'05 Girls FBR Rankings - 11-18-14

Ok this ranking system doesn't make any sense. Last weeks number 10 ranked team tied Solar Abbott and were beat by #6 and #1. So all results as expected. Yet they move down two. Solar Abbott tied to two lower ranked teams and lost to the #6 team and they didn't move at all...can someone explain? Not to mention several other teams lost this weekend and that isn't reflected either.

Not questioning position, but I notice Pulpaneck shows 1 loss where I remeber they lost to solar Abbott and dfeeters this past weekend. DTS shows 0 losses yet they lost to rush.

If I remember correctly, games that go to OT or maybe it's PK's, show up as ties in FBR.
Gopher- how did I do?

What about that shot my DD took that hit the post. Would have gone in had it not been for that stupid gust of wind. Does that count as a win in the FBR?

That makes no sense!

Does it show they have one loss or does it show they have 1 draw and 3 losses?

BWG can explain it much better (I think he's been explaining it to every age group for the past few years), but it's basically a sliding window of games that's being ranked. So if a bad losses slides outside of that sliding window of games you are being ranked on then you may jump up in the rankings even though you haven't done anything. Same things if you happen to have a big win drop off and replace it with a couple of (statistically) mediocre wins, then you might actually drop a few points even though you are winning.

I think you are right, if it ends up going to PK's then I think it's counted as a draw in the FBR. Some may not like the FBR but it's actually pretty darn accurate when it comes to LH's year one qualifiers.

Slammed at work today (yes contrary to popular belief, I actually do have a life and am gainfully employed... ).

I will address all of the questions when I get free.

Soccer1108 wrote:Ok this ranking system doesn't make any sense. Last weeks number 10 ranked team tied Solar Abbott and were beat by #6 and #1. So all results as expected. Yet they move down two. Solar Abbott tied to two lower ranked teams and lost to the #6 team and they didn't move at all...can someone explain? Not to mention several other teams lost this weekend and that isn't reflected either.

It's not that your results were terrible, it's that Gerrard got a big 7 point push and D'Feeters got 6 points. Their better results dropped you down a few. Nothing to worry about being ranked 12th in this group. As you can see from tourney results anything can happen. If you have enough of a point spread in the rankings a loss may not move you down at all....I think lol

O.K.  From the top...

The primary purpose for my publishing the FBR rankings is to provide a reasonably accurate snapshot of the relative strength of teams in the age group.  Hopefully, coaches/managers and league/tournament directors  can use the information from FBR, to select and set up appropriate divisions/brackets that will provide the proper levels of competition that optimize player development.

Hopefully, parents will also find FBR useful as a GUIDE for scouting out potential teams that may be an appropriate fit for their DD's.  Although I would NEVER recommend that a parent use a team's FBR rank as a primary reason for joining a team.  First and foremost, keep in mind that FBR is nothing more than a mathematical formula based on game results.  FBR will tell you NOTHING about the style of play, or whether the coach develops players vs. simply recruits the best athletes, whether the coach is a great guy/gal or a complete raving lunatic a$$hole, or whether the parents on the team are such.

Now, of course, one of the primary side-benefits of publishing FBR, is that a large number of you appear to enjoy FBR, and the endless forum debate that it generates.

When we started FBR back with the '01's, the challenge for me, was to see how accurate FBR would be at predicting the outcome of QT at U11.  Turns out, it's been pretty good... to the tune of close to 90% accurate at predicting D1 Supergroup qualifiers (that's predicting and average of 18 out of 20 D1 qualifiers correctly), and ~86% accurate at predicting all qualifiers for LHGCL at U11 (that's predicting an average of 26 out of 30 total qualifiers correctly) over the past 4 years ('01's-'04's).

Now, with all of that said, I am NOT trying to crown some kind of mythical academy champion with FBR.  While I strive for accuracy and a defensible ranking for every single team in FBR, it doesn't always happen.  Sometimes teams will be off a few spots.  No ranking is system is perfect, and certainly FBR isn't.  It's just pretty darn good.

So... if you are getting all worked up about a team being ranked a few spots higher or lower than you think it should, well, frankly, you are missing the point.

O.K. Off my soapbox. More to come...

05girls wrote:Not questioning position, but I notice Pulpaneck shows 1 loss where I remeber they lost to solar Abbott and dfeeters this past weekend.  DTS shows 0 losses yet they lost to rush.  

futbollove wrote:If I remember correctly, games that go to OT or maybe it's PK's, show up as ties in FBR.
Gopher- how did I do?

05girls wrote:What about that shot my DD took that hit the post.  Would have gone in had it not been for that stupid gust of wind.  Does that count as a win in the FBR?  

That makes no sense!

You nailed it futbollove.

I made the decision 4+ years ago when I started doing FBR, that since it's a mathematical formula, and since I'm not trying to crown any kind of "champion" with FBR, that it made the most sense to use as much game data as possible, and to treat all games as equally as possible. So, for the sake of treating all games equally, I don't give any different weight to league games vs. tournament pool games vs. tournament knockout games, and I only count the score during regulation time since league games and tournament pool games are not allowed to go to OT or PK's.

Now, that is incumbent on my actually knowing that a tournament knockout game, or a league playoff/championship game went to OT or PK's. I have to rely on what's reported on the forum for that. If I know about it, I'll enter it as a draw. If I don't know about it, it will end up going in as a W-L. If someone brings it to my attention a few days or a few weeks down the road, I'll go in and correct it in FBR.

By the way, similar logic applies to forfeits. If I know a game was a forfeit, I won't include it in FBR as it serves no purpose in what I'm trying to accomplish with FBR. However, I don't have time to track down if every 3-0 game was a forfeit or not (most aren't), so it's always possible that one may sneak into the rankings. Again, if I find out about it after the fact, I'll go in and correct it.

Anarchy wrote:

Soccer1108 wrote:Ok this ranking system doesn't make any sense. Last weeks number 10 ranked team tied Solar Abbott and were beat by #6 and #1. So all results as expected. Yet they move down two. Solar Abbott tied to two lower ranked teams and lost to the #6 team and they didn't move at all...can someone explain? Not to mention several other teams lost this weekend and that isn't reflected either.

It's not that your results were terrible, it's that Gerrard got a big 7 point push and D'Feeters got 6 points. Their better results dropped you down a few.  Nothing to worry about being ranked 12th in this group. As you can see from tourney results anything can happen. If you have enough of a point spread in the rankings a loss may not move you down at all....I think lol

Anarchy has it explained pretty good. Based on "body of work" heading into the weekend, teams 8-15 from last week were bunched up fairly tight in total ranking points (~7 points total separating #8 from #15), while Solar Abbott was ~6 points clear of FCD Premier. D'Feeters and LP Gerrard had better weekends than DTMC and Sting Scarfone, and when the points are that close, you'll end up with teams shuffling some spots. Abbott and DTMC both had about the same movement in total points from last week to this week, which is correct given that they had similar performances. It's just that Abbott had enough of a cushion that even though FCD Premier, D'Feeters, and LP Gerrard gained on them, it wasn't enough to swap spots.

Bottom line, depending on what point of emphasis you want to make, you could pretty much make an argument for just about any order of the teams from ~8-15, and most of those arguments would likely put DTMC somewhere between 9-13. It just so happens that the formula that FBR uses puts them at #12 this week. Again, if you are worried about a couple of spots one way or the other in FBR, you really are missing the point.

Actually thanks, I am a numbers person and just trying to understand the logic. Not worried about movement within a few just trying to understand why some with same record moved while others did not.

How accurate were the final FBR rankings versus the final fall LH standings (top 10) for this year's u11s? Was there much variance?

WisdomTeeth wrote:How accurate were the final FBR rankings versus the final fall LH standings (top 10) for this year's u11s? Was there much variance?

I think it was 90% (18 out of 20) that made but can't remember all of the variations on who settled where.