a greater chance of winning compared to lower ranked recruits, but there are so many other variables that need to be accounted for to more accurately predict an outcome of a player, a game and a team. That is why gambling exists, no one can completely predict all the variables (except maybe illinoisloyalty members ;))
 
#65      
What all of these charts show is pretty clear: The rating services are intentionally "over-ranking" the players that Illinois will eventually sign, in a concerted effort to keep us from our rightful place among the Blue Bloods.
I'm pretty sure this was posted in jest, but it raises an important point IMO. While we can use the recruiting ratings from a service like 247 or even a broader aggregation of evaluations like RSCI as a proxy for any given team's talent, it's still only a proxy based on highly imperfect qualitative information, and as such is difficult to use.

Consider the success rate for, say, NBA scouts in identifying talent. They have the advantage of looking at players who are more mature and who have typically logged significant game time in a much more level and predictable environment than those scouting high school juniors, and their evaluations still miss at a pretty high rate. The probabilities that kids in the top 20 or 30 of the RSCI will be productive in college are pretty strong, but that's a fraction of the 1000-odd kids who sign with NCAA teams each year. Evaluation becomes quite a bit more ragged after that, and gets there in a hurry.

This means a couple of things to me. First, I don't think that it's particularly accurate to assume the talent level of a team is directly and most cleanly related to how kids were ranked coming out of high school and that any failure to live up to the rankings is a result of what came after. I'm more inclined to say that we know more about a player's raw talent level if we consider what they produce at the college level than if we don't. That doesn't mean that player development doesn't matter, among other things, but it seems odd to exclude data points that are frankly quite a bit better than what we have at the time a player signs from the analysis.

The second thing is that if you acknowledge just how hard it is to evaluate the talent level of high school kids, the argument that ratings are biased towards or against certain groups of teams starts to hold some water. And it's possible that it's not a coincidence that IU and Illinois are the two teams showing the biggest underperformance in the league. If you're a long-term national observer, your opinion of those two programs is probably a good deal higher than those who suffer through watching, say, Illinois day in and day out. (To say nothing of the opinions of kids who have only been watching college basketball for five or ten years.)

I don't think that player ranking data really allows for that hypothesis to be proven with any confidence, so that's just conjecture just like everything else is when we talk about recruiting. My personal opinion is that Illinois has, in fact, underperformed its talent level over the past decade, but also that the incoming talent level was not as good as indicated by the ranking services. I apologize because that's not a particularly interesting observation, but I have a tough time believing that we've had first-division B1G talent all along based on what we've actually seen happen on the court.
 
#66      

JFGsCoffeeMug

BU:1 Trash cans:0
Chicago
What all of these charts show is pretty clear: The rating services are intentionally "over-ranking" the players that Illinois will eventually sign, in a concerted effort to keep us from our rightful place among the Blue Bloods.

Wow, you're right! And it's even worse than what you've described. We know our recruits always drop in the rankings after they commit. So, really, the rating services are consistently over-over-ranking our recruits so that they continue to be over-ranked even after the drop. Why is everyone out to get us? :cry:
 
#67      
Also want to be clear that I'm not dumping on the work that the OP has done -- it's been an interesting read and a fun exercise, so thanks for pulling it all together.
 
#68      

Tevo

Wilmette, IL
This means a couple of things to me. First, I don't think that it's particularly accurate to assume the talent level of a team is directly and most cleanly related to how kids were ranked coming out of high school and that any failure to live up to the rankings is a result of what came after.

One thing for sure is that "talent" is not universal. A kid who has the right skills for a given system may or may not fit in well in another system. I think that's the source for Ryan and Wisconsin so consistently "outplaying" their talent -- because he was excellent at identifying guys for his system and then getting them to buy in and do things his way. I think Beilein does that too. Not that they don't get some good players, but they really have a skill in getting the players they have to mesh into a team. Which is, you know, coaching.
 
#69      
One thing for sure is that "talent" is not universal. A kid who has the right skills for a given system may or may not fit in well in another system. I think that's the source for Ryan and Wisconsin so consistently "outplaying" their talent -- because he was excellent at identifying guys for his system and then getting them to buy in and do things his way. I think Beilein does that too. Not that they don't get some good players, but they really have a skill in getting the players they have to mesh into a team. Which is, you know, coaching.
Yes, that's all true, and I would agree that coaching plays a huge role in a player's eventual success. (Not that you needed me to pull out a telescope and show you the sun, right?)

I don't think that fact does anything to make the 247 composite or RSCI a better dataset, though, and it's likely that the perception that Wisconsin gets meh players and coaches them up has a self-reinforcing impact on the talent gurus' evaluations that make up that data. It's the best data we have, but it's not nearly good enough to provide a reliable baseline that allows us to isolate the impact of coaching on player development.