How do leagues stack up?

I will say this. SC del Sol 07s have dominated SW GA conf didn't lose a game last year. Winning league games 4-0, 9-0, 13-0. They played at a showcase at Silverlakes against SW ECNL and lost every game to mid table teams and an 08 team. In the SW ... GA isn't close to ECNL.

This is one of the tournaments were they mixed GA and ECNL on the girls side. Scores speak for them selves

Link to described bracket

Yes, that particular tournament seems to show what you're suggesting, but not sure that is in conflict with any other belief of rankings/strength, right? SC Del Sol are 42nd in the country, with a 47.85. The 2008 team they lost to is the #4 best team in the country, #1 in state. The Eagles 07 show #43 nationally, with a 47.84, so the likelihood is it would be a close game, but Eagles ended up taking it. Slammers are #23 in the country (48.41), and would be expected to end up on top. As far as I can tell, the top-rated AZ GA team (SC Del Sol) would be expected to show mid-pack or slightly lower in SC ECNL; that's the prediction, and that was borne out in the showcase.

The comparing league to league metrics are a challenge - not because the math is impossible, but you need to define what would be a proper comparison. Comparing best teams vs best teams? Average strength of all teams vs average strength of teams? Average strength if you drop the outliers on the top and the bottom? Comparing a specific team vs. another specific team doesn't leave nearly as much wiggle room to define terms. The footnote in one of the images earlier in the thread says it all, in that the best teams from "lower" brackets can sometimes compete well into the higher brackets. In addition, some of the lowest teams in the highest brackets, might fall towards the bottom of the table even if they were placed into a lower bracket.

I appreciate Mark posting up those league comparison numbers, but I would be curious if he could share just a little of the metrics he used. The FAQ in SR goes into pretty deep detail about the formulas used for Club Rating, etc; so he is generally comfortable in sharing his thinking on these.
 
Link to described bracket

Yes, that particular tournament seems to show what you're suggesting, but not sure that is in conflict with any other belief of rankings/strength, right? SC Del Sol are 42nd in the country, with a 47.85. The 2008 team they lost to is the #4 best team in the country, #1 in state. The Eagles 07 show #43 nationally, with a 47.84, so the likelihood is it would be a close game, but Eagles ended up taking it. Slammers are #23 in the country (48.41), and would be expected to end up on top. As far as I can tell, the top-rated AZ GA team (SC Del Sol) would be expected to show mid-pack or slightly lower in SC ECNL; that's the prediction, and that was borne out in the showcase.

The comparing league to league metrics are a challenge - not because the math is impossible, but you need to define what would be a proper comparison. Comparing best teams vs best teams? Average strength of all teams vs average strength of teams? Average strength if you drop the outliers on the top and the bottom? Comparing a specific team vs. another specific team doesn't leave nearly as much wiggle room to define terms. The footnote in one of the images earlier in the thread says it all, in that the best teams from "lower" brackets can sometimes compete well into the higher brackets. In addition, some of the lowest teams in the highest brackets, might fall towards the bottom of the table even if they were placed into a lower bracket.

I appreciate Mark posting up those league comparison numbers, but I would be curious if he could share just a little of the metrics he used. The FAQ in SR goes into pretty deep detail about the formulas used for Club Rating, etc; so he is generally comfortable in sharing his thinking on these.
Ok, here is the methodology used by Mark at the Soccer Rankings app for his analysis of the comparison between each league. I just needed to read a bit lower on his FB post and I would have seen it! Like I said before, partisans of one league or another can question the methodology, but Mark is the only one doing this in a scientific way for the time being, so I think this has to be the starting point for any discussion. Looking at one tourney, or one age group, or cherry-picking examples can get you any result you want to find, so take all those posts with a grain of salt. Here is what Mark did, per his FB post:

"
We are regularly asked "Which is the best league?" USA Sport Statistics analyzed the Soccer Rankings database to find the answer. We calculated the rating of the average team in each league for each age group and gender. We looked at the 2006 - 2010 age groups and found the rankings were the same. We also calculated the difference between the average team in each league and the average team in the best league. For example, the average boys MLS Next team would beat the average boys Elite 64 team by about 2.3 goals.
Our takeaway is that there is a lot of overlap in quality between MLS Next and ECNL National boys. However, Elite 64, EA and ECNL Regional boys are quite similar and a lot weaker. For girls, ECNL National is by far the strongest competition.
Download the app by visiting https://usasportstatistics.net or find the Soccer Rankings app on Google Play and Apple."

He's really good about replying to emails and tweaking things based on user comments, so download his app - pay for the pro version - and then dig further by asking him questions if you really want to get to the bottom of the question: "which league is best?"
 
I think it’s hard to compare individual teams when they aren’t regularly playing each other. On the girls side, I pay attention to the 07 age group since my DD plays in that age group on an 07 ECNL team. Last year’s 07 GA final saw FCV beat NEFC 2-1. We live in the north east, and those two have played some local ECNL teams in the last few months:

SUSA 2 -NEFC 0
Scorpions 1- NEFC 1
PDA Blue 4- FCV 1
Penn Fusion 1- FCV 0
Penn Fusion 1- FCV 2
World Class 2- FCV 1

That’s 1W-1D-4L for the two 07GA finalists, however most of those games were competitive. I think there are 5-6 other 07 GA teams who would have similar competitive results against these same teams, but I’d bet the depth drops off substantially after that.
 
Ok, here is the methodology used by Mark at the Soccer Rankings app for his analysis of the comparison between each league. I just needed to read a bit lower on his FB post and I would have seen it! Like I said before, partisans of one league or another can question the methodology, but Mark is the only one doing this in a scientific way for the time being, so I think this has to be the starting point for any discussion. Looking at one tourney, or one age group, or cherry-picking examples can get you any result you want to find, so take all those posts with a grain of salt. Here is what Mark did, per his FB post:

"
We are regularly asked "Which is the best league?" USA Sport Statistics analyzed the Soccer Rankings database to find the answer. We calculated the rating of the average team in each league for each age group and gender. We looked at the 2006 - 2010 age groups and found the rankings were the same. We also calculated the difference between the average team in each league and the average team in the best league. For example, the average boys MLS Next team would beat the average boys Elite 64 team by about 2.3 goals.
Our takeaway is that there is a lot of overlap in quality between MLS Next and ECNL National boys. However, Elite 64, EA and ECNL Regional boys are quite similar and a lot weaker. For girls, ECNL National is by far the strongest competition.
Download the app by visiting https://usasportstatistics.net or find the Soccer Rankings app on Google Play and Apple."

He's really good about replying to emails and tweaking things based on user comments, so download his app - pay for the pro version - and then dig further by asking him questions if you really want to get to the bottom of the question: "which league is best?"
I never denied that ECNL has some top girls teams/clubs. What bothered me was the "GA is slowly sinking statement" which was based on data that doesnt exist because for the most part GA teams are playing league games against themselves right now. (Not tournaments against ECNL teams)

This summer at events like Surf Cup the chips will fall where they may + assessments can be made with real data.

My suspicion is that ranking apps weight current data over older data + if GA teams arent doing tournaments right now their rankings would slowly move down.
 
I think it’s hard to compare individual teams when they aren’t regularly playing each other. On the girls side, I pay attention to the 07 age group since my DD plays in that age group on an 07 ECNL team. Last year’s 07 GA final saw FCV beat NEFC 2-1. We live in the north east, and those two have played some local ECNL teams in the last few months:

SUSA 2 -NEFC 0
Scorpions 1- NEFC 1
PDA Blue 4- FCV 1
Penn Fusion 1- FCV 0
Penn Fusion 1- FCV 2
World Class 2- FCV 1

That’s 1W-1D-4L for the two 07GA finalists, however most of those games were competitive. I think there are 5-6 other 07 GA teams who would have similar competitive results against these same teams, but I’d bet the depth drops off substantially after that.
Sure there's a drop off, but there's a substantial drop off in any league as you go down the standings. Also, the two best teams don't necessarily play each other in the final every year. Those 2 are near the top of 07 GA, but looks like there are 4-5 teams ranked above them both. That's the problem with picking examples, it doesn't necessarily represent the entire data set. But anyway, no one here is arguing GA tops ECNL, but the question posed was whether GA "is slowly sinking lower" and I don't think there's any evidence of that yet. More cross-league play would be wonderful, but each league has their reasons for not allowing more of it.
 
Ok, here is the methodology used by Mark at the Soccer Rankings app for his analysis of the comparison between each league.

Thanks! I should have read down that page as well. Average of all teams is a defensible option to rank 'em.

He's really good about replying to emails and tweaking things based on user comments, so download his app - pay for the pro version - and then dig further by asking him questions if you really want to get to the bottom of the question: "which league is best?"

Yep - he's quite responsive, and a couple of the tweaks I have suggested over the past few months have made it into the app.

I think it’s hard to compare individual teams when they aren’t regularly playing each other.

Maybe, but that is the question that SR tries to partially solve as objectively as possible, rather than just subjectively. With enough game data, you can fairly compare two teams - even if those two teams themselves never play eachother directly. The farther away in age, geography, league, all factors in to how valid the ratings comparison might be - but overall the predictability appears to be higher than some would want to admit.

On the girls side, I pay attention to the 07 age group since my DD plays in that age group on an 07 ECNL team. Last year’s 07 GA final saw FCV beat NEFC 2-1. We live in the north east, and those two have played some local ECNL teams in the last few months:


SUSA 2 -NEFC 0
Scorpions 1- NEFC 1
PDA Blue 4- FCV 1
Penn Fusion 1- FCV 0
Penn Fusion 1- FCV 2
World Class 2- FCV 1

That’s 1W-1D-4L for the two 07GA finalists, however most of those games were competitive. I think there are 5-6 other 07 GA teams who would have similar competitive results against these same teams, but I’d bet the depth drops off substantially after that.


SUSA 2 -NEFC 0 (11/25)
Scorpions 1- NEFC 1 (11/26)
PDA Blue 4- FCV 1 (11/19)
Penn Fusion 1- FCV 0 (11/20)
Penn Fusion 1- FCV 2 ?
World Class 2- FCV 1 ?

I can't find the last two games looking through any of the team histories, what dates should they show up on? But regardless, the ratings for all 7 of these teams appear to fairly represent their performance, even in just these individual game examples. In only 1 case of the 6, did a lesser rated team beat a higher rated team (Penn Fusion over FCV 11/20). (World Class and FCV are within .02 of each other)

GA teams.jpg

With GotSport down, it was harder to get an official list of GA teams to review, but SR was helpful to provide a decent list. It looks like the ratings follow your intuition here. There are only 4 GA teams nationally with a higher rating (Nationals GA Blue, SC Wave GA, Lou Fusz, and Ukrainian Nationals Philly. Below FCV there are teams like Cincinnati United, and IMG Elite. Looked at another way, GA shows zero teams in the top 10, 3 in the top 20, 12 in the top 50, 26 in the top 100 (see below).

GA1.jpgGA2.jpgGA3.jpg

My suspicion is that ranking apps weight current data over older data + if GA teams arent doing tournaments right now their rankings would slowly move down.

While that makes intuitive sense, I have seen the opposite over time for teams that don't play for a month or 3. It's not that ratings go down much over time, it's that they stay the same numerically, or even move up slightly, right up until they drop out entirely to unranked if the team stops playing for 7 months. I believe there are a couple factors that could explain that behavior. At all ages other then the very young, over time there are more teams dropping out and stopping playing entirely rather than new teams created, so as teams drop out of the rating systems, the remainder continue to get bumped up. I also think there is some age-weighting factor that is continually applied to teams throughout the year. It's shown by why 2007's are higher than 2008's are higher than 2009's etc. Yes - any interplay between the ages causes those ratings shifts directly for the two teams immediately involved, but it seems there is an upward nudge for teams that haven't played at all as well. In a specific case, one of our teams hasn't had a scored game since 11/20, and their rating has gone up almost a full point even prior to their first spring tournament in a week or two.
 
Sure there's a drop off, but there's a substantial drop off in any league as you go down the standings. Also, the two best teams don't necessarily play each other in the final every year. Those 2 are near the top of 07 GA, but looks like there are 4-5 teams ranked above them both. That's the problem with picking examples, it doesn't necessarily represent the entire data set. But anyway, no one here is arguing GA tops ECNL, but the question posed was whether GA "is slowly sinking lower" and I don't think there's any evidence of that yet. More cross-league play would be wonderful, but each league has their reasons for not allowing more of it.
 

Attachments

  • A3919CB2-213B-4B98-B8FE-4615F341C4AE.jpeg
    A3919CB2-213B-4B98-B8FE-4615F341C4AE.jpeg
    560.5 KB · Views: 16
  • 042F69C2-2DAF-47BC-905A-6AA9C3E589B2.jpeg
    042F69C2-2DAF-47BC-905A-6AA9C3E589B2.jpeg
    1.3 MB · Views: 17
Michigan Hawks and Nationals scrimmage each other every year, which I think is awesome. I wish the clubs where we lived would do that, but it does not happen very often. This weekend, for example, FC Stars & Scorpions 2007 ECNL & 2007 NEFC & South Shore Select GA are all driving 4 hours to NY to play in the SUSA showcase, and not playing a single game against each other.
 
When your club hosts a tournament you get to define the brackets. This is why its important if it's a GA or ECNL club hosting.

Again this summer after league games + when GA teams start participating in the big (mixed) tournaments rankings will define themselves.

Just to reiterate you stated that "As a side note on the girls side, the GA teams in the Southwest have been in a slow decline in the rankings for 2007 and younger throughout this season." Then used Summer and early fall tournament results to backup your statement..

The truth is that you have no idea how GA teams are performing after summer / early fall + are making statements not base on facts/results.
One data point. Albion 06 GA National Champs lost to Surf 06 ECNL by a score of 8-0 in November last fall. That’s all the data needed to see the difference between ECNL and GA.
 
Soccerwire did an article on this same subject last fall.


"While 7 of the top 10 clubs represent the ECNL, GA clubs occupied 2 out of the top 3 spots"

IMO, it's a huge step backwards for Soccerwire. They stopped using YSR (and now SR) because they believe it is too focused on current results - but in doing so they are ignoring the realities of the differences between leagues. Their algorithm now treats how well you finish in GA equivalent to how well you finish in ECNL, by definition. Then tries to fix it by adding in things like how many players are pulled up. It ignores the fact that on the field right now, the average ECNL team is significantly better than the average GA team, and the best ECNL teams are significantly better than the best GA teams. Just review the top 10 (Girls clubs) of each to see the wide discrepancies:

ga ecnl soccerwire.jpg

The only real agreement is that Solar SC is top dog either way. But Soccerwire's #2 club barely makes the top 20 in actual results. Their #3 club is #16 on SR. Soccerwire puts the Blues at #23 rather than #3. Puts DKSC at #40 rather than #5.

At the end of the day, the rankings only matter so much (and perhaps the answer is not much). But Soccerwire's are now moving in the direction of irrelevancy like GotSoccer/GotSport rankings, if the goal of the ranking/rating is to answer the question who would beat who on the field this afternoon.
 
(Slammers HB Koge shows as a low outlier in SR, as the club doesn't field a 2012 or 2011 team, so is given an average rating of "30" for that year. SR requires 5 teams of 7, from 2012-2006, but that is the penalty for the missing teams).
 
I asked Mark from Soccer Rankings to run his league analysis with only the California Region. He has previously stated that he can not separate NorCal and SoCal on the App. So California as a whole is as specific as we can get for now. Here is a link to Facebook with the results. https://www.facebook.com/1000882331...L2Jj9cRzzq6jjUVfQyVdjRieDGol/?mibextid=Nif5oz
I still think California GA will slip into #3 by the end of this summer. I'll ask him to update it again after the Labor Day tournaments are uploaded. Looks like currently ECNL averages 2GD above GA. GA, E64, ECRL, NPL are all within 1GD of each other.
 
I asked Mark from Soccer Rankings to run his league analysis with only the California Region. He has previously stated that he can not separate NorCal and SoCal on the App. So California as a whole is as specific as we can get for now. Here is a link to Facebook with the results. https://www.facebook.com/1000882331...L2Jj9cRzzq6jjUVfQyVdjRieDGol/?mibextid=Nif5oz
I still think California GA will slip into #3 by the end of this summer. I'll ask him to update it again after the Labor Day tournaments are uploaded. Looks like currently ECNL averages 2GD above GA. GA, E64, ECRL, NPL are all within 1GD of each other.
Amazing, even after your ranking app guy said GA isnt bad you keep pressing your unsubstantiated narrative that GA league is regressing.

Try being less of a homer + be objective when recognizing talent. When/if players make it to the college level nobody will remember or care which league they played in as a youth.
 
New visual reference that reflects the current SR California League Analysis data. If the average was within 2 GD of the Top League, I placed them in the same Tier. I figure within 2 GD is same level of play. Any more than that, and they move down to the next Tier (Tiers are in 2 GD increments). Get your excuses ready.
Capture.PNG
 
Amazing, even after your ranking app guy said GA isnt bad you keep pressing your unsubstantiated narrative that GA league is regressing.

Try being less of a homer + be objective when recognizing talent. When/if players make it to the college level nobody will remember or care which league they played in as a youth.
Like you said, we will see after the summer tournaments.
 
Like you said, we will see after the summer tournaments.
My money is on GA & E64 rising slightly and ECRL falling off mostly based on kids not wanting to play on a "B" team when "A" teams in strong leagues are recruiting them. Nothing says a club believes in your kid like putting them on the top team.
 
Back
Top