247 Final Update Bias, tldr warning

[LouderVol]

I crunched the numbers to see what the implied "bias" might be of 247 vs a given team in the SEC.
I did this by checking how much each teams commits moved up or down, and then averaging. So that even if some players dropped, overall a school could have improved.

Granted this doesnt actually prove anything but the final numbers and some math, but if you buy into247, or others, being biased this gives one metric that MIGHT back up the theory. Basically the more a school averages move, the more "biased" 247 was for/against that school in this one metric A low number means the service got it pretty much correct. Again this is based on the already flawed system of a recruiting service so take with a grain of salt. And I only did the Top247, not the composite, or those not in the top 247.

Remember the farther from 0, the further the theoretical "bias". And theoretically a negative average would mean they are biased FOR the school, a positive would mean they are biased AGAINST the school as the players got ranked higher.

Alabama average: +5.1
Arkansas average: -29.5
Auburn average: -12.3
Florida average: 2.8
Georgia average:1 (least "biased" in SEC)
Kentucky average: -22
LSU Average: 18.4
Mississippi State average: -10
Missouri average: N/A no 247 commits
Ole Miss average: 44
South Carolina average: 31 2
Tennessee average: 63.25
Texas A&M average: 25
Vanderbilt Average: n/a, no 247 commits.

So yes, according to this one incredibly limited metric 247 hates us the most in the SEC. We have 8 in the top 247, so we arent like Olemiss with 2 and one big mover. We had 5 guys make moves of more than 50 spots. And one of those was down. So more than half our class made big, bias correcting, moves at the end.

Let the debate!

 

[RollerVol]

thanks, wow...that's a lot of work...so now, we know 247 and ESPN are biased against us...

GO BIG ORANGE!

 

[Pridedad]

I crunched the numbers to see what the implied "bias" might be of 247 vs a given team in the SEC.
I did this by checking how much each teams commits moved up or down, and then averaging. So that even if some players dropped, overall a school could have improved.

Granted this doesnt actually prove anything but the final numbers and some math, but if you buy into247, or others, being biased this gives one metric that MIGHT back up the theory. Basically the more a school averages move, the more "biased" 247 was for/against that school in this one metric A low number means the service got it pretty much correct. Again this is based on the already flawed system of a recruiting service so take with a grain of salt. And I only did the Top247, not the composite, or those not in the top 247.

Remember the farther from 0, the further the theoretical "bias". And theoretically a negative average would mean they are biased FOR the school, a positive would mean they are biased AGAINST the school as the players got ranked higher.

Alabama average: +5.1
Arkansas average: -29.5
Auburn average: -12.3
Florida average: 2.8
Georgia average:1 (least "biased" in SEC)
Kentucky average: -22
LSU Average: 18.4
Mississippi State average: -10
Missouri average: N/A no 247 commits
Ole Miss average: 44
South Carolina average: 31 2
Tennessee average: 63.25
Texas A&M average: 25
Vanderbilt Average: n/a, no 247 commits.

So yes, according to this one incredibly limited metric 247 hates us the most in the SEC. We have 8 in the top 247, so we arent like Olemiss with 2 and one big mover. We had 5 guys make moves of more than 50 spots. And one of those was down. So more than half our class made big, bias correcting, moves at the end.

Let the debate!

So your metric implies that because a recruit committed to TN they were purposely rated lower? Then why now are they moved up?

 

[Vols410]

So your metric implies that because a recruit committed to TN they were purposely rated lower? Then why now are they moved up?

This.

If anything it’s just a testament to the scouting ability of the staff. 247 wasn’t biased against the guys before they were committed, they just didn’t know who they were at the time.

 

[SmokeyDaBear]

thanks, wow...that's a lot of work...so now, we know 247 and ESPN are biased against us...

GO BIG ORANGE!

I feel like ESPN is just ignorant all the way around, but I like numbers so thanks to OP.

 

[Pridedad]

This.

If anything it’s just a testament to the scouting ability of the staff. 247 wasn’t biased against the guys before they were committed, they just didn’t know who they were at the time.

That is kinda my point. Look at Arkansas. Their composition drop was -29.5.

 

[volsagain]

I crunched the numbers to see what the implied "bias" might be of 247 vs a given team in the SEC.
I did this by checking how much each teams commits moved up or down, and then averaging. So that even if some players dropped, overall a school could have improved.

Granted this doesnt actually prove anything but the final numbers and some math, but if you buy into247, or others, being biased this gives one metric that MIGHT back up the theory. Basically the more a school averages move, the more "biased" 247 was for/against that school in this one metric A low number means the service got it pretty much correct. Again this is based on the already flawed system of a recruiting service so take with a grain of salt. And I only did the Top247, not the composite, or those not in the top 247.

Remember the farther from 0, the further the theoretical "bias". And theoretically a negative average would mean they are biased FOR the school, a positive would mean they are biased AGAINST the school as the players got ranked higher.

Alabama average: +5.1
Arkansas average: -29.5
Auburn average: -12.3
Florida average: 2.8
Georgia average:1 (least "biased" in SEC)
Kentucky average: -22
LSU Average: 18.4
Mississippi State average: -10
Missouri average: N/A no 247 commits
Ole Miss average: 44
South Carolina average: 31 2
Tennessee average: 63.25
Texas A&M average: 25
Vanderbilt Average: n/a, no 247 commits.

So yes, according to this one incredibly limited metric 247 hates us the most in the SEC. We have 8 in the top 247, so we arent like Olemiss with 2 and one big mover. We had 5 guys make moves of more than 50 spots. And one of those was down. So more than half our class made big, bias correcting, moves at the end.

Let the debate!

Well that looks like our coaches find diamonds faster than most. !!!

 

[Pridedad]

Well that looks like our coaches find diamonds faster than most. !!!

Or a classic one final "pruit bump"

 

[Devo182]

The fix is in for the usual big money bluebloods...Arkansas and Kentucky

 

[Devo182]

Wouldn't our recruits moving up suggest a bias for us, if anything? Don't see how it is the other way around.

 

[MadWorld]

@LouderVol Who moved down the 50+ spots?

 

[LouderVol]

So your metric implies that because a recruit committed to TN they were purposely rated lower? Then why now are they moved up?

just the numbers. when looking at it its hard to say Morven Joseph was ranked low because of his commitment to UT, and he was one of our biggest movers.
if you look at their explanation for Bailey they quote what he did in the season, not in the All American games/practices. so they have been sitting on his change for a while. why wait on the bump if not to make TN look bad?
if you buy they are biased they are moving them now to make it look like they aren't biased. if you ever argue that perception matters throughout the year, this is how they can keep us down, but still accommodate for players doing well.

personally I don't think these numbers prove anything. its just one way to look at the numbers to confirm my own confirmation bias

 

[LouderVol]

@LouderVol Who moved down the 50+ spots?

Len'neth Whitehead, -54 spots.

 

[LouderVol]

Wouldn't our recruits moving up suggest a bias for us, if anything? Don't see how it is the other way around.

depends on how you look at the timing. you could argue either way.
they are keeping us down during the season to hurt momentum, but at the end of the day can't deny the players some recgnition.
your argument is just the flip. they want to help us right at the end.

 

[2xVol]

I crunched the numbers to see what the implied "bias" might be of 247 vs a given team in the SEC.
I did this by checking how much each teams commits moved up or down, and then averaging. So that even if some players dropped, overall a school could have improved.

Granted this doesnt actually prove anything but the final numbers and some math, but if you buy into247, or others, being biased this gives one metric that MIGHT back up the theory. Basically the more a school averages move, the more "biased" 247 was for/against that school in this one metric A low number means the service got it pretty much correct. Again this is based on the already flawed system of a recruiting service so take with a grain of salt. And I only did the Top247, not the composite, or those not in the top 247.

Remember the farther from 0, the further the theoretical "bias". And theoretically a negative average would mean they are biased FOR the school, a positive would mean they are biased AGAINST the school as the players got ranked higher.

Alabama average: +5.1
Arkansas average: -29.5
Auburn average: -12.3
Florida average: 2.8
Georgia average:1 (least "biased" in SEC)
Kentucky average: -22
LSU Average: 18.4
Mississippi State average: -10
Missouri average: N/A no 247 commits
Ole Miss average: 44
South Carolina average: 31 2
Tennessee average: 63.25
Texas A&M average: 25
Vanderbilt Average: n/a, no 247 commits.

So yes, according to this one incredibly limited metric 247 hates us the most in the SEC. We have 8 in the top 247, so we arent like Olemiss with 2 and one big mover. We had 5 guys make moves of more than 50 spots. And one of those was down. So more than half our class made big, bias correcting, moves at the end.

Let the debate!

This also implies that our staff is the best in the conference at finding underrated but top talent. Thaks fellow stats nerd, I'll have something coming out after NSD (once the class is complete) about why Pruitt's recruiting classes are different than anything we have had in the last decade (hint: it could be what makes Pruitt the first coach since Fulmer to have sustained success here)

 

[LouderVol]

This also implies that our staff is the best in the conference at finding underrated but top talent. Thaks fellow stats nerd, I'll have something coming out after NSD (once the class is complete) about why Pruitt's recruiting classes are different than anything we have had in the last decade (hint: it could be what makes Pruitt the first coach since Fulmer to have sustained success here)

i know enough about stats to know you can make them fit just about any argument, like the other side of my OP. 247 is biased for us, instead of against us. its all in how you want to look at the numbers. this was my way.

i always have fun running numbers like this.

 

[Devo182]

just the numbers. when looking at it its hard to say Morven Joseph was ranked low because of his commitment to UT, and he was one of our biggest movers.
if you look at their explanation for Bailey they quote what he did in the season, not in the All American games/practices. so they have been sitting on his change for a while. why wait on the bump if not to make TN look bad?
if you buy they are biased they are moving them now to make it look like they aren't biased. if you ever argue that perception matters throughout the year, this is how they can keep us down, but still accommodate for players doing well.

personally I don't think these numbers prove anything. its just one way to look at the numbers to confirm my own confirmation bias

They do reranks all at once, not for individuals. They were waiting till all the all star games ended to make 1 last rerank. They get hell from every fanbase on every rerank, so I'm sure they try to consolidate them and minizimize how many occur

 

[Devo182]

depends on how you look at the timing. you could argue either way.
they are keeping us down during the season to hurt momentum, but at the end of the day can't deny the players some recgnition.
your argument is just the flip. they want to help us right at the end.

Ok. Just saying the most common idea of bias in recruiting is that some schools get bumps after they commit and later in the process. Your view is simply opposite of the typical view of bias. Like, complete opposite.

Both are just theories and not saying you're wrong. Hell, it all could just be good or bad early evaluations by staffs. Or bias. Or services following known elite evaluators (Saban, Pruitt). It's too much to disagreggate honestly.

I'll continue to follow them (as well all obviously do), with a small grain of salt.

At the end of the day, the services are in competition against one another to be "most right". Their incentive is to survive and be seen as the most reliable/accurate. That is what keeps them going, not being completely wrong by giving 5*s to all Texas or Michigan (2 biggest fanbases) recruits.

 

[bignewt]

Ugh to it all.

 

[drvenner]

I crunched the numbers to see what the implied "bias" might be of 247 vs a given team in the SEC.
I did this by checking how much each teams commits moved up or down, and then averaging. So that even if some players dropped, overall a school could have improved.

Granted this doesnt actually prove anything but the final numbers and some math, but if you buy into247, or others, being biased this gives one metric that MIGHT back up the theory. Basically the more a school averages move, the more "biased" 247 was for/against that school in this one metric A low number means the service got it pretty much correct. Again this is based on the already flawed system of a recruiting service so take with a grain of salt. And I only did the Top247, not the composite, or those not in the top 247.

Remember the farther from 0, the further the theoretical "bias". And theoretically a negative average would mean they are biased FOR the school, a positive would mean they are biased AGAINST the school as the players got ranked higher.

Alabama average: +5.1
Arkansas average: -29.5
Auburn average: -12.3
Florida average: 2.8
Georgia average:1 (least "biased" in SEC)
Kentucky average: -22
LSU Average: 18.4
Mississippi State average: -10
Missouri average: N/A no 247 commits
Ole Miss average: 44
South Carolina average: 31 2
Tennessee average: 63.25
Texas A&M average: 25
Vanderbilt Average: n/a, no 247 commits.

So yes, according to this one incredibly limited metric 247 hates us the most in the SEC. We have 8 in the top 247, so we arent like Olemiss with 2 and one big mover. We had 5 guys make moves of more than 50 spots. And one of those was down. So more than half our class made big, bias correcting, moves at the end.

Let the debate!

Lol at Andy

 

[jasonvol18]

depends on how you look at the timing. you could argue either way.
they are keeping us down during the season to hurt momentum, but at the end of the day can't deny the players some recgnition.
your argument is just the flip. they want to help us right at the end.

I hihly doubt 247 purposely keeps Vol recruits ranked low during the season to hurt momentum. Especially when guys like Joseph, Slaughter, Holiday and others were committed to other schools when they were rated lower..The more sensible and least sensitive answer is those guys had great Senior years and earned more high program offers. These movements if anything show they don't hate TN or other SEC schools. Way too many sensitive Vol fans these days. Nobody is out to get us or intentionally trying to sabotage our recruiting. Now I've heard it all.

 

[2xVol]

i know enough about stats to know you can make them fit just about any argument, like the other side of my OP. 247 is biased for us, instead of against us. its all in how you want to look at the numbers. this was my way.

i always have fun running numbers like this.

Either way you look at it I think it reflects well on the class. I do agree that there is bias against re-ranking UT commits. And while the late bump is nice it hurts recruiting when it comes so late because other recruits care about rankings and it would be a lot easier to sell them on a class full of 4 stars.

 

[firedog28]

I saw where Evans just after visiting UT lost his 5 star on one service and dropped in the rankings.The evidence is in,they hate UT.

 

[bluetick2]

Cocaine OD the night/morning after he was drafted by the Celtics. Many thought he might of had similar impact as Jordan who was a year or two ahead of him. Bird was so pumped about the Celtics drafting Bias that he was going to go through rookie camp with Bias.

Do you think he might of?

 

[MadWorld]

i know enough about stats to know you can make them fit just about any argument,

There is an old joke "74.5% of all stats are made up"