Online Now 670

Wolverine247 Board

The place for discussion on the Michigan Wolverines

On this Board 412
Record: 2671 (1/26/2013)

Online now 595
Record: 7264 (3/12/2012)

Reply

We're #2! We're #2!

  • shrug

    2013 Football Recruiting Team Rankings

    http://247sports.com/Team/Ranking

    247sports.com

    buttesnake

  • I dont understand 247 calculations and methodology....you would figure take you average player ranking and multiply by total commits, and that equals your point total, but I guess not.

    signature image signature image signature image

    Awink2

  • We'd still be number one if offering a Long Snapper didn't tank our average numerical ranking by more than a whole point. For example, a 24 commit class with an unranked LS is weaker than a 23 commit class featuring the same players minus the LS. Does not make any sense, only way to keep it fair would be to not factor in ST players that are rarely ranked and, if so, are not higher than two stars. Taking the LS could very well put us out of a Top-5 finish just because of broken methodology. Basically, not taking a contributor is better for your ranking than taking an unranked or two-star recruit that at least has an opportunity to contribute in some way. In actuality, even a very minimal contribution is better than no contribution.

    UMFB

  • UMFB said...

    We'd still be number one if offering a Long Snapper didn't tank our average numerical ranking by more than a whole point. For example, a 24 commit class with an unranked LS is weaker than a 23 commit class featuring the same players minus the LS. Does not make any sense, only way to keep it fair would be to not factor in ST players that are rarely ranked and, if so, are not higher than two stars. Taking the LS could very well put us out of a Top-5 finish just because of broken methodology. Basically, not taking a contributor is better for your ranking than taking an unranked or two-star recruit that at least has an opportunity to contribute in some way. In actuality, even a very minimal contribution is better than no contribution.

    The average numerical ranking isn't part of the calculation for their team rankings.

    buttesnake

  • UMFB said...

    In actuality, even a very minimal contribution is better than no contribution.

    Correct. Our point total did increase after the LS commit.

    buttesnake

  • Awink2 said...

    I dont understand 247 calculations and methodology....you would figure take you average player ranking and multiply by total commits, and that equals your point total, but I guess not.

    247 explains it thusly:


    Explanation

    Each recruit is weighted in the rankings according to a Gaussian distribution formula (a bell curve), where a team's best recruit is worth the most points. You can think of a team's point score as being the sum of ratings of all the team's commits where the best recruit is worth 100% of his rating value, the second best recruit is worth nearly 100% of his rating value, down to the last recruit who is worth a small fraction of his rating value. This formula ensures that all commits contribute at least some value to the team's score without heavily rewarding teams that have several more commitments than others.

    Readers familiar with the Gaussian distribution formula will note that we used the value 6 for σ because this was the standard deviation for total number of commits between schools as they were ranked during the 2011 recruitment year, the year this formula was developed. This standard deviation creates a bell curve with an inflection point near the average number of players recruited per team.

    Below is a graphical representation of how our formula works. You can see that the area under the curve gets smaller both as the rating for a commit decreases and as the number of total commits for a school increases. The y-axis in this graph represents the percentage weight of the score that gets applied to an overall team ranking.

    And I attempt to describe it in layman's terms in the linked thread.

    What will Smith bring to Michigan?

    What will DeVeon Smith bring to the Wolverines? Click inside for a detailed analysis of Michigans latest verbal pledge from Wolverine247.

    michigan.247sports.com
    attachment

    buttesnake

  • buttesnake said...

    The average numerical ranking isn't part of the calculation for their team rankings.

    Thought the average factored back into some sort of formula to determine the the overall points. That was just an assumption to make sense of it. Only other thing I can think of is that there must be a cap at 15 (Alabama just hit that number) and every recruit after a team's best 15 do little to affect the rankings. They have a couple five-stars and a few highly ranked four-stars. Their best 15 are ranked higher than ours. That would explain it, can't think of anything else.

    UMFB

  • UMFB said...

    Thought the average factored back into some sort of formula to determine the the overall points. That was just an assumption to make sense of it. Only other thing I can think of is that there must be a cap at 15 (Alabama just hit that number) and every recruit after a team's best 15 do little to affect the rankings. They have a couple five-stars and a few highly ranked four-stars. Their best 15 are ranked higher than ours. That would explain it, can't think of anything else.

    When the team rankings were released last year, I believe JC mentioned that 17 was the magic number, after which any subsequent commit is weighed significantly less. Don't quote me on that number though, they may have adjusted it since.

    buttesnake

  • UMFB said...

    Thought the average factored back into some sort of formula to determine the the overall points. That was just an assumption to make sense of it. Only other thing I can think of is that there must be a cap at 15 (Alabama just hit that number) and every recruit after a team's best 15 do little to affect the rankings. They have a couple five-stars and a few highly ranked four-stars. Their best 15 are ranked higher than ours. That would explain it, can't think of anything else.

    There really isn't a cap. Each committ will add to your overall team point total but the more recruits you get the less it adds to your team score. With each commit the percentage of their overall rating that counts toward a team point total goes down.

    signature image signature image signature image

    WillyWolverine

  • Basically, don't expect Michigan's point total to increase significantly unless Michigan gets Kendall Fuller or Leon McQuay. Those are the targeted recruits that can noticeably increase Michigan's point total because of where they are ranked (Fuller at 9 and McQuay at 38) compared to our committed recruits.

    But if Michigan adds Joe Mathis and/or Josh Augusta, Michigan's point total will increase incrementally.

    This post was edited by Frank C on 6/8/2012 at 3:05 AM

    signature image

    Frank C

  • buttesnake said...

    When the team rankings were released last year, I believe JC mentioned that 17 was the magic number, after which any subsequent commit is weighed significantly less. Don't quote me on that number though, they may have adjusted it since.

    I sort of figured it out once. It's actually weighed down quite a bit per recruit. If you look at Michigan team point total we are only averaging about 36 points per recruit. That has to mean that by commit 10 each commit is weighed down pretty heavily. You can't be getting anywhere near 50% of a commits overall point ranking at that point.

    signature image signature image signature image

    WillyWolverine

  • Frank C said...

    Basically, don't expect Michigan's point total to increase significantly unless Michigan gets Kendall Fuller or Leon McQuay. Those are the targeted recruits that can noticeably increase Michigan's point total because of where they are ranked (Fuller at 9 and McQuay at 38) compared to our committed recruits.

    But if Michigan adds Joe Mathis and/or Josh Augusta, Michigan's point total will increase incrementally.

    Another way our point total could go up is if a decent number, say 6 or more, of our recruits received bumps in their ratings.

    signature image signature image signature image

    WillyWolverine

  • WillyWolverine said...

    I sort of figured it out once. It's actually weighed down quite a bit per recruit. If you look at Michigan team point total we are only averaging about 36 points per recruit. That has to mean that by commit 10 each commit is weighed down pretty heavily. You can't be getting anywhere near 50% of a commits overall point ranking at that point.

    Yeah you can see that by checking out teams with about that number of recruits and comparing it to other teams that are clustered right around them.

    buttesnake

  • That makes sense to give every team an equal chance at receiving a highly ranked class. In a given year, some teams can take 28 and others may fill at 20. Though, 28 players will likely do more for a team on a football field than 20. It can be argued either way, no formula can marry quantity and quality in a way that is both fair to all programs and accurate to how much will be contributed on the field. The overall rankings seem futile until every program hits the magic number of weighted recruits.

    UMFB

  • So what would/will the effect be when they finally get around to giving Shane his fifth star? Big? Small? Nothing at all?

    inuyesta

  • Alabama picked up another recruit Maurice Smith a highly rated CB from Texas.

    HelloHeisman 21

  • inuyesta said...

    So what would/will the effect be when they finally get around to giving Shane his fifth star? Big? Small? Nothing at all?

    There will be additional points awarded but it's hard to quantify it without doing the math, and I don't feel like doing that. I suspect that the bump to 5* status will accompany a nonlinear increase in points as compared to a situation where the recruit's rating is increased but the star status isn't.

    buttesnake

  • buttesnake said...

    There will be additional points awarded but it's hard to quantify it without doing the math, and I don't feel like doing that. I suspect that the bump to 5* status will accompany a nonlinear increase in points as compared to a situation where the recruit's rating is increased but the star status isn't.

    I have a post I started this morning, but I was running late for work. I will finish it up when I get a chance here in a little bit. It covers recent thoughts on recruiting ratings and rankings

    xxmgobluexx

  • xxmgobluexx said...

    I have a post I started this morning, but I was running late for work. I will finish it up when I get a chance here in a little bit. It covers recent thoughts on recruiting ratings and rankings

    Cool, looking forward to it.

    buttesnake

  • All you har to do to figure out that the team rankings are broken is look at USC and Oklahoma. They each have 5 four stars, Oklahoma also has 3 three stars whereas USC also has 3 five stars. Somehow Oklahoma is ahead of USC.

    CVille Blue75837

  • It really doesn't matter to me how, or why, 247 does their rankings the way they do.

    What matters is what Brady and his guys think and do in evaluating recruits. They know more football than any 'expert' for any of the recruiting websites. That's why they're employed as D-1 football coaches at a legendary program like Michigan, while the other guys post website stories for a living.

    'We are attacking that like animals right now.'

    AAStagg

  • buttesnake said...

    There will be additional points awarded but it's hard to quantify it without doing the math, and I don't feel like doing that. I suspect that the bump to 5* status will accompany a nonlinear increase in points as compared to a situation where the recruit's rating is increased but the star status isn't.

    If Morris receives a bump, Michigan's overall points will get the same bump. The reason being is that your top recruit's (i.e. Morris) rating is multiplied by a factor of one. So if Morris goes from 97 to 98, Michigan's total points will increase by 1.

    signature image

    psubills62

  • C-Ville Blue said...

    All you har to do to figure out that the team rankings are broken is look at USC and Oklahoma. They each have 5 four stars, Oklahoma also has 3 three stars whereas USC also has 3 five stars. Somehow Oklahoma is ahead of USC.

    Not broken, it's simply a side effect of classes not being completed yet.

    When looking at teams with fewer commits ( < 10), the total number of commits become that much more important. The team points system essentially ranks the commits according to rating, highest to lowest. Then they multiply each rating by a factor. The factor decreases as you move down the list. If a team doesn't have many commits, then they're still using large factors, and thus every commit comes with a relatively large point total.

    With where Michigan is at, you guys already have 21 slots filled, which means any further commits will simply be a relatively small increase, even if they're highly rated.

    Harder to explain without graphs, but hopefully that covers it. Basically, with fewer commits, it's nearly impossible to overtake a team with more commits - the fact that USC is only 3 points away from Oklahoma (who has a very good class in their own right) just goes to show how great USC's class is. Things will balance out once classes are finished.

    signature image

    psubills62

  • psubills62 said...

    Not broken, it's simply a side effect of classes not being completed yet.

    When looking at teams with fewer commits ( < 10), the total number of commits become that much more important. The team points system essentially ranks the commits according to rating, highest to lowest. Then they multiply each rating by a factor. The factor decreases as you move down the list. If a team doesn't have many commits, then they're still using large factors, and thus every commit comes with a relatively large point total.

    With where Michigan is at, you guys already have 21 slots filled, which means any further commits will simply be a relatively small increase, even if they're highly rated.

    Harder to explain without graphs, but hopefully that covers it. Basically, with fewer commits, it's nearly impossible to overtake a team with more commits - the fact that USC is only 3 points away from Oklahoma (who has a very good class in their own right) just goes to show how great USC's class is. Things will balance out once classes are finished.

    Gotcha, didn't realize OU had an extra sub 3 star commit.

    Though I feel like when everything evens out this system will punish Michigan compared to top heavy classes. We might have a bunch of four stars, but since your top recruits are weighted kind of heavily a hypothetical class with a few five stars and more three stars would pass us. That's a legitimate valuation of high end, top 25 players by 247 though

    CVille Blue75837

  • I think we all knew that Bama was going to pass us. They have a legitimate shot at like 4-6 5* recruits. Quite frankly, I don't think anyone can keep up with them in recruiting right now outside of maybe USC or Texas.

    We are still going to have an awesome class even if it ended right now.

    jimtaco