VolFreakJosh
“Don’t you put that evil on me Ricky Bobby!”
- Joined
- Feb 2, 2013
- Messages
- 28,127
- Likes
- 71,539
When was that written? There’s no date/time stamp on it or even an author (that I could see at least).
-VQWe believe Chandavian Bradley will be in town at the end of the month which is something that will go a long way to helping to wrap things up for the Vols in our opinion.
He's a composite 5* on 247. It doesn't make sense because I don't think he's a 5* on any particular service, but he is listed as one in the composite.
It changes but the cut-off is usually around 32. Some years it might be 30 some it might be 34.I may be wrong, but I thought the top 35 or so was the cut-off for 5 star status on 247 by the end of the year. Bradley is very close to that line at 37.
It’s similar to the Seldon situation. At the beginning they have less 5*s than they do at the end of the year. They always end up with 32 (I think that’s the number, been awhile since I looked but close to that) 5*.
So what’s happening is the composite rating is taking sites like Rivals that have him in the top 32 and counting that as a 5* rating
It changes but the cut-off is usually around 32. Some years it might be 30 some it might be 34.
It's usually around 32 because there's 32 first round picks and they are "projecting" five stars to be first round picks in the draft.
They have 37 or 38 composite 5*s rn. And I think 32 would be high enough on Rivals to finish as a 5*.He's a composite 5* on 247. It doesn't make sense because I don't think he's a 5* on any particular service, but he is listed as one in the composite.
It doesn't matter to me. BOT was saying he wasn't a 5*, so I checked all the services. I hope we get this kid because he could end up being the best pass rusher in this class.They have 37 or 38 composite 5*s rn. And I think 32 would be high enough on Rivals to finish as a 5*.
Regardless,
Mathematically it's technically possible to not be a 5* on any site but still be a composite 5* if you're close and have low variance across sites, which he does.
Assume for simplicity's sake only top-10 players are 5*s and compare between 2 guys:
One is
#1
#10
#40
#200
5 star on 2 sites. Average is 62.75 though.
Compared to
#11
#12
#13
#14
5 star on 0 sites but average is 12.5
Ofc the average isn't their ranking, but determines slotting hierarchy. So the 2nd kid would actually have a higher composite ranking than the 1st kid (this isn't exactly how it all works ofc, but bear with me for simplicity's sake). Now imagine all other 9 5*s are ranked similarly wonky to kid #1. It's at least conceivable that kid #2 could actually be the #1 composite player, in fact.
This is obviously an extreme example, but just to prove the point.