Tiering + Fluidity = Retention and Development

Twitter can be a fantastic forum for conversation around a topic, the immediate exchange of ideas and the presentation of media for others to consider. I use it heavily and have benefitted from it considerably in terms of being introduced to people and their ideas around youth soccer development.

The downside is the ‘silicon tower’ effect that is long on, often vaguely supported, theory and short on practicality and overall reality. Take the idea of “talent selection vs talent identification” raised in an article by John O’Sullivan who runs a site called Changing the Game Project.

It became a source of Twitter interest last night amongst several of us who wring our hands over the state of youth soccer here in Vancouver.

Boiled down the article chastens, rightly, the win at all costs approach in teams sports before high school and warns of the perils of selecting players based on current abilities rather than potential future abilities. O’Sullivan advocates for non-tiered playing environments that keep a larger pool of players engaged for longer periods of time and not over-training.

First thing to note is that O’Sullivan is American and coming at this from an American youth sport point of view which is a somewhat, though not entirely, different perspective from those of us in Canada. What is true is that the club structure in the States is generally delineated into elite clubs and recreational clubs with elite clubs being much smaller and much more expensive while recreational clubs generally have lower costs but fewer resources in terms of professional coach or technical director support.

In Canada, generally, soccer clubs offer all, or most levels of play available within each age group. The number of levels of play averages around eight on the boys side and six on the girls side in Vancouver.

My reading of the article gleaned these assumptions being put forward by Mr. O’Sullivan:


  • Win at all costs youth culture is the status quo
  • Kids quit or move to a different club if they don’t make the top level
  • Kids all aspire to the highest level of play
  • Kids are content playing with players who are either far better or far worse at an activity than they are
  • Kids will develop just as well in a non-tiered environment up until U14 than they will in a tiered environment
  • Talent identification is an art

The article interests me primarily from the angle of tiering vs non-tiering and how that affects development and player retention.

First off, the environment here is becoming more progressive, perhaps not in a linear fashion and perhaps not across all clubs and coaches but its moving away from early ‘success’ and towards valuing coaches who have evolved past this and are committed to longer term development. We also see this in leagues not keeping standings or having Cup competitions before U13. We’re moving in the right direction and sometimes need to remind ourselves of that.

Looking at player retention, the statistic stating 70% of kids stop playing organized sport by age 13 is shocking. It’s also un-supported by citation in the article and I’d imagine getting that number like that measured across all sports in an entire country would be a massive research project. I’m not denying it’s true but I’ve yet to see a reference to the research that supports it.

My experience is that if a club provides multiple levels of play and can offer a level that is suited to a players current technical and tactical level of play that also fits with their commitment level, you will likely retain that player. If you just tell them they didn’t make the one team you can offer and cut them loose to go find another club, yes, you are more likely to see that player quit the sport due to a lack of support and a lack of ability for them to continue playing in their community. I find having up to eight levels of play at our disposal when we form teams to be a highly beneficial tool.

If kids were all as focused on playing at the highest level it would make sense that participation in lower levels of play would be minimal but a casual look at league tables indicates that there are many, many kids content to play at levels that represent the bottom third of what is available in their age group. Kids don’t quit because they didn’t make a particular team. Kids quit because adults don’t facilitate their continued play on terms that make sense for them.

Perhaps most contentious for me is O’Sullivan’s assertion that kids should not be tiered before high school age and there should simply be larger pools of players playing together. This is where theory seems to like the sound of its own voice too much and conveniently ignores practicality, logistics and reality. It should be noted that O’Sullivan does not say put players of a similar ability together for training and games and allow quick and easy movement between these groups to ensure players are always suitably challenged. Beyond that, there’s no detail on how to organize these teams, if there’s leagues that make sense to put them in and the difficulties of coaching to a wide range of ability and motivation.

I have worked with youth soccer players in Vancouver since 1997 and professionally since 2000. I have a very clear idea of the range in ability at each age group and at what point that range starts becomes too wide and begins to warp how the game should look. The assumption that O’Sullivan’s model is an enjoyable experience for kids and that it doesn’t hinder their development is entirely wrong in my experience.

This expectation we have that we should just let kids play without guidance from knowledgeable coaches and that somehow they will learn from their mistakes is wishful thinking. Just because we can pull quotes from a handful of exceptional people who managed to do this, or at least believe they managed to do this, does not prove that its a sustainable model for the majority. The reality is that the model for elite development is defined by top European and South American clubs and they provide excellent coaching in a highly structured environment for the best players they can find. Yes, there is a definitely an element of talent identification involved and often patience in waiting for some players to physically develop but there is not a wide gulf in ability in their training groups and the amount they train is higher not lower than the average player who likely trains twice per week.

The situations you see in a U11-13 bronze game are monumentally different than those you see in a U11-13 gold. Everything from time on the ball, quality of first touch, defensive shape, attacking shape, everything is massively different. So how does it make sense to expect that U11 bronze player to enjoy and thrive in an environment with U11 gold players? Why would we expect them to be able to combine with gold level players at this stage in their development? Why would we expect that they feel they are contributing to their team and develop a love for the game when they are clearly in over their heads and increasingly marginalized by stronger, naturally competitive teammates who don’t involve them in play? Conversely, how does it benefit the stronger player who is being told to work co-operatively on the field and move the ball around the field to teammates when it breaks down as soon as the ball goes to certain teammates who haven’t developed the necessary technical skills yet? How many times will they continue to look up and pass to a teammate when they know it will, 9 times out of ten, lead to a dispossession? This is definitely a source of frustration for kids at both the higher end of the spectrum and the lower end and it does nothing to help either develop.

And I’m afraid it’s far from convincing to use, as O’Sullivan does, a non-contact, individual sport like tennis to support an argument about contact team sports like soccer. But even if we do use tennis as a discussion point, would it work to have a ten year old who’s been playing tennis since she was five and receiving excellent, professional coaching from that age, rally and play against another girl who just picked up a racket six months ago and is still learning the basics even if it’s also with the same excellent coaching? One gets bored, one gets embarrassed at being put in that environment by adults who must surely know it does neither an ounce of good in terms of their development as players nor their engagement in sport in general.

We have to stop pretending that tiered environments are wrong and elitist and accept that the solution for all players begins with an environment that feels safe and nurturing in terms of parent support, adequate coaching and being able to play with and against peers of a similar ability. That means creating multiple playing environment to meet those needs rather than jamming them all into one environment.

What would make our current system better is to build in more fluidity between levels so that players aren’t locked in for a year before they can move. I’ve been an advocate for this for some time. And once that tool is made available, over use it early on to normalize it so that young players aren’t overly-excited or overly-nervous about being moved up a level or overly-stigmatize by moving down a level.

Do this and the whole idea of talent selection vs talent identification moves away from being a systemic issue rooted in a ‘desire to win at all costs’ to ensuring sufficient resources are put into coach education so that correct decisions about talent identification are actually being made. This then helps us move them up and down through levels of play as they acquire skill and knowledge of the game.  This benefits both their enjoyment of the game and their development.

To finish, I’ll open another can of worms. As alluded to above, moving from ‘talent selection’ to ‘talent identification’ is not as simple as just changing gears. O’Sullivan says talent identification is an art form. Another way to say it is that it’s a very, very difficult art to guess what an 11 year old will look like as a player when they are 14. The degree of acumen needed is not just a function of coach education but a long track record working with players at these ages and levels. We already have a dearth of coaches truly capable of working with young players to develop them effectively. I’d suggest the number of coaches capable of accurate ‘talent identification’ is even less.

This entry was posted in Coaching, Uncategorized and tagged . Bookmark the permalink.

12 Responses to Tiering + Fluidity = Retention and Development

  1. Imccarth says:

    Another nice piece. Just tweeted it.

     I would ‘like’ this and other past posting but I give up when WordPress asks me to login. 

    Sent from mobile device please excuse brevity and typos 

  2. daffydyl says:

    I’ve read the article numerous times now and i would be a little hesitant to suggest that simply because tiering wasn’t mentioned that he must not support it. I think we have to look at some of the realities out there in our footy landscape and accept that many selection processes start far too young and are predicated upon picking the most physically able at that moment in time. That usually ends up being the most chronologically/biologically mature kids. Why? Results man. And the sad truth is that, as most will acknowledge, we have a dearth of sophisticated coaches at top youth levels, let alone the layers below catching the non-selected. So here comes the self-fulfilling prophecy when we take the oldest and most mature, give them more training of a higher quality and the fall off is huge. Kids become demotivated in many instances because they feel no sense of improving or see no path to catch up. Of course some will always choose rec level, I think that’s missing the larger issue though. I think if given the opportunity, Mr O’Sullivan just might take tge opportunity to agree about tiering. Question become how we do it.

    • Gregor says:

      The article, for me, is short on details and a bit sensational. I looked into the Sweden hockey bit he referenced. Doesn’t bear scrutiny too well based on quotes from their head of hockey development in this article: http://www.nhl.com/ice/news.htm?id=598251

      The follow up article about the Danish player overlooked by UEFA A coaches for a spot in an Academy who then went on to play for Denmark in the 2010 World Cup is also not exactly a ringing endorsement for talent identification. As I said, it’s a very difficult thing to do (pick players now who you think will blossom in 3-4 years) and if a progressive, development oriented country like Denmark miss a player like this despite having several UEFA A license coaches evaluate him…is it a realistic model to go for in Canada?

      • daffydyl says:

        I’m not here to defend every element cited in the piece, simply to point out many over-arching points are fairly spot on. Just as I agree with many of your points, I do find it a bit disingenuous to suggest he must be against tiering because he didn’t mention it. I’ve seen him comment on the book Top Dogs where he said something to the effect that he used to think grouping players of similar abilities wasn’t the best way to go until he read the book. I think the bottom line is all people involved in grassroots sports ned to take a long hard look at how and why we do things.

      • Gregor says:

        You O’Sullivan apologists are all the same 😉

        I think if you’re going to say there’s a problem in the current tiering model due to biases toward bigger, stronger players and then say keep larger groups together for a longer period of time (ie until high school) you either have to specify what that model looks like or leave people to assume you’re talking about a continuation of non-tiered play.

        If he does favour tiering what criteria is being used to pick, say a U12 team. Do you bypass the players that he acknowledges have, “the current ability to participate and be successful” to pick players that may be better players in three years? Do you pick both? I don’t know how such a model should look and O’Sullivan is short on details. As I alluded to, it’s a lot easier to solve youth soccer problems on screens and paper but at some point it needs to be implemented in real life.

      • daffydyl says:

        I’ll be happy to answer to some of those questions on the 23rd 😉 Merry Christmas!

  3. Red Hot Coach says:

    Excellent blog!

    -Describes the realities of different talent levels and the need for tiering to make sure that all players (strong and developing) are challenged at their appropriate levels and abilities. Great tennis analogy to illustrate why tiering is needed.
    -debunks the ‘theory’ that kids are leaving the sport (every sport) in droves because they aren’t playing in the top tiers; I’ve seen no research to that either.
    -addresses the ‘wishful thinking’ that kids learn and become adept at a sport (any sport) by just having coaches stand there with their hands behind their backs watching the kids practice or play.

    I also agree with your suggestion that a better system of fluidity/movement between levels – during each playing season – is needed for coaches to be able to move players who are developing as the season progresses.

    One thing you didn’t touch on is the responsibility of parents to explain to their children that they have to work hard to get to the top tier and work hard to stay. Too many articles on Twitter – like Mr. Sullivan’s – suggesting that clubs should do away with tiering,and lowering the playing standards so that players will not have to work at becoming better and moving up. Parents also have a big responsibility to be realistic with their little athlete so that he or she is realistic about their goals and abilities.

    Again, well done. I can refer your article on twitter when this topic comes up there again.

    Next Blog topic – the poor defensive attitudes demonstrated professionally on EPL (and elsewhere) because of reliance on the no responsibility, “he’s not my man”, flat four defense that is prevalent today.

  4. NW says:

    Great article and I agree with tiering with the caveat that it is done for development . This requires recognition of skills and potential,(not just current form) and, as you suggest , fluidity between levels. The way our upper tiers are currently done is to pick the maximum allotted number of players with greater emphasis on the bigger and stronger players . This allows limited ability for fluidity between levels and hence fewer players appropriately developing.
    I feel tiering allows players to get more touches and develop more when matched with the appropriate level of skill . The problem is getting enough great coaches at all the levels to allow development and not a drop off in skills when movement occurs between tiers

  5. Gregor,

    Thanks for the chance to reply. My blog gets ping backs when the articles are mentioned or linked to, and while I usually ignore many of them, with a site called Monday AM Centerback I had to look. When I read your well thought out article,and the comments of your readers, I wanted to reach out and respond. This is my type of discussion!

    First, the 70% of kids drop out stat is well documented, you can find sources either through the National Alliance of Youth Sports, and Michigan State University Institute for the Study of Youth Sports. The hockey stat I got from a man named VJ Stanley who is the author of a book Stop the Tsunami in Youth Sports, I did not verify his source though.

    As one of your commentators said, I did not really discuss tiering in this article, and I will below. The main point I was trying to make was in the US, our elite clubs (the ones with the best coaches) all too often only keep small numbers of players at pre-pubescent ages. They cut far too many kids whom they have no way of predicting whether they will be good players or not. They are then removed from the top coaching and proper player development cycles, do not have the chance to move up and down amongst teams, and usually leave the game (or sport in general). The emphasis on winning usually leads to the selection of bigger, stronger, faster and older kids, and not necessarily the ones who bloom late, have a lot of technique but are short on athleticism, etc. We shrink the player pool far too early. Thus, I argue that we select current talent instead of keep enough players around to truly select talent when it counts, at a much older age. And because it is nearly impossible to actually identify future talent, we better keep a lot more players around to select from once they grow.

    As for tiering, I am 100% behind it, and in my clubs that I ran I often fought many battles to make sure it happened. If you have 50 kids in an age group, player 1 and player 50 will not benefit from competing against each other on a daily basis. As you say, the gap in talent, motivation and athleticism is far too big to have much of a benefit. I believe in breaking players up by ability groups, exposing them to the same exact training and top level coaching, and the fluid movement of players between groups. We have all seen top players go through awkward periods where they struggle, and weaker players once exposed to proper training and a little growth blossom and become top performers. So I totally agree with you, tier all you want, just be sure to offer proper training and monitoring of the kids who are initially put in the lower levels. In my experience, in this type of environment, the kids who start out weaker make faster and far more significant improvement, thus strengthening the entire pool. For further reading, I suggest you Google Tom Byer, an American coach who has spent 20 years revolutionizing soccer in Asia, specifically Japan. I am a big fan of his methods and philosophy.

    I do recommend you read Top Dog, they discuss the science behind the need for tiering of students whom are not too far apart in ability at the US Air Force Academy. I see no reason why sports would be any different. All performers in a given group should be able to be the best on a given day, and the worst on a given day. When they are never too far from the top or bottom, they improve significantly. The study shows that when players are not tiered, the top are not pushed because they will never be the worst, and the bottom give up because the top is so far off.

    Thanks again for the invite to be a part of the discussion Gregor, I hope you will continue to read my articles with the Changing the Game Project, and that they promote such lively discussion and great articles int he future. Happy Holidays everyone!

    John O’Sullivan

    • Gregor says:

      Hi John,

      Sorry to take so long to continue the discussion. Holidays etc..

      Thanks for clarifying some of your points I’d referenced. When I started the post I didn’t really want to get into questioning each stat but unfortunately to some extent did. For me this is the crux of it, quoting from your reply above

      “They cut far too many kids whom they have no way of predicting whether they will be good players or not.”

      As I said it is a bit different here in Canada compared to the more clear division of “elite travel clubs” vs “recreational clubs” that exist in the States but the reality in both is that there are multiple levels of play and at least a perception that the better coaching exists at the higher levels.

      To go back to your quote, if there’s no way coaches can predict future performance, and there is the reality that they can, by league rules, only carry a certain maximum number of players on their roster, then how can there be much space for players who may show they are capable of playing at that level in a year or two? The key for me, given that, is to change the environment so you’re not restricted to just one elite team. Sometimes that’s not possible but most of the time it can be done with co-operation and some leadership in defining the youth soccer culture wherever you live and play.

      I realize that the focus of your argument is players being select based on physical traits rather than more soccer specific attributes and I couldn’t agree more. I run evaluations each spring for all our U11-14 players (about 1100 kids each year) and over the past decade have developed a good idea for what level of ability is necessary to compete and enjoy the level we are selecting for. We view these years (and I’ve been fortunate to have worked with a generally very progressive Boards that shares this view) as key developmental years. Factor in a generally progressive governing body and league structure that now doesn’t keep standings or have Cup play until U13, and I have the ability to pick teams that really just need to be able to compete. So I have the latitude to go with a second or third team in a division when I feel its warranted rather than just, say, picking the top 12 U11 players in the age group and knowing they will likely win almost all their games. By diluting the talent pool a bit, we give those kids a chance at the younger tiered ages (U11 and U12) to play with and against the better players in their age group without the pressure of winning. We create roster spots so that we can carry both the very athletic kid who wants to be a player but needs to refine touch and vision as well as his/her opposite: the smaller, slower technical player who sees the field well and makes good decisions.

      Clubs need to be cognizant of how quickly kids can improve given the right environment but also how some kids lose interest because of other things that catch their interest and fight for environments that keep as many players engaged as possible. Here’s some ways to do that:

      1. Develop a community amongst other progressive clubs to establish that development environments (no standings, no Cup, equitable playing time policies) rather than competitive environments should be the norm at U11, U12 and, for me, U13 (which is the age here that we move to full size 11v11 games). Pressure governing bodies collectively if necessary to ensure rules and formats reflect this.

      2. Form teams at these age groups with the overriding goal that the teams should be able to compete and will likely be .500. I communicate with other surrounding clubs that, like us, are big enough to have the ability to field multiple teams at the top tiers of the leagues we play in. We get on the same page and try to establish divisions where our teams will be suitably challenged.

      3. Manage parent expectations by communicating often and in various ways so they are not going into a season overly concerned about the win-loss record. I have a meeting with parents who are heading into their first year of tiered soccer and usually open with something like “We really don’t care about winning games at U11 gold. We want our coaches to teach your kids how the game needs to be played so they will be successful at older ages when it does start to matter if you’re winning or not.” This is in other written info that goes out to parents as well.

      4. That said, the counterpoint to this is that you can only put some much air in a balloon before it bursts. By that I mean if you dilute teams too much you cannot expect them to (a) play effectively as a unit due to the increased differences in ability, (b) expect them to compete reasonably in the division you have affiliated at. There has to be a degree of common sense and the knowledge of leagues to know when you are forming one too many teams for the level of play. Kids also lose engagement and get frustrated when adults put them with teammates who are not of a similar ability and in divisions where they are unable to compete.

      While I would agree with you that we should more and more kids to ‘top level’ coaching the reality is that there are more players needing and deserving of this coaching than we currently have in most North American cities.

      For now, I think the focus is on respecting young players by providing sensible learning and playing environments that focus on their continuing development rather than winning (U11-U13). If we do that, we retain a higher percentage of them and have a larger pool as they hit high school and puberty.

      By engaging and, hence, retaining we can then increase training loads to further their development as they are enjoying the experience. By U14 we are ready to introduce the idea of training and playing to win. I’ve written about this once or twice previously here.

      Thanks for taking the time to reply and I think it’s safe to say we agree on at least 80% of what’s been covered. I’ll eventually get around to reading Top Dog and I still regret not going to see Tom Byers presentation at the NSCAA conference in Kansas City two years ago.

  6. Gregor,

    Suffice it to say we are on the same page here. Thanks for your response. You are right that in Canada, you have many more teams under your umbrella, while in the US clubs are fond of saying “we are special so we only have one team” and get rid of the rest. I agree with your four points as well. I worked for a very large organization called Rush Soccer for 6 years, are stated goals was to be 6-3-1 record wise. More wins and we were not challenged enough; all losses and it was discouraging. Too many coaches find places to play where they can win all their games and say look at me.

    I also agree that we just do not have enough top coaches, which is why I think better curriculum’s and an emphasis on touches on the ball are imperative at the young ages. Too many average coaches spend their time trying to teach a kid positions, or teach tactics, set pieces, etc instead of comfort on the ball. I see too many 11 year old teams playing a zonal back 4 with a high line when the kids have no clue when to step, when to drop, how to position their body, what side of the man to be on, etc. I hear the coach screaming instructions from the side, and half the time he/she is wrong. We need to invest more in helping coaches understand exactly what young players need, and spoon feed it to them so that even average coaches are disseminating good information.

    I wish you luck with your blog Gregor, and if you are in Philly for the NSCAA Convention this year please come by my Changing the Game Project booth. Thanks again.

Leave a Reply to John O'Sullivan Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s