In Part 1 of our Roster Assignment series, I highlighted how a fairly flexible system of roster assignments was great for making soccer easy on parents schedules, letting kids play with friends, but also allowed the gradual creation of farm systems which over time became powerhouses. In this post I’ll cover all the options the officers came up with during a recent brainstorming session and also look to you all for ideas that you know to work elsewhere. Got that red pill? OK then, here we go…
It should be noted that all these options assume age divisions that are two years in size (U8, U10, U12, etc) and younger players get to stay with their team the next year. Only players aging up or who are new to the league are ‘assigned’ via the options below. The league felt it would be too disruptive to break up the teams every year and would present major difficulties to the sponsors.
Option 1 – Leave It Alone
Our current system allows parents to make special requests to play for a specific coach, play with certain friends, play with people they can carpool with, or to avoid being assigned to a certain team (i.e. a coach they didn’t get along with in a previous season). Division coordinators assign the rosters using the following criteria in priority order [taken from the MYSA Team Assignment Guidelines]:
- Coaches children should be on their parents team IF the coach wants them there. Most do, but every once in a while you’ll have a coach who would rather their child be coached by someone else to avoid distractions, etc.
Kids returning to the same age division should be returned to their original teams unless:
- Their old team has dissolved due to a coach leaving, most of the team moving up to the next division, etc.
- The parents request a new coach (this happens – and can be a sensitive subject. You will inevitably have to answer questions from the original coach as to why a specific player did not come back to their team.)
- Roster size. Teams should have the same # of kids. This is very important. Players drop out, etc. So starting with equally sized teams will go a long way to reducing problems down the road with dropouts. You should never have teams +/- more than 1 player after making assignments. After dropouts it may be worse, but you can fill those holes with waiting list kids (more below).
- Attempt to achieve gender parity on your teams. An ideal goal (sometimes hard to achieve in smaller divisions like U10 and U12) is to have teams come within a difference of 1 or 2 gender wise. Note that the All Rosters screen in SoccerSys [ed. this is our league mgmt system] will show you both the gender breakdown and average age of each team which can help during assignments. A good idea is to open the All Rosters screen in one browser window while having the Assign Players window open in another as you shift kids around.
- Attempt to achieve age parity. There are two aspects to age parity. Overall and gender based. A team with overall age parity compared to other league teams that has all older boys and younger girls, may be at a competitive advantage compared to other teams which have age parity across both the entire team and each gender. Obviously this isn’t guaranteed – the younger girls may have better skills than all the older boys! But again – team assignments are a best effort. Since it is tough to know the skill level of every player you can only go based on what data you have (age, gender and experience)
- Special Requests. Yes, special requests come last. In reality you probably can meet over 90% of your special requests. But in trying to make EVERYbody happy, you will likely have trouble. A single special request can cascade and wreck half your teams. You will find requests which conflict with each other. So be prepared to tell parents NO sometimes. While the MYSA has some hard and fast rules (no age ups, maximum roster size, etc.) the rest is left up to the discretion of the division coordinators. So you have to remember that your job is to create competitive divisions and teams that ensure a challenging environment for the players and don’t discourage some teams at the expense of others. Your job is not to make every single parent happy – it simply can’t be done. But be prepared to discuss with some parents WHY you couldn’t meet a special request (it would have created 3 super teams that would dominate, or we can’t have that many children on a team, and be prepared to use "if we do that just for you, everyone will want us to and it’ll be chaos") You’ll also find dealing with special requests, that some tend to outweigh others. A sibling playing together request probably outweighs a ‘I want to play with my friend’ request.
One thing not covered here is the concept of ‘cores’. We generally allow cores of kids to stay with a coach moving up and sometimes they merge with an existing core of older kids. If a coach moves up to an older division, but leaves behind a sizable core, we’ll often try to find them a new coach rather than break them up.
Pros: Most of the parents of younger children like the flexibility of having their kids play with friends, people they can carpool with, or even just someone they happen to know from their neighborhood. Doing this definitely gives our league a social aspect it otherwise might not have. It likely increases our player pool since more kids who can carpool are able to make practices their parents might not be able to get them to reliably. It fosters the creation of farm systems of teams across most age divisions where kids funnel through a set system of teams and coaches providing continuity. Farm system coaches tend to be some of the most experienced as well.
Cons: Farm systems will generally lose less skilled players who decide soccer isn’t their thing and they will be replaced by stronger players whose parents specifically asked for teams with experienced coaches. New coaches tend to have teams filled entirely at random from the overall pool while established coaches often get requested by parents. Parents often feel coaches are trying to ‘stack’ teams by telling specific parents to ask for their team though this has never been observed to date. There is an overall perception that certain teams are trying to rig the system, especially when a core from a top team in one division ages up and merges with a core of returning players who also finished near the top. Parents who do not get special requests, even when told repeatedly there are no assurances they will, often get very upset. Division coordinators who reject certain special requests to keep a team from having all top players will often face immense pressure from parents.
Option 2 – Assign Randomly
Allow coaches and sponsors to keep their own children on their team. Returning players within an age group can stay on their old team. All other slots are filled at random by drawing, etc. No special requests are allowed. Players likely would be split into groups by age and gender and assigned to open slots at random (i.e. if a team needed two 9 year old boys to meet the overall age and gender targets, then two would randomly be selected from the 9 year old boy pool)
Pros: Impossible to stack a team. Teams overall would probably be more balanced than with Option 1.
Cons: Makes it difficult to carpool with friends and neighbors. Power teams could still be created by chance. No assurance the teams will be balanced since age and gender parity is inexact at best, especially as the players get older.
Option 3 – Random Plus
Teams would still be filled at random as outlined in Option 2, however the age/gender pools would be sorted by estimated skill level. The sorting would be done using available data in the system including previous records of teams a given player was on, the number of years they have played soccer, have they tried out for Challenge teams, or possibly even a ‘skill score’ assigned by previous coaches.
Pros: Same as Option 2, except the teams probably would be more balanced than under Option 2.
Cons: Requires investment in software to sort the pools. Available data is subjective and may not accurately represent a players relative skill level.
Option 4 – Draft
Allow coaches and sponsors to keep their own children on their team, or possibly allow each team X exemptions/special requests. Returning players within an age group can stay on their old team. All other slots are filled via a draft where the division coaches select from the available pool of players in a round robin format. The order that teams choose could be set at random or based on previous records.
Pros: By involving the coaches who can judge skill level, the round robin format likely would spread the talent evenly across the teams. Difficult to stack a team unless many other teams make mistakes in drafting. Best players would go to teams that need a boost.
Cons: Requires significant time investment by coaches to ‘scout’ upcoming talent from younger divisions. No way to know the skill level of new players who join the league. New coaches are at a serious disadvantage since they do not know the players. Difficult for coaches to know the talent pool in larger leagues. Encourages parents to lobby coaches to pick their kids if they want to be on a given team.
Option 5 – Limited Exemptions
For options 2, 3, and 4 – allow coaches to ‘protect’ a certain number of players beyond their own child when moving a core up to an older age division. The remaining players from the core would go into the assignment pool. The rest of the open slots would be assigned as outlined in the above options.
Pros: Allows assistant coaches to keep their children on a team while ensuring more random players get assigned.
Cons: Reduces the number of open slots each Fall to 2 or 3 which can make it difficult to balance a team due to less flexibility.
Option 6 – Limited Special Requests
For options 2, 3, and 4 – allow coaches to accept a set number of special requests. The number of requests could be a set number (2, 3, etc) or set in proportion to the number of open slots available on their team. Any further roster slots would be filled as outlined in the above options.
Pros: Allows parents to have a chance at getting a special request for carpooling, getting an experienced coach for their skilled player, etc. while still providing for some random assignments. Prevents farm systems from locking out players who don’t specifically request them.
Cons: So few slots would be assigned at random, little parity benefit would likely be seen. Certain teams would improve due to even a limited number of special requests while other teams would still get all their players from the pool.
Option 7 – Skill Rating
Assign skill ratings to players. This rating could be assigned by previous coaches or via a day-long assessment similar to a tryout where players are rated based on drills and match play. These ratings would be used to sort the assignment pools and the slots would be filled round robin from the sorted pools.
Pros: The teams would likely be well balanced since this type of ranking would be more accurate than the sort methods described in Option 3. Allows coordinators to have a feel for the skill level of players new to the league (if skill assessment is used)
Cons: Very labor intensive. Overkill and inappropriate for Recreational level. Difficult to get everyone to an assessment at the same time. Ratings would be relative since it would require multiple assessors.
Option 8 – Handicap Top Teams
Teams who finished the previous season in the top X% of the standings would not be allowed to take kids via special request. All of their assignments would be done at random. The teams lower in the standings would be allowed to take special requests and then fill the rest of their slots at random.
Pros: Can help prevent ‘dynasty’ teams from winning year after year since they would get random assignments each year which could balance out any stars they may have.
Cons: Doubtful parents would request that their kids play on the teams lower in the standings so overall it would likely end up being all random assignments. Previous season standings aren’t a good indicator of how a core (either the core staying back or the core moving up) will do since the team is being split.
Scary isn’t it? We came up with these during a brief brainstorming session. Obviously some things aren’t even something you’d consider at the Rec level, but again, it was brainstorming so nothing was too outlandish. None of the options is perfect – they all have downsides. The implications of some of these options would be significant.
Random assignments are very attractive because you can always tell parents who complain that a team was ‘stacked’ that it was random chance, nothing else. But that’s a cop out. Nobody said running a soccer league was easy. A draft has similar appeal. However the non-random assignment methods will always be subject to parents who see favoritism.
So which option should we consider? I’m sure many of you have other ideas on how to handle it and I’d love to hear them. In my next post I’ll discuss what our board decided to do for the Fall season.