mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-13-17 02:04 PM - Post#233929
In response to TheLine
I have an Ivy recruiting database that spans back to the 2002 class (the first one where scout and rivals really blew out their tracking to the mid level) and includes a little north of 600 recruits. That database also includes the outcomes from all of those recruits in win shares (offensive, defensive, total), offensive and defensive rating, PER, FIC, HOB and overall games/minutes played.
What that allows me to do is empirically demonstrate the relationship between ratings and actual output over time. There is a weak, but significant, relationship between the recruiting ratings and future output. The relationship tends to bucket recruits into three different levels:
No rating/maybe one outlet has him listed at all
Some rating (1*, 2*, 2+*)
Consensus (i.e. average) 3* or better
We have had 18 consensus three-star recruits since 2002. Here is that list:
Team Player Year
Harvard Zena Edosomwan 2013
Harvard Bryce Aiken 2016
Princeton Jaelin Llewellyn 2018
Princeton Harrison Schaen 2003
Harvard Chris Lewis 2016
Harvard Spencer Freedman 2018
Harvard Seth Towns 2016
Yale Jordan Bruner 2016
Harvard Wesley Saunders 2011
Harvard Robert Baker 2016
Harvard Kenyatta Smith 2011
Harvard Siyani Chambers 2012
Penn Justin Reilly 2006
Penn Ryan Pettinella 2003
Yale Sam Kaplan 2003
Harvard Michael Hall 2012
Columbia Ben Nwachukwu 2004
Penn Steve Danley 2003
Out of 16 players that actually have production to track, seven either finished their careers with 10 win shares or more or are on pace to do so using their career-to-date observed production. For the remaining 500+ that have production to track, we have 42 players that either hit 10 win shares or are on pace to. 21 of those came from the 1 star to 2+ star group which encompasses 144 recruits with some performance to track. The final 21 came from the 410 recruits that didn't average at least a star across the different sites.
The odds of getting a 10 win share player by group then are:
No rating/less than 1-star: 5%
1*,2*,2+*: 15%
Consensus 3*: 44%
Mike Wang is not a consensus 3*. He is a 3* at ESPN. He is a 3* at 247 (which is now sorta Scout). He is an NR at Rivals. He is not listed at Future 150. He is in that middle group.
The next layer of this is to add in real offers, as when I look down the list of the relative busts, there are a lot where the offer list (and FWIW, what the coaches were saying too) lines up better with the relative lack of productivity than the rating assigned to the player. The opposite is also true. Sometimes there are 2* players that have much stronger offer lists and interest, and those players tend to perform better than expected.
The combination of this empirical work and conversations with those who would know gives me the confidence to make projections and to try to provide useful information to fans that genuinely want an accurate representation of their team - not just the rosiest view of the situation.
I'm genuinely confused by a lot of the comments in this thread that seem to be accusing me of being a troll and basically trying to smear any team that isn't Harvard. Anyone that knows me or is taking a true, holistic view of what I share knows that I care about being right above all else.
While I know the longer-tenured members on this board know this, for the newer folks, what I share on here is a blend of empirical work and source info. I try to make it clear when I'm sharing my own gut opinions, but I don't post many of those, because I like to be right. Anyone that attempts to paint me as having an agenda or being malicious really doesn't get what I do.
|
westcoast
Senior
Posts: 302
Reg: 03-08-16
|
10-13-17 02:48 PM - Post#233932
In response to mrjames
Great post.
|
AsiaSunset
Postdoc
Posts: 4358
Reg: 11-21-04
|
10-13-17 03:38 PM - Post#233934
In response to westcoast
Mike - I'm not accusing you of being a troll; however, for a guy that I consider a straight shooter, your post was incredibly disingenuous.
NR means not rated and nothing more. It does not equate to a 1* or a 2*. Consensus 3*** is a term you've invented. There are plenty of 3 stars in the Ivies not on your list. They may not have averaged out to 3*** because they had a NR or a 2** on a particular site. That's often the case with Ivy recruits. I can think of a bunch in the league today. Names that come to mind are Cannady, Bell, Boudreux, Brodeur, Smith and a bunch on Harvard players besides the names you listed. I've probably missed a few others.
Mike Wang is not only a 3***, he's rated at the very top of the 3 star category. To suggest he is anything less is both demeaning and inaccurate.
To state that he's not on your list of "consensus 3***" is accurate, but basically meaningless.
I don't know Wang's complete offer list because none of the sites are really that up to date. His high major offers were in the fall of 2016. I didn't even know we were recruiting him until I read it on this board.
|
PennFan10
Postdoc
Posts: 3584
Reg: 02-15-15
|
10-13-17 05:11 PM - Post#233943
In response to AsiaSunset
Great post.
|
Mike Porter
Postdoc
Posts: 3618
Loc: Los Angeles, CA
Reg: 11-21-04
|
Mike Wang to Penn
10-13-17 05:13 PM - Post#233944
In response to Okoro Dude
Agreed - definitely don't take anything that Mike says as trolling. He always seems to be imparting the most accurate information he has on a topic. That doesn't mean Penn fans will always like it... truth is that we are behind HYP in recruiting and sometimes the truth hurts and we don't like hearing it but that isn't trolling. Plus of course we all have our biases. This also doesn't always mean his information will also lead to what plays out in the future.
Personally I like Mike Wang on the team and it's good he is ranked by a few of the sites (his offeres definitely aren't up to date though) and that's more indicative that he will be successful. Much better than being non-existent on those sites. He just isn't ranked by all of them (per the list above not many are ranked by all), so that's another piece of info.
Let's just hope Mike Wang exceeds expectations and then we will all be happy. 
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-13-17 05:42 PM - Post#233945
In response to AsiaSunset
My methodology is as follows (and I've shared it in articles and podcast episodes before):
I average the rating provided by the prevailing recruiting websites of the time - for a while that was Rivals, Scout and Hoop Scoop, then ESPN joined the fray then Hoop Scoop left and 247 and Future150 joined. I take the average of stars across those sites, and that determines the final rating. A consensus 3-star is a recruit is a recruit that averages a 3.0 or above from those systems. I set definitions like this to avoid data mining, which can lead to overfitting.
You're ability to come up with anecdotes is a common counter, but I could pull just as many from the database that were 3-stars in some/many places but below 3.0 on average who were the duds to your studs. That's the problem with anecdotes that having a 600+ Ivy recruit database can help solve. We can observe large sample averages instead of availability-bias-driven counters to real, time-intensive, empirical work.
As I always say, you can believe whatever you want. My work is available here, in podcasts, on twitter and on my old blog. For those that judge it to be useful, they can view this work in that context. It's always funny, though, how I'm a straight shooter until someone doesn't like my answer. HA!
|
Bill Lewis
Senior
Posts: 304
Reg: 12-23-04
|
10-13-17 05:48 PM - Post#233946
In response to mrjames
Mr. James
A year or two ago you posted that one of the rating sites was better than the other sites. I don't recall which site you listed as the best. Could you please refresh my memory? Thank you.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-13-17 06:02 PM - Post#233948
In response to Bill Lewis
IIRC, Scout and Rivals were the best but ESPN was making up serious ground quickly. Hoop Scoop was a joke.
247 Sports didn't have enough runway to get a good sense and now that it's merged with Scout, I don't know what to say there. The average outperformed all of them, though. Wisdom of crowds, as it were.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-13-17 06:10 PM - Post#233950
In response to mrjames
Also, just so we're clear, "trolling" would be saying something like: "It's gonna be funny when Penn can't even qualify for the Ivy Tourney the last time it's held at The Palestra before getting moved around."
|
PennFan10
Postdoc
Posts: 3584
Reg: 02-15-15
|
10-13-17 06:53 PM - Post#233953
In response to mrjames
Yea, trolling is not how I would characterize it. That is more trying to rile up the home board for no good reason.
You are riling up the home board, but it's generally within reason!
|
AsiaSunset
Postdoc
Posts: 4358
Reg: 11-21-04
|
10-13-17 07:18 PM - Post#233954
In response to mrjames
Mike - your post is beyond ridiculous. You posted that Michael Wang was a low rated player because his Rivals page hasn't been updated since the fall of 2016. He is clearly a 3***.
You can devise whatever system you choose, but that doesn't change that fact. There are no andidotes here or in any of my posts above. Simply the facts.
Now had you said Mike Wang's game is less attractive to the typical power 5 team, I would agree. Those teams seem to focus on athleticism. And - had you said there are several other Ivy recruits who are more athletic and therefore more attractive to power 5 teams, I would also agree. But that doesn't make Wang a roll player or a less significant piece of the puzzle for Donahue. Nor does it negate the fact that he's been rated a high 3*** recruit.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-13-17 08:22 PM - Post#233956
In response to AsiaSunset
Despite the fact that we're having a chat about recruiting ratings and the model, which I hope I've explained clearly, the reason why I don't have Wang amongst the Top 5 and likely Top 10 recruits in this class has nothing to do with those.
|
AsiaSunset
Postdoc
Posts: 4358
Reg: 11-21-04
|
10-13-17 09:40 PM - Post#233957
In response to mrjames
Mike - you should understand that where you rank Wang has very little significance because, unlike your statistical work which is excellent, you have no special creds when it comes to this stuff..
|
UPIA1968
PhD Student
Posts: 1120
Loc: Cornwall, PA
Reg: 11-20-06
|
10-13-17 11:11 PM - Post#233958
In response to AsiaSunset
This is what we don't know for sure in this debate: how good Mike Wang will be. We also don't know for sure how accurate Mike James' predictions will be about the likely performance of 18-year-old rookies in the challenging world of college basketball. There simply isn't enough good data about their performance in high school or the transferability of that skill to college, save for a tiny minority of players, none of whom go Ivy. So agonizing over whether Mike Wang is a true "3" might fill boring October evenings, but it tell us little about what he will do in the Cathedral of College Basketball in a little over a year.
Here's what we do know for sure. Mike Wang is a better than average Penn recruit, easily good enough to be in the group that has a decent chance of playing well in the Ivies. If Donahue can keep getting such players, as Franny did, we will be happy.
We also know for sure, damn sure, that Mike James is the best, most reliable analyst of this stuff, whether recruits or active players. Sure he is wrong lot's. As a professional forecaster, I know all about what it is like to be wrong, frequently. I just hope that my customers understand that I am usually somewhat less wrong than my competition. I for one know that Mike James is less wrong than most of the rest of us.
|
SomeGuy
Professor
Posts: 6404
Reg: 11-22-04
|
10-13-17 11:25 PM - Post#233959
In response to mrjames
Yes, it certainly appears that you are separating out Wang from guys who are lower rated overall like Djuricic and Kirkwood, just to make it about Penn and Harvard. The question for me is why you do that. Wang does well in the ratings, and he does well in terms of offers. So it seems like there must be a word of mouth component to the evaluation. You often seem to have pretty good info on these sorts of things. However, calling him a likely role player seems questionable to me -- i really doubt Penn is bringing him in with that in mind.
|
section110
Masters Student
Posts: 847
Loc: south jersey
Reg: 11-22-04
|
10-13-17 11:30 PM - Post#233960
In response to SomeGuy
Help us out here. I don't see any of these recruits play in high school or AAU; I doubt any of us sees them or many of them The consensus in the entire world is that Sebastian Much is going to be the rookie of the year. Per Verbal Commits, he was rated 3* by ESPN, unrated by Scouts and Rivals and a composite 2.3*. Michael Wang is 3* ESPN. unrated by Scouts & Rivals and a composite 3* and not a legitimate consesnsu 3*?
|
SomeGuy
Professor
Posts: 6404
Reg: 11-22-04
|
10-13-17 11:46 PM - Post#233961
In response to UPIA1968
Following up on UPIA1968's post, I tend to view what recruits become in terms of odds. While Mike has suggested Wang isn't as good as his ratings and offers, I assume Mike would say there is a significantly above zero chance he develops into an All-Ivy player. And while Asiasunset calls BS on Mike's evaluation, I very much doubt he would say that Wang is a lock for All-Ivy. There may be significant difference in how each would handicap his odds for eventual All-Ivy. Mike could say 10% and Asiasunset could say 50%. Even with that huge difference, to me the ultimate data point of what happens with Wang is by itself a nearly meaningless anecdote. If he makes All-Ivy, it may in reality have been a long shot; if he does not, that may have been a bad coin flip. Chris Egi could have been an 80% chance to be a star. Maybe Harvard rolled the dice with the odds in their favor and lost on that one.
Where this matters for everyone but Harvard right now is the simple math that Harvard is getting more guys with that higher percentage shot to be All-Ivy. Egi doesn't have to be All-Ivy for Harvard to win. But AJ Brodeur probably has to for Penn. at least right now, Penn has more pressure to hit on the higher ranked recruits, because we get fewer bites at the apple. And that means people are going to react when you are critical of Wang, because there is simply more pressure for Penn to be right.
|
SomeGuy
Professor
Posts: 6404
Reg: 11-22-04
|
10-14-17 12:04 AM - Post#233962
In response to mrjames
Curious who the top 10 are right now. Seems like Llewellyn and Freedman are right at the top, and Cotton and Kelly are next among the 3 star guys. I think there are some 2s in VC that might be better than the other 3s -- namely Kirkwood, Forbes, Wight, and Princeton's other wing. Wang and Ellis seem like they would round out a top 10.
|
Mike Porter
Postdoc
Posts: 3618
Loc: Los Angeles, CA
Reg: 11-21-04
|
10-14-17 01:19 AM - Post#233965
In response to SomeGuy
FYI I wouldn't trust VC ratings these days. Honestly doesn't seem to always be aggregating right / updating with latest details.
|
AsiaSunset
Postdoc
Posts: 4358
Reg: 11-21-04
|
10-14-17 07:40 AM - Post#233966
In response to UPIA1968
Just to be clear about what I've said and what I think, I have no idea whether or not Mike Wang will play at a level that deserves all Ivy consideration. If you look at past All Ivy teams you'll find certain players who emerged from high school with big reputations and others who entered virtually unnoticed. This is not anecdotal. It's factual.
If you look at MJ's "concensus" 3*** list in this thread you'll see some hits and misses. Look at our guys - Justin Reilly and even Steve Danley. Steve was a fine player at Penn; however, Steve played with Ibby, Mark (both All Ivy) and even Grandieri and a young Tim Begley (who later became Ivy POY). You don't see these other guys on his list. Importantly - even though win share is designed to measure an individual's impact, it's really a very misleading statistic in this particular sport. The points you score, the assists you get and ever your offensive and defensive rebounds are very team dependent.
I think All Ivy is a better measurement of success but we all know even that's imperfect. Sometimes politics plays a role.
So - to be clear - I think MJ's knows more about Ivy basketball than anyone else on this board, but his abilities don't extend to evaluating hs players potential as college players, at least not to the extent he is better at it than anyone else. I do think he talks to some people who at least have seen these kids and formed opinions. But - history shows they are as wrong as much as they are right about this stuff.
As for Mike Wang, it's not hard to understand why Kentucky would not be interested. That doesn't mean he isn't a 3*** with a real chance to play at a high level in our league. 6'9 / 6'10 kids who can put the ball on the floor and shoot from range are not all that common in our league. Take Luke Petrsek for example. There's an All Ivy kid who had a real impact in our league, yet entered as a yawner coming out of Suffolk County. And he's counterbalanced by guys like Bryce Aiken who enter with big reps and superior skillsi and it Is obvious from the get go. But some of those guys stumble and there just are not that many choosing our mid major level league. Even Aiken hasn't shown he can win championships although he'll clearly have that opportunity.
|
