I heard that too, though I don't know where the goalkeeper is going. There is just such a shortage of good female goalkeepers. I know firsthand if one doesn't think that one's girl is being treated fairly in that position, the demand is off the charts, and there is lots of opportunity. Evolution's RL badge is definitely a limitation, and their answer seems to be to expand in the city. Feel like they are just going to keep getting poached in the western suburbs. FWIW I think that 13 Evo team holds its own versus Eclipse and Inter ECNL right now. Too bad we will probably never see it.
I’ll use all message board fervor to again beat the drum that it’s ridiculous that we don’t know because they don’t play each other. Michigan has it right. All play Friendlies ECNL/GA. Much better ecosystem. No wonder they provide more talent. It’s easy to dismiss them playing a second tier schedule. But that team has navigated from being a good grassroots team to getting to the highest spot the system would allow. I’ll continue to applaud it. We need more of it. Hopefully they can make a run in the RL playoffs this summer. FWIW I do think this Evo team would compete with Inter and Eclispe NL squads. And beat any of the GA teams. App seems to agree. Has them #2 in the state. Keeper talent at this age group seems thin compared to all the studs locally at older ages.
Amen. Wouldn’t that be nice? Never going to happen. ECNL promotion is club based. They aren’t even close at a club level to that.
IMO, having the #2 team in the state being an RL team ultimately says less about that team, and more about the (not-very) competitiveness of the leagues in general in the area, at least at that age group. Of course people can argue 'til the cows come home that until a team directly plays another team, it's impossible to "prove" which one is stronger. But if math can tell you that given all prior game data of all teams involved, one team is 90% likely to beat the other - it's not smart money to believe that "everyone's equal" until they directly play each other.
In 2013 age group, I would take Inter and Rockford vs Evo RL team. Inter & Rockford are just more battle tested with the 82 and 81 hardest schedules nationally vs 462. It’s a massive difference that cannot be accurately adjusted for until a team actually plays competitive games on a regular basis. That said, Teams can go to tournaments like Jefferson Cup or Surf Cup and get into more difficult brackets. Nothing is preventing this team ( or others) from getting a more challenging schedule. Inter literally has tournaments each May and July that their top teams play in, clubs can just sign up and play an ECNL team.
I don't have insight into the algorithm, but it does seem the rankings app gives significant weight to beating lesser teams by large goal differences. I've noticed a few teams who have dominated weaker brackets have very high rankings, and though they occasionally do play games against top bracket teams who are much (~50 places) lower ranked and which usually end in close (e.g., 1-goal) games, these few games are not enough to significantly impact rankings. It would be interesting to take a look at how accurate the predictions are specifically for cross-league games.
I don't have any specific insight into the algorithm either - but I do know a couple things about it that can be assumed. First - everything to do with the algorithm and assigning a rating to teams has to do with only one, relatively simple, objective. Rate these xx,xxx teams over these xxx,xxx games, predict which team should win, let's keep track. Now look back over the last two weeks of actual game results - and see how many predictions were correct (did a team with a better rating win, or did they instead lose). Now optimize that algorithm so it can predict the highest number of correct predictions. The more accurate it can be in predicting wins successfully, the more accurate it can be perceived to be in rating teams vs. each other. If it could predict 100% of wins correctly, it would have to be assumed that it is 100% correct in rating each team in order. Since of course it can't get to 100%, how high does the threshold need to be for the inferences made from the data to be considered poor, average, pretty good, or shockingly surprising? Predictivity is the measure they calculate to describe how accurate the predictions are. If it is 100% accurate, it is picking every game. If it is 0% accurate, it is picking only 50% of the games, meaning it's no better or worse than a coinflip - it isn't predictive at all. So if it is 50% predictive, that means it is really picking 75% of wins correctly. Here are some stats that they shared with me from a two-week time period last fall: To go from predictivity to % of games predicted successfully, just divide predictivity by 2, and add .50 to it. So for 2018G - it ran at 80.8% predictivity for those 2 weeks, meaning it picked (.808/2)+.5 = 90.4% of games correctly (a higher rated team beat a lower rated team). And it picked only 9.6% of games incorrectly (a lower rated team beat a higher rated team). You can run the same math on any of the results above - the least predictive age/gender over those 2 weeks looks like 2008B, at 66.4% predictivity, meaning it picked 83.2% of games correctly. And that's the worst one of the bunch. There is always going to be criticism in all directions, from some people who are quite sure that their team isn't getting enough "rating benefit" by beating a team by a certain amount of goals. And there are also plenty of people who believe that other teams are getting too much of a "rating benefit", by having their higher score differential count too much. But neither of them are ultimately correct - the goal of the algorithms are always going to be, how can this matrix of results be tweaked, so it results in the highest number of predicted wins. Optimizing how much teams get for winning by X goals compared to Y goals is certainly a part of that. Their team continues to tweak the variables in the algorithm over time, to keep predictivity as high as possible. It shifts during times of year, from pre-season to playoffs, and year over year there are always small differences about how the current game scores are predicting future game scores. I've asked them some questions about the data I've had over the years, and probably the most interesting finding is that when they ran the predictivity numbers comparing games that were played between teams in the same state, and games where the teams were from different states - it turns out that the latter games (different states), were actually slightly *more* predictive by a few tenths of a percent. Teams that saw each other less had more predictable games than teams that saw each other more. More than any other piece of data I've looked at with them, that was one that made it clear that at this point the algorithm is about as optimized as you can get, given the data constraints of knowing nothing more than game scores.
The 2013 Evo team played in a friendlies with SLSG recently. Lost 1-0 to their NL team and tied their top RL team. In my opinion they can compete with the best.
It's a good performance, and certainly validates your opinion. For what it's worth, here's what SR predicts about the matchups: They would be predicted to lose 3-1 to the top team, instead they actually were within a goal. And they would be predicted to win 2-1 to the RL team, yet they actually tied. They are friendlies, which can certainly field any roster they want, and perhaps there isn't as much on the line as on a "real" league game, or even a playoff game. But it's still some reasonably good info that the team can certainly compete. (SLSG is a very large, complicated club - if I picked the wrong teams in that club, let me know and I can correct them. All of the other team choices I was considering had ratings very close, so the analysis wouldn't change much either way.)
Correction. Forgot SLSG has two NL teams. Evo played their Green team, 1-0 L. For what it's worth, their Navy team is excellent, but the best team I saw was the OK Energy FC NL team. I think OK beat the Navy team 2-1 in the friendlies. Fantastic soccer.
They would be expected to lose by a goal to the Green team. Sounds like they lost by a goal. Their rating seems pretty accurate.
Playoff season. Looking at qualifiers on the girls side at top 2 leagues (ECNL, GA). A little easier to do now that all the IL GA’s are in the same conference. Somebody could add in RL or Aspire if they were feeling froggy. I was not. 07/08 Galaxy (defending Natty champs) 09 Eclipse FC United 10 Galaxy 11 Eclipse Galaxy Honorable mention to Inter who finished 5th in conference and just missed out on champions league but made North American cup. 12 FC United (defending natty champs) Inter 13 Eclipse Observations : No reason for a metro like Chicago not to have more teams qualifying. Does make you wonder if there are too many badges. Eclipse with some strong teams but seem to be fading. Inter doing ok, still only 2 years in to NL and imagine they will continue to build. FC United has their 2 solid teams but meh everywhere else. Galaxy on the surface seems to be fading from their high point last year. Likely them entering a tougher conference. I imagine they’ll adjust and be fine. Need a talent infusion with that younger group it seems. Rockford Raptors struggled. Not a team finishing above the bottom half at any level. They need to figure out how to expand to Tri-Cities and NW burbs in a more thoughtful way if they want to remain competitive. Same for Sockers. Mid table for the older groups and then bottom of the table for the younger groups. Just simply more of a resource allocation to the boys side it seems. Maybe the Empire thing helps the player pool. Moving all IL GA’s to the same conference with MI and WI had an impact. Galaxy dominated the west last year. That conference was much stronger top to bottom.
This year there is a bit of an asterisk for the ECNL playoffs as the format and number of qualifiers has changed due to the World Cup. For the Midwest conference, only the top 4 go this year then the next 2 go to the NA Cup whereas in previous years all 6 would qualify for the main playoff. Also the main playoff is single knockout, there is no pool play. If you lose right away, you get 2 more friendlies to reach a 3 game minimum. All of the games are at the same time in July, vs historically 1st round in June and finals in July.
I think Eclipse 2010 team qualified, I think Chicago Inter 2010 RL team qualified and Chicago Inter’s 2010 team qualified for the Cup.
The initial post said if you want to add RL or GA Aspire so I was just throwing that club out there because I remember seeing their post. .
Does anyone have the breakdown of the clubs that the kids the Fire signed, and who? I've seen public postings of the following, but I'm not sure beyond this. It would be interesting to see the pathway of who is actually getting called in and from where. It would also be curious to know which league/division these kids were playing in and how long they were at each club. Mason Riley - Indiana Elite Nolan Pomagier - Chicago Fire Sawyer Walton - Chicago Fire Adam Borzecki - Chicago Fire (2015) Evan Rouches - Chicago Fire Renat Branavitski - Chicago Fire Kamden Davis - Bloomingdale Lightning Giovanni Valle - Bloomingdale Lightning Lee Caddell - Illinois Magic (2015) Anthony Cruz - Illinois Magic Jacob De Leon Jacob Field - Sockers FC Edvin Guva - Galaxy SC Lewis Hernandez Jr Jayden Ledesma - Illinois Magic Ikenna Madueka - Illinois Magic (2015) Elston Ofori - Galaxy SC James Padgett Omar Quinones - Deportivo Leonardo Sanchez Kain Galeana - Illinois Magic Cameron Watson -
Hopefully, someone can help fill in the other gaps. It's interesting to me, looking at this that I know there were 6 Illinois Magic kids and 1 Deportivo. That means that 31% of the kids had no access to the "higher" leagues. I also find it interesting only 1 ECNL team (Indiana Elite), put a kid on there. We did see 2 ECNL RL clubs contribute.
Good observations......... With regular season over I think it is going to be very interesting to see how these teams look next year given what sounds like some heavy transfer activity in certain years AND age group changes. I think some teams stay about the same, some get much stronger and some weaken significantly. Sounds like many of the top Galaxy rising seniors moving over to Eclipse? I assume the FCU 09s stay about the same. Eclipse and Galaxy 10s both losing some high quality youngers to the team below I hear. Eclipse 11s getting even stronger than they already, which is scary. Galaxy 11s should also be improved at 2011 so that age group is stacked. Not sure about the 2012s but I assume mostly status quo with Inter and FCU still being good teams. Rockford 2012s and Galaxy 2012s showed some life towards the end of the season according to rankings app. Speaking of Rockford, the 13s might be the surprise of the area in that their current 13s were very competitive in games against their in state rivals and are adding some younger 12s that are good players. No clue about the 2014s next season as I am sure all the badges are consolidating up the community clubs top players as we speak! Will take the ranking app the entire fall season to catch up more than likely. Did I miss anyone?
I was surprised to see so many Galaxy players moving over to Eclipse. Not really sure what to make of it. Are kids just looking for something different in their senior year? Do they think playing ECNL their last year will help them keep their scholarships? It seems like colleges are pulling scholarships before girls even set foot on campus, which is just crazy…but that’s the state of the game. All the Eclipse teams will benefit from the age group changes, but it’ll vary. Overall, success will depend on how well the coaches integrate and use the players, which is easier said than done. I’m really curious to see how college recruiting plays out with June 15th coming up. I expect Eclipse to do pretty well. Do the other clubs with badges make consistent improvements, or is it still just their one or two standout players, with the balance being late cycle auditions by programs? It’s only going to get more competitive for everyone with fewer spots available (transfers, roster cap, 5th year..).
In the abstract that might be true. In this case, having an RL team ranked 96th in the nation tells you quite a lot about that particular team.
It seems much crazier that most kids would get an offer before their final year. But that's been the state of the game for too long. I'm glad to see that change.