Nonsensical team rankings

[redeye]

I just want to point this out for everyone to see, while it's available.

Oregon gained a commit this weekend and it allowed them to pass us in the 247 composite rankings.

http://247sports.com/Season/2015-Football/CompositeTeamRankings

Now, note how Oregon has the same number of commits (14), fewer 4+ star players (5 to our 6) and a lower average player rating (88.17 to our 88.35) as of the time this was written.  And yet, Oregon has 200.11 points to our 199.07, which means they're ranked above us, apparently for reasons other then "star ratings".  It may be that 247 believes that Oregon has filled it's needs better, but I don't know (does anyone?)

I realize many of you are familiar with these things and I've noticed it many times this year, myself.  I'm just putting it here for those who may be unfamiliar with the curious way that rankings work.

[greenie]

I think you have to look at the numerical rating of each player. I'm just guessing, but the 4*'s they have committed probably have a higher numeric rating resulting in more points.  One of the gurus on here will surely be able to explain.

[redeye]

I think you have to look at the numerical rating of each player. I'm just guessing, but the 4*'s they have committed probably have a higher numeric rating resulting in more points.  One of the gurus on here will surely be able to explain.

Others have noted before that the best players are worth more, while the worst may not count at all.  In this case, I believe they all count some, but their higher rated 4's likely give them the edge, so you're right, but our average player rating is still higher (i.e. our average player is considered better.)  Personally, I prefer a higher average player rating, because that means we have better players across the board, although I also think a few superstars are a necessary component to a championship team.

Anyway, I just thought it was interesting and wanted to point it out.

update: If you remove the highest rated player from each class, Arkansas would jump back in the league.  I used the class calculator to do that and it resulted in us gaining a 2+ point lead on Oregon.

If you remove the lowest rated player from each class, Oregon retains the lead by 00.05 points and their lowest rated player was also ranked higher.

[31to6]

Others have noted before that the best players are worth more, while the worst may not count at all.  In this case, I believe they all count some, but their higher rated 4's likely give them the edge, so you're right, but our average player rating is still higher (i.e. our average player is considered better.)

Anyway, I just thought it was interesting and wanted to point it out.

update: If you remove the highest rated player from each class, Arkansas would jump back in the league.  I used the class calculator to do that and it resulted in us gaining a 2+ point lead on Oregon.

If you remove the lowest rated player from each class, Oregon retains the lead by 00.05 points and their lowest rated player was also ranked higher.

Click through and look at their individual ratings. Their 4*'s are .98, .96, .94 and .92. Ours are .95, .92, .92, .92, .91 & .89.

Then look at the bottom few 3*s. Their "worst" 3*'s are .82s whereas ours are all .85 or better (ignoring the punter).

Then go into the class ranking calculator for both and look at the points they give for the top players in each. Frojholt and Hill contribute 25.32 and 21.72 to our overall score, respectively. For Oregon, Griffin contributes 28.02 and 25.81.  So their 5 4*'s are actually worth more points than our 6.

All of these ranking systems give far more credit for star power than for overall depth, which weights them toward the top ranked players in the class.

That is why Oregon's class has more "points" and is therefore ranked higher even though our 3*'s are higher quality than theirs (which is why our average star rating is higher).

[redeye]

Click through and look at their individual ratings. Their 4*'s are .98, .96, .94 and .92. Ours are .95, .92, .92, .92, .91 & .89.

Then look at the bottom few 3*s. Their "worst" 3*'s are .82s whereas ours are all .85 or better (ignoring the punter).

Then go into the class ranking calculator for both and look at the points they give for the top players in each. Frojholt and Hill contribute 25.32 and 21.72 to our overall score, respectively. For Oregon, Griffin contributes 28.02 and 25.81.  So their 5 4*'s are actually worth more points than our 6.

All of these ranking systems give far more credit for star power than for overall depth, which weights them toward the top ranked players in the class.

That is why Oregon's class has more "points" and is therefore ranked higher even though our 3*'s are higher quality than theirs (which is why our average star rating is higher).

Yea, I'd forgotten that at first, but you're right.  They didn't always do it this way and it was simply the more players, the better score.  I like how they do it now better, although I'd still rather have a higher average star rating, then a higher overall score.

Think of it this way; some of those high 4's will never materialize for Oregon, and then you have injuries, and whatnot.  When Oregon's class is re-ranked down the road, that's going to drastically change their ranking more then ours and, most likely, their level of success on the field.  I'm not saying that I wouldn't love to have those players at Arkansas, but just that I'd rather have a higher average star rating, even though the services don't value it as much.