• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

BBall Recruiting Megathread 3.0 - The Manning era begins: The Harry Giles Chronicles

If you know anything about basketball and standard deviation, there's no way the standard deviation is around 6.
 
ChrisL68 said:
It's not a statistical study, but the more data points you have showing a different average PPG among the star ratings, the more likely it is to be valid. The data group, which is what this study is approximating, sets the standard deviation for analysis of a data point.

Taking the averages of 19,000 data points probably yields a pretty good approximation of what the standard deviation is. i.e., the differences are probably real.
Click to expand...

Kudos to the author for compiling a large number of data points and generally that yields a more accurate picture. It does not necessarily mean the differences are statistically significant unless you have information about the spread of the data. Large studies with apparent differences can be statistically insignificant if the spread of data is large. Those 19000 data points are spread over 24 categories and could range from zero to around 22 for ppg. You also have the problem that those data points are not independently variable in that they represent the same player over time. I'm not saying that they aren't real differences, just that we don't know for sure. If they aren't different then the whole arguement is moot.
 
PhDeac said:
If you know anything about basketball and standard deviation, there's no way the standard deviation is around 6.
Click to expand...

Doesn't have to be 6. Even an SD of 3 might make the differences nonsignificant (only one standard deviation difference between the means). Point is that the analysis is incomplete and renders subsequent arguments suspect.
 
PPG is not a good measure of success at all. Using that metric, Travis McKie had a better first two years at Wake than AFA. And that doesn't address the problem of the competition that highly ranked players will face versus those who can only get mid-major scholarships. But there have been much better studies that confirm the obvious, there is a strong correlation between ranking and production. No one can seriously argue otherwise, right?
 
PPG just started his discussion. Offensive rating was a good case as well.
 
Fuck Your Couch said:
PPG is not a good measure of success at all. Using that metric, Travis McKie had a better first two years at Wake than AFA. And that doesn't address the problem of the competition that highly ranked players will face versus those who can only get mid-major scholarships. But there have been much better studies that confirm the obvious, there is a strong correlation between ranking and production. No one can seriously argue otherwise, right?
Click to expand...

You would be utterly shocked at what people can argue about. To me there is just absolutely no way a person could try to claim less talented players have an equal impact on a team's productivity as the more talented star players.
 
Lower rated doesn't necessarily mean less talented when you get past the Top 20-25 players. I've asked many scouts, raters and coaches if they could describe what the difference between #30 and and #75, many laughed. Some said personal taste. After the top group, it's really a crapshoot.
 
Well, that is probably why only the top 25 or so players get 5 stars. #30 and #75 usually have the same number of stars. And lets not forgot there are a lot more players lower than #75. rj you would admit there is usually a big difference between #30 and an unranked 3 star, right? Or usually even #75 and an unranked 3 star.
 
Fuck Your Couch said:
Well, that is probably why only the top 25 or so players get 5 stars. #30 and #75 usually have the same number of stars. And lets not forgot there are a lot more players lower than #75. rj you would admit there is usually a big difference between #30 and an unranked 3 star, right? Or usually even #75 and an unranked 3 star.
Click to expand...

The study says that the difference between a 3 star and a 4 star is 1.6ppg as a senior. Is that a big difference? Still think this is a good study?
 
1.6 PPG isn't a huge deal, but if you have three 3-stars instead of three 4-stars that is 5 ppg there. That can be the difference in several games over the course of a year could it not be?

The study is not exactly a scientific gym, but it reinforces what most of us already know: higher rated players on average are better than lower rated players on average. That clearly makes sense.

There are aberrations, but most of the time a 4-star will perform better than a 3-star.

It's also more about offensive rating and usage rates than just PPG, which I believe the article touches on as well.
 
Fuck Your Couch said:
Well, that is probably why only the top 25 or so players get 5 stars. #30 and #75 usually have the same number of stars. And lets not forgot there are a lot more players lower than #75. rj you would admit there is usually a big difference between #30 and an unranked 3 star, right? Or usually even #75 and an unranked 3 star.
Click to expand...

It depends on what 3*. Is the guy just out of the Top 100? Is it a guy who didn't play much AAU?

Look at Devin. Lots of teams have guys like him.
 
RJKarl said:
It depends on what 3*. Is the guy just out of the Top 100? Is it a guy who didn't play much AAU?

Look at Devin. Lots of teams have guys like him.
Click to expand...

Of course there are individual, anecdotal cases that point to everything. I'll tell you I'm confident on average #75 does better than unranked 3-star. Arguing otherwise seems to just be an effort to be contrarian.
 
Top PPG scorers in college basketball last year:

Points Per Game
25.031 Erick Green (VT)
23.167 Doug McDermott (CREI)
22.618 Lamont Jones (IONA)
22.273 Nate Wolters (SDSU)
22.121 Travis Bader (OAK)
21.806 Isaiah Canaan (MURRAY)
21.667 Tyler Haws (BYU)

Look at all those likely 2 and 3 star players. Obviously those were the best players in college basketball. 3 stars > 5 stars.
 
I think Joachim Noah was about #75. :)

Let's look at some #75s

Melvin Johnson, VCU 6.9 ppg, 1.5 rpg, 1/0 apg
Mike Shaw, IL 0.9/0.8/0.1
Tariq Black, Memphis 8.1/4.8/0.4
Roger Franklin, OKST went form 1.7 ppg at OK State to 11.7 at N TX State
Steve Tchiengang, Vandy 3.4/3.3/0.5
Scoop Jardine, Cuse about 10/2/5 for three years

There's no question you can find a really good player at #75. But it is hit and miss.

Obviously #75 is likely to be better than #140 or #150, but the difference between him and #105 or #110 may be slight.
 
Sorry to have derailed this thread. I just wanted to stop people from blindly accepting the conclusions of an article just because it has numbers in it. I'm sure there is a correlation, on the whole, between recruiting stars and productivity, although I suspect it is not as strong as many believe. Consider the source of the article. His livelihood depends on recruiting stars being relevant. Anyway, I think that's enough of a diversion. Let's talk WFU basketball recruiting...
 
I was actually surprised the stats weren't more skewed. Not to mention 5 stars and high 4's likely won't be around for their senior year, so if you establish a program that picks up high 3's and mid-tier 4's but keeps them as seniors, the interesting comparisons are really between different class years. A 2 star junior, on average, is better in every measurement than any freshman other than a 5 star.

If anything that "study" supports just how dangerous it is to rely on freshmen, and how drastic the leap is from freshman to sophomore year. And that's just the averages - if you really believe you have recruiters that are better at recognizing under-the-radar talent, that makes it look like a no-brainer to take upper 3/lower 4 stars and ensure stability in your junior/senior players.

My biggest problem with all those averages is that basketball isn't football. You play 5 guys and one player can make you a national championship contender. To be a Final Four quality team, it can be argued what you need is an outlier - and I'd be willing to bet that outlier players like CP3 are massively weighted into the 5 star and high 4 star pools.
 
For the morons:

Scout's rankings are still available through 2002. Since that time, here are our recruits, broken down by rating.

5-star
Chris Paul
AFA
Ty Walker

4-star
Justin Gray
Richard Joyce
Trent Strickland
Eric Willilams
Kevin Swinton
Ish Smith
Jamie Skeen
Anthony Gurley
Jeff Teague
James Johnson
Tony Woods
Ari Stewart
Travis McKie
Carson Desrosiers
JT Terrell
Melvin Tabb

3-star
Chris Ellis
Kyle Visser
Jeremy Ingram
Todd Hendley
Cam Stanley
David Weaver
Harvey Hale
LD Williams
Casey Crawford
Chas McFarland
Gary Clark
CJ Harris
Tony Chennault
Chase Fischer

2-star
Joe Dukes
Konner Tucker
Anthony Fields
Daniel Green


Among 5-star players, 2/3 made All-ACC (67%) with three total selections, 1/3 made first-team (33%), and 2/3 were draft picks (67%). One out of 3 was a bust (33%).

Among 4-star players, 6/16 made All-ACC (37.5%) with nine total selections, 1/16 made first-team (6%), and 2/16 were draft picks (12.5%). Jamie Skeen also made the All-CAA second team and Anthony Gurley made the All-A-10 third team in 2011, which would bump up to 50% with all-conference honors. To date, 2/16 are indisputable busts (12.5%), two (Stewart and Tabb) look likely (12.5%), and others are up in the air.

Among 3-star players, 2/14 made All-ACC (14.3%) with two total selections, 0/14 made first-team (0%), and 0/14 were draft picks (0%). Incidentally, both Harris and Visser, the two All-ACC honorees, were named to the third team following their senior year. To date, 4/14 are indisputable busts (28.6%), I'd give the same label to Ellis and Weaver (14.3%), and Chennault and Fischer appear likely (14.3%).

Among 2-star players, 0/4 made All-ACC (0%) with 0 total selections, 0/4 made first-team (0%) and 0/4 were or will be draft picks (0%). To this point, 3/4 are busts (75%).


ETA: Just looked back at the stats and would probably remove Ellis from the bust discussion (averaged 5.6 and 4.3 as a senior, played regularly except when hurt as a sophomore). Weaver, with a senior year line of 2.6 ppg and 2.1 rpg, qualifies, IMO.
 
Last edited:
You must log in or register to reply here.
Back
Top