mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-14-17 10:33 AM - Post#233968
In response to SomeGuy
Right now, here's how I'd rank the incoming 2018ers:
1) Llewellyn
(gap)
2) Kirkwood
3) Freedman
4) Samuels
5) Forbes
6-8) Wright/Cotton/Ellis
Next tier) Kelly, Lanford, Mitchell, Slajchert, Wang, Catchings, Friberg
Obviously senior seasons will tell more as well... that (and the PG year) is when Oni went from a fringe D1 player to a top ROY candidate in a loaded class.
|
TheLine
Professor
Posts: 5597
Age: 60
Reg: 07-07-09
|
Mike Wang to Penn 10-14-17 10:41 AM - Post#233969
In response to AsiaSunset
Luke is an interesting case. He wasn't highly recruited because he had a late growth spurt and was playing in a close to off-the-radar conference. Recruiting services do poorly with that sort of player. If you happened to watch Petrasek play his senior year - and it was easy to do that since several of his games were on SNY - it was clear how talented he was.
Nevertheless Luke's first 2 years at Columbia were non-descript. His Win Shares for the 4 years were 1.3/0.7/3.4/3.0. That's below Mike's arbitrary standard for what he's considering a full-blown Ivy success.
Michael Wang plays on a high profile team in a high profile conference - much more of a known quantity. I trust more that the recruiting services pegged him correctly. The one caveat is that they do tend to grade taller players higher.
I have no idea if that makes Michael Wang a top 5 recruit in this class. Not that it matters. Let's hope he's not the next Harry Schaen.
I'm in full agreement with UPIA1968's take. And I'd be happy if Michael Wang is as successful as Luke Petrasek, even understanding that would mean enduring 2 years of growing pains.
|
SomeGuy
Professor
Posts: 6391
Reg: 11-22-04
|
10-14-17 10:44 AM - Post#233970
In response to mrjames
thanks. I'm a bit surprised to see Samuels so high -- the public offers I've seen don't really support the high rating. Where in your matrix is Samuels getting out in front of Wang/Kelly/Cotton/Elliis? Seems like those are similarly rated guys with better offers. Is it that I'm missing offers, or is something else going into the equation for him?
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-14-17 10:57 AM - Post#233971
In response to TheLine
I just picked 10 because it's a nice round number. Happy to run the odds at any level of win shares. In fact, one of the models I run at the player level is focused around the odds a player can get to 5 win shares.
10 career win shares is an absolute star, but the relationship should hold at all levels of win shares (though the relative gaps in odds may change). Throw out a number and I can give you the odds at that level of win shares.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-14-17 11:05 AM - Post#233972
In response to SomeGuy
Samuels is an interesting case. I think people were surprised he pulled the trigger so early and went for Dartmouth. My perception from that is that more offers might have followed the longer he stayed on the board. He might get a bit overlooked because the one thing the 2018 class was useful for was quality, high-academic PGs and combos, but that pickup gets a lot of high praise. Like on a different level than the kudos over the Chris Knight pickup last year.
|
Streamers
Professor
Posts: 8141
Loc: NW Philadelphia
Reg: 11-21-04
|
10-14-17 11:43 AM - Post#233973
In response to mrjames
This is always an interesting time of the year on the board. Before we see our teams on the court, but with most of the recruiting results in, so we obsess over every little piece of news we get. It's fun and I'm not knocking it, but ranking the recruits is a complete exercise in futility for all the reasons that have pointed out that make such predictions in our league difficult. Best we can do is see is what each team gets in the context of their needs. This is not the NFL draft where you hunt for best athletes. They don't come to us.
Penn fans - that list of 3*s is telling. We rejoiced when we got Pettinella. He had that kind of talent, but never fit in. Reilly was a huge talent, but the injuries and the coaching did him in. Danley worked out beautifully. Then you have all the hidden gems: Begley (made for the Ivy game), Schiffner (football recruit), Zoller (Fran knew what he had if nobody else seemed to), Ibby (insiders knew, ESPN did not.) to name a few.
The comments about the 2018's senior year are important. I suspect we will all revise the projections after we see that. As for Wang, he clearly appears to be a good fit with the potential to get a lot better on paper. We do not need a big class. I'm happy for now, although a PG would have been nice.
|
TheLine
Professor
Posts: 5597
Age: 60
Reg: 07-07-09
|
10-14-17 12:30 PM - Post#233974
In response to mrjames
Yes understood - and 10 is a reasonable number to pick. But also wanted to point out to everyone that Luke Petrasek is an example of a success - he was well above average for 2 years - and didn't reach that number. Some might've given up on him after his first 2 years.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-14-17 12:51 PM - Post#233976
In response to TheLine
Yeah, you'd want to use Win Shares per 40 to identify potential breakout candidates that have only received rotation minutes.
As for the rest of this, I continue to be amused by those that confidently assert that we can't tell anything about recruiting despite the only person on these boards that has done the work to test that hypothesis telling you that you can, actually, predict the performance of recruiting classes. Is what it is.
|
AsiaSunset
Postdoc
Posts: 4350
Reg: 11-21-04
|
10-14-17 03:14 PM - Post#233978
In response to mrjames
Mike
Most people believe that you can tell a lot from recruiting and that there is some correlation between rankings and success; however, that certainty increases significantly when you get to the high major area. In our league, it's a bit more complex.
Others, like myself, don't believe you have any special expertise that makes your observations anymore reliable than others in terms of ranking recruits. And - you actually have done nothing to disspell that thinking. This, of course, is the exact opposite opinion that almost everyone has with respect to the quality analytics you provide on this site.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-14-17 03:51 PM - Post#233979
In response to AsiaSunset
You provide quality analytics.
You have an analytical model that demonstrates a statistically significant relationship between recruiting rankings and future output.
You have done nothing to dispel the thinking that you have no special expertise that makes your observations any more reliable than others.
That's why I chuckle.
|
Penndemonium
PhD Student
Posts: 1877
Reg: 11-29-04
|
Mike Wang to Penn 10-14-17 04:37 PM - Post#233980
In response to mrjames
I'm not sure I like this. Two posters who are among my favorites (mrjames and AsiaSunset) going through the age-old arguments about predictive analytics. I don't get why anyone is upset here. Mrjames is an unabashed believer in analytics and agrees that his models are predictive but leave vast room for over and under achievers. Asia is a believer that the models don't capture the human realities of Ivy basketball, but recognizes that statistics do tell a story. I'd love for you both to contribute your own style insights and to encourage each other to keep supplying the opposite type. I disagree strongly with either of them being labeled as trolls, apologists, and such.
I share worries about what this recruiting class signals, as it appears we whiffed on many of our top targets. The problem is that our program is at an inflection point and a single bad recruiting class seems like a big deal. If we experience institutionalized success for a few years, we wouldn't worry so much about this class.
I also think there is some inherent stereotyping going on here. We have a recruit whose videos are pretty good, who has a certain skillset, and has played in top quality competition. He is rated highly by some services, but not covered by all. I agree with Asia that many are being way too dismissive of him - clearly our coaches see something in him. I also agree with mrjames that he probably isn't the top 5 recruits in the league. Everyone is worried about a first signing of a recruit who appears to be a solid three star recruit on ESPN. Normally that would delight Penn fans. Hopefully Mike Wang has enough mettle to silence the critics like Jeremy Lin did!
Edited by Penndemonium on 10-14-17 05:06 PM. Reason for edit: No reason given.
|
TheLine
Professor
Posts: 5597
Age: 60
Reg: 07-07-09
|
10-14-17 04:43 PM - Post#233982
In response to mrjames
Mike provided his data and it shows a real strong correlation. It's not like he's making anything up out of thin air.
Nothing is absolute. Over half of the "Consensus 3 Stars" didn't accumulate 10 win shares. Some in the lower rated categories did.
We've been told year after year we're behind H and now also YP in recruiting. We lose head to head recruiting battles. We lose in the standings. There's a correlation.
Am I missing something?
|
AsiaSunset
Postdoc
Posts: 4350
Reg: 11-21-04
|
Re: Mike Wang to Penn 10-15-17 08:35 AM - Post#233988
In response to Penndemonium
Don't worry. I respect and appreciate Mike, but I don't need to agree with him about certain things. I certainty think his definition of a 3*** is both imperfect and for the most part non sensical, but I do believe in a correlation between ratings and performance (although I strongly disagree that what he has posted does much to support a point which should be pretty obvious to everyone).
Not sure why he's chuckling since his analytical work is clearly outstanding, but that work in no way demonstrates he has the ability to pick a list of top ten recruits coming into the league. That's an entirely different exercise. My comments should be clear to everyone and haven't been inconsistent.
I do agree with his first 3 picks. i watched a little tape on the Princeton pg. He is clearly an elite athlete with a terrific handle and a potent step back jumper. I'd be surprised if he doesn't have a career consistent with what I saw on tape and I doubt we'll need to measure win shares to document that fact - that is, if he enjoys good health and plays out his 4 years at Princeton. As we all clearly realize, relative to our own program, stuff happens that sometimes gets in the way independent of a kid's rating entering college.
|
SomeGuy
Professor
Posts: 6391
Reg: 11-22-04
|
10-15-17 11:18 AM - Post#233995
In response to AsiaSunset
In regard to this anecdotal vs. factual debate, it seems to me that the instances where highly rated guys bust or obscure guys hit are something Mike's thoughts on recruiting take into account. Whatever you call these instances (facts or anecdotes), it seems like the basic criticism of Mike is that he isn't right all the time. However, he isn't claiming to be.
|
AsiaSunset
Postdoc
Posts: 4350
Reg: 11-21-04
|
10-15-17 11:35 AM - Post#233996
In response to SomeGuy
Correct - but my point is Mike's thoughts are no more insightful than others, go way beyond his area of expertise and don't have much to do with the analytics posted on this site.
I don't pretend to be an expert but I think projecting the top recruits in the class is something any of us might do, although our lists might disagree on certain individuals. And all of us will have hits and misses because stuff happens.
Importantly - it differs from his analytical work in that it includes subjective thinking and is not void of emotion. Did you watch Mason Forbes video? My reaction - might not get PT time at Harvard based on the young players already there and depending on who follows behind him. We'll see, but I don't see him on any list I might put together.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
10-15-17 12:10 PM - Post#233997
In response to AsiaSunset
Correct - but my point is Mike's thoughts are no more insightful than others
I definitely wouldn't take my opinions on recruits seriously at all. Why is that relevant to this conversation though?
|
Silver Maple
Postdoc
Posts: 3765
Loc: Westfield, New Jersey
Reg: 11-23-04
|
10-15-17 12:16 PM - Post#233998
In response to mrjames
Correct - but my point is Mike's thoughts are no more insightful than others
I definitely wouldn't take my opinions on recruits seriously at all. Why is that relevant to this conversation though?
Honestly, I've lost track of what this conversation is even about. However, I'll just lob in the observation that, while I'm not really a Pink fan, 'What About Us' is a pretty good song.
|
Streamers
Professor
Posts: 8141
Loc: NW Philadelphia
Reg: 11-21-04
|
10-15-17 12:16 PM - Post#233999
In response to AsiaSunset
Seems to me we are over-analyzing this. Mike has shown a correlation, but not a strong one, and that is exactly what I would have expected at this level.
All other things being equal, i'd rather see us get:
- More 3* then 2*, why? because the league has gotten to the point where you need at least a couple of 3* athletes on your roster to compete for a title
- Players who fill needs and will like get into the rotation by the time they are juniors
- Players who are a system fit
- Players with real potential beyond what was displayed in HS (aka easy to coach and room to grow)
- Players with several offers, especially if PU, Yale or Harvard are among them
|
Silver Maple
Postdoc
Posts: 3765
Loc: Westfield, New Jersey
Reg: 11-23-04
|
10-15-17 12:23 PM - Post#234000
In response to Streamers
Players who are ideal for a team's particular system are especially valuable. A 2* guy who's perfect for your style could easily look like a 4* recruit. Often, when we refer to 'under the radar' recruits, I think this is what we're actually talking about-- guys who might ride the bench at another school, but who are stars at yours. Princeton has had a number of these in recent years. Might Ryan Betley also fit into this category?
I'll also add that this is a factor that probably can't be expressed quantitatively.
|
JDP
Masters Student
Posts: 559
Reg: 11-23-04
|
10-15-17 01:37 PM - Post#234004
In response to mrjames
Mike, how adaptable is your model to the women's game? I assume you would have to preform some recalibration (the recruit rankings may not be as robust) relative to the men's model? Does the data exist to undertake the task? does the will exist?
|