'04 Girls FBR Rankings - 02-28-12

Istanbul2005 wrote:

Busby Babes wrote:I would liken these rankings to a scaled down version of the FIFA rankings.. Everyone knows the FIFA rankings are shite and full of inaccuracies. The Elo ranking system is a much more truer version, and considered the gold standard for football rankings..

Of course, the ranking of U10 football in itself is the stuff of fantasy land....

Having said that, Bw gophers enjoys taking the time to do these, so that deserves an A for effort..

Interesting and something to talk about if nothing else. If Scotland or Ireland were subject to this sort of calculation, they'd be forever ranked top 10 because they play top European countries and lose. Only flaw I see in these are teams that lose to higher ranked teams are rewarded the A for effort. Also, Brazil wouldn't rank so well because they beat up on their poor soccer neighbors.

KnKsDad wrote:

Istanbul2005 wrote:

Busby Babes wrote:I would liken these rankings to a scaled down version of the FIFA rankings.. Everyone knows the FIFA rankings are shite and full of inaccuracies. The Elo ranking system is a much more truer version, and considered the gold standard for football rankings..

Of course, the ranking of U10 football in itself is the stuff of fantasy land....

Having said that, Bw gophers enjoys taking the time to do these, so that deserves an A for effort..

Interesting and something to talk about if nothing else. If Scotland or Ireland were subject to this sort of calculation, they'd be forever ranked top 10 because they play top European countries and lose. Only flaw I see in these are teams that lose to higher ranked teams are rewarded the A for effort. Also, Brazil wouldn't rank so well because they beat up on their poor soccer neighbors.

poor soccer neighbors like Argentina and that country that Soares comes from..

Master Bates wrote:

KnKsDad wrote:

Istanbul2005 wrote:

Busby Babes wrote:I would liken these rankings to a scaled down version of the FIFA rankings.. Everyone knows the FIFA rankings are shite and full of inaccuracies. The Elo ranking system is a much more truer version, and considered the gold standard for football rankings..

Of course, the ranking of U10 football in itself is the stuff of fantasy land....

Having said that, Bw gophers enjoys taking the time to do these, so that deserves an A for effort..

Interesting and something to talk about if nothing else. If Scotland or Ireland were subject to this sort of calculation, they'd be forever ranked top 10 because they play top European countries and lose. Only flaw I see in these are teams that lose to higher ranked teams are rewarded the A for effort. Also, Brazil wouldn't rank so well because they beat up on their poor soccer neighbors.

poor soccer neighbors like Argentina and that country that Soares comes from..

Brazil are currently ranked 18th

go99 wrote:

KnKsDad wrote:

go99 wrote:You guys got caught sleeping on Scott.

Just a general comment/question - what's to get caught sleeping on? Is it possible that the Scott-Deleon hyphenation may have something to do with the teams success? I really know nothing about them, but just asking the question based on something that seems obvious.

Furthermore, I suspect that when the rubber meets the road in SDL 9v9 it'll probably me more than just City showing up, and understandably so, that is welcomed. The question I have is in that scenario if City beats the "higher" ranked team how does that effect the higher ranked teams ranking? That's where there could be some murkiness in the rankings. Not sure BW can account for all of those details though. All in all, still pretty accurate.

Nope nothing at all to do with the deleone hyphenation. Was always one of the most talented, athletic groups in the 04's then they got the coaching and its starting to pay off. Don't know anything about that city nonsense

SoftballDad wrote:I heard that DT Scott only just scraped by a Sting team ranked outside the top 20 by 2-1 today
Heard the 2 goals were PKs too. Can anyone confirm??

GinkoBiloba11 wrote:Someone correct me if I'm wrong but Scott the highly ranked team is totally separate from Scott/Deleon city and Scott/Deleon city2? 3 totally different teams. Notice Deleon's name is not on the Scott highly ranked teams sdl schedule . Don't get them confused.

SoftballDad wrote:I heard that DT Scott only just scraped by a Sting team ranked outside the top 20 by 2-1 today
Heard the 2 goals were PKs too. Can anyone confirm??

rattler99 wrote:Any score for Armstrong/Dalglish?

Busby Babes wrote:I would liken these rankings to a scaled down version of the FIFA rankings.. Everyone knows the FIFA rankings are shite and full of inaccuracies. The Elo ranking system is a much more truer version, and considered the gold standard for football rankings..

Istanbul2005 wrote:
Yes, I know. Speaking more historically.
They are ranked 18th because they haven't been winning. But in the days of #1 for consecutive years, they would play third world countries and win 5-0 and be on top (pre recalibration of the formula)

Losing to Argentina twice, England, Mexico, and a draw to Colombia should put Brazil in the top 8!

South Africa lost to Brazil 1-0, that'd put them in the top 20 in the FBR..
Don't get me wrong, I'm like the majority of the hundreds of viewers and enjoy the latest release of FBR, but, win 2 out of 10 games in SDL2 and you've got yourself a place in the Top 30 on this page!

O.K. Let me make sure I've got this straight...

I've got one "expert" (Busby Babes), telling me that FBR is shite** because it's like FIFA rankings, where Brazil is ranked 18th, and I should use an ELo method, where Brazil is ranked 3rd...

I've got another "expert" (Istanbul2005), telling me that FBR is shite** because FBR would have Brazil ranked in the top 8, which means that FBR is actually designed to be much closer to ELo, than to FIFA...



Of course, the real reason is that I've looked at all 3 formulae and determined that FBR is shite** because it's nothing like either of these two ranking systems...

Istanbul2005 wrote:I think the top 10 to 15 are very accurate, like the example you provided in the older group.
It's the 20-40 range where the flaws are glaring. Can't stop teams borrowing to do their best to keep it competitive v the top tier, or top teams taking it easy on these teams by playing their bench, being respectful, etc. Teams that have won just 2 out of 10 (2-6-2), or have played just 3 games (and winning one of them) shouldn't be ranked as high as they are. Perhaps a minimum # of wins or the minimum # of games played should be altered.

Istanbul2005 wrote:I think the top 10 to 15 are very accurate, like the example you provided in the older group.
It's the 20-40 range where the flaws are glaring. Can't stop teams borrowing to do their best to keep it competitive v the top tier, or top teams taking it easy on these teams by playing their bench, being respectful, etc. Teams that have won just 2 out of 10 (2-6-2), or have played just 3 games (and winning one of them) shouldn't be ranked as high as they are. Perhaps a minimum # of wins or the minimum # of games played should be altered.

bwgophers wrote:

Istanbul2005 wrote:I think the top 10 to 15 are very accurate, like the example you provided in the older group.
It's the 20-40 range where the flaws are glaring. Can't stop teams borrowing to do their best to keep it competitive v the top tier, or top teams taking it easy on these teams by playing their bench, being respectful, etc. Teams that have won just 2 out of 10 (2-6-2), or have played just 3 games (and winning one of them) shouldn't be ranked as high as they are. Perhaps a minimum # of wins or the minimum # of games played should be altered.

While I am not going to agree that FBR has "glaring flaws" in the 20-40 range, I will agree that the rankings definitely get "noisier" once you get outside the top 25-30.

I will also say that ANY ranking system... FBR, Human Polls, Power Rankings, ELo, FIFA, etc. will all struggle with the same issues that are mentioned above (guest players, mismatches/blowouts, large disparity in total # of games played, how to handle relative strength of teams playing in different leagues/divisions, etc.) It's the nature of the game when you are trying to do something as absurd as ranking 8-10 year old girls academy soccer teams in NTX.

If the ranking methodology is based on some kind of consistent, valid, mathematical formula, or, in the case of human polls, there is enough averaging involved to dampen out individual subjectivity, my experience is that all of these methods provide roughly equivalent levels of overall accuracy.

bwgophers wrote:

Istanbul2005 wrote:I think the top 10 to 15 are very accurate, like the example you provided in the older group.
It's the 20-40 range where the flaws are glaring. Can't stop teams borrowing to do their best to keep it competitive v the top tier, or top teams taking it easy on these teams by playing their bench, being respectful, etc. Teams that have won just 2 out of 10 (2-6-2), or have played just 3 games (and winning one of them) shouldn't be ranked as high as they are. Perhaps a minimum # of wins or the minimum # of games played should be altered.

While I am not going to agree that FBR has "glaring flaws" in the 20-40 range, I will agree that the rankings definitely get "noisier" once you get outside the top 25-30.

I will also say that ANY ranking system... FBR, Human Polls, Power Rankings, ELo, FIFA, etc. will all struggle with the same issues that are mentioned above (guest players, mismatches/blowouts, large disparity in total # of games played, how to handle relative strength of teams playing in different leagues/divisions, etc.) It's the nature of the game when you are trying to do something as absurd as ranking 8-10 year old girls academy soccer teams in NTX.

If the ranking methodology is based on some kind of consistent, valid, mathematical formula, or, in the case of human polls, there is enough averaging involved to dampen out individual subjectivity, my experience is that all of these methods provide roughly equivalent levels of overall accuracy.

bigtex wrote: I found the average FBR placement variance to be 2.3 for the top 10 teams, 3.5 for FBR teams ranked 11-20 and 5.5 for FBR teams ranked 21-30 and 5.1 for FBR teams ranked 31-40. Obviously too much time on my hands for lunch and this is not a bid to take over FBR when bwgophers moves on- I am not that crazy.