Why Your Soccer Practice Feels Like Chaos (And How to Fix It)

Give a group of eight-to-ten-year-olds five coaching points before they practice, and some of them will get better. Others will get worse. Researchers ran more or less exactly that with 90 children, and the difference came down to how much each child could hold in their head at once. Any team you coach has both kinds of player in it.

That's a big part of why, when a practice feels chaotic, the environment is the first place I look, before the coaching.

Young players who can't sit still for five minutes of instructions will chase a ball for an hour without anyone telling them to, then fall apart the moment you try to organize what they're doing. A lot of that comes down to what they've been given to learn from.

I've been coaching for over 15 years, and I still catch myself asking a player for one or two things more than the session needs, more than they can hold in their head at that age. It's an easy habit even when you know better, because in the moment, saying one more thing always feels like coaching.

Two ways to get a good decision

There are two broad ways to get a player to make a good decision. You can tell them what to do: more instructions, more stopping the game, more of your voice competing for their attention. Or you can build a practice where the right decision is the obvious one, because of how you've set up the space, the task, the equipment and the players.

The second way asks more of you before training and less of you during it. It also asks less of the player's memory, which matters more the younger they are. Most sessions need some of both, and later in this piece I get into where the first one earns its place. But the second is the one most of us under-use.

I didn't come up with this on a car journey home. It lines up with a researched approach to practice design called the constraints-led approach, built on decades of ecological dynamics work (the idea that players learn by interacting with their environment, not only by being told what to do) from sport scientists like Keith Davids and colleagues. The FA teaches a version of it through Ben Bartlett's "restrict, reward, relate" framing. Different name, same idea: design the game rather than narrate it. (That research treats perceiving and acting as one coupled process rather than separate steps. It matters later: the scanning piece picks it up.)

Here's an example built for U8–U10 players. Look at where the goals sit on the grid below, and what the scoring rule asks of a player the moment their team wins the ball back.

View full size Diagrams like this are built in PitchLabs →

The STEP framework

Make the area bigger or smaller. Change the rules or how players score. Swap the goals or balls. Change the number of players on each team. The FA teaches a version of this today, so it isn't mine and it isn't new. I keep coming back to it because it turns "this session isn't working" into four specific questions instead of one vague worry.

• Space. Can I change the area to create more time, pressure, or movement? Squeeze a 3v3 from 25x20 down to 25x15 and you get faster decisions and more pressure on the first touch. Stretch it to 30x25 and you reward the player brave enough to take their first touch forward instead of sideways.
• Task. Can I simplify it, add one condition, or change how players score? Not four conditions stacked on each other, but one constraint that pushes the specific decision you want to see. Make a goal worth double when it's scored on the opposite side to where you won the ball, and the team in possession has to scan for space and a wide option before they commit, instead of just driving at the nearest goal. An old mentor of mine put it simply: chase five rabbits and you catch none. Chase one and you've at least got a chance.
• Equipment. Can I use goals, gates, zones, or different balls to shape the behavior, instead of shaping it with my voice? Two small goals at each end instead of one big one give a switch-of-play rule somewhere to go. The reward for scanning has to be real, not just a coaching point shouted from the sideline.
• Players. Can I adjust numbers, add a neutral, or create an overload? Overloads buy space and time for the team in numbers-up. Underloads force quicker decisions out of the team defending.

Get these four right and you need your voice a lot less. Get them wrong and you can end up narrating over a practice that's working against you.

When a session isn't working

This is how I work through it. It starts with what's actually going wrong.

Too easy? Players are succeeding without effort. Try adjusting Players before you shrink the game for everyone. Often it's only part of the group finding it easy.

Too difficult? Constant breakdowns switch young players off fast. Simplify the Task first, then rebuild the challenge gradually.

Too slow? That's rarely lazy players. Usually it's too much standing around, and adjusting Players is the fastest fix: smaller teams, more games running, nobody waiting for a turn.

Right game, wrong behavior? If players aren't doing the thing you designed the session for, ask whether the game actually requires that behavior to win. If they can succeed without it, they will. That's game intelligence, not defiance. Reward the behavior you want in the scoring rather than banning the workaround.

That four-way split is how I use STEP when a session goes sideways. It's a practical checklist, not a research finding, but two of the four have research pointing the same way. The too-easy and too-difficult ends match what motor learning researchers call the challenge point framework, which proposes that learning stalls at either extreme and peaks somewhere in the middle (Guadagnoli & Lee, Journal of Motor Behavior, 2004). It's a theory rather than a measured result, but it's a well-used one. The too-slow fix has direct support with young players: in one study of ten-year-olds, switching from 7v7 to 4v4 doubled the actions per player per minute, and the least involved players gained the most (Hintermann et al., PLOS ONE, 2021). Fewer players per team means more of the ball for each of them, and standing in a line kills a session faster the younger the players are.

Why more coaching points can mean less learning

Back to the study I opened with. A young player who can chase a ball for an hour but can't take in a long explanation isn't being difficult. A lot of it is capacity.

Researchers took 90 children aged eight to ten, measured their working memory, and had them practice basketball shooting across 240 shots, with five explicit instructions about technique before every block. The children with higher working memory improved steadily. The children with lower working memory got worse (Buszard et al., Frontiers in Psychology, 2017).

The same five instructions helped one group and hurt the other, and any team you coach will have both kinds of player in it. Five was just the number that study tested, not a discovered ceiling, and it was instructions given before a shooting block rather than coaching points during a live game. What carries over is that a countable number of instructions was enough to make some children measurably worse at a skill they were practicing 240 times.

Remember the five rabbits from earlier? This is them. Stack four or five coaching points on a busy activity, and for the players in that study with less working memory to spare, the result wasn't four or five things learned. They got worse at the one thing they were practicing.

A packed, multi-part session looks like good coaching. Lots happening, lots planned, a plan you could hand your DOC without embarrassment. A simpler check I'd use instead: could a player tell you what they were working on after today's session? If not, too much was probably competing for their attention, however well it read on paper.

What a game can't teach on its own

Everything above assumes something that isn't always true: that players can already do enough to take part in the game you've built. A switch-of-play rule only teaches switching play if a player can get the ball across the pitch. If they can't yet pass that far, or stop the ball well enough to look up, the decision you designed never happens. The game just breaks down, and no amount of adjusting Space will fix it.

There's a well-known argument from education research that backs this up. Beginners, it says, learn better from explicit guidance than from being left to discover things, and the advantage of guidance only fades once learners know enough to guide themselves (Kirschner, Sweller & Clark, Educational Psychologist, 2006). That's classroom learning, not sport, but the logic travels. Young players are beginners at a lot of what we ask them to do.

The sport research is more encouraging for game-based practice than that paper would suggest. In the studies that compared game-based and drill-based practice head to head, none of the technical outcomes favored drills, and games came out ahead on tactical skills (Bromilow et al., Sports Medicine – Open, 2025). But it's nine studies, mostly small, with few young children, and several of the game-based programs used task simplification and focus cues themselves, so the comparison was never "instruction versus none."

In practice, that's why I'd hold this piece's argument loosely at the edges. When a game keeps breaking down because players are choosing badly, that's a STEP problem. When it keeps breaking down because they can't yet do the thing, that's a teaching problem: simplify the Task a long way, or step out and teach the skill directly with a demonstration and some repetition, then put it back into a game. Teach It or Let the Game Teach It? goes deeper on telling the two apart, and How Much Should You Actually Say? covers when Instruct and Demonstrate are the right call.

What I'd cut from a session plan

Look at your session and ask which pieces are shaping the environment on purpose, and which are there because they make you feel prepared. Usually it's:

• The fourth activity, when three would have done the job properly
• Two or three conditions running at once, none of them clearly doing work on its own
• The backup plan you wrote in case the first one didn't land, which is really a plan to abandon the environment you designed the moment it gets noisy

Cutting these isn't lowering your standards. You're choosing to run one environment well instead of three half-attentively.

None of that is from a study. It's what I see. Watching a lot of other coaches' sessions as well as my own, the pattern is almost never bad coaching. It's sessions carrying one activity more than they need.

What's next

Building the environment is half the job. The other half is what you do once it's running: when you step in, how you say it, and whether you need to say anything at all. That's the next piece in this series, followed by one on coaching without stopping the game every two minutes.

For now, pick your next session. Before you add another condition, ask whether adjusting Space or Players would do that work for you.

Common questions about youth soccer practice

Why does my youth soccer practice feel so chaotic? Often the environment is working against what you're asking for. If the space, the numbers and the scoring don't require the behavior you want, instruction from the sideline has to fight the game to get it. The other common cause is a game players can't yet execute, in which case simplify it or teach the skill directly first.

How many coaching points should you give a young player? There's no research-backed number. One study found five explicit instructions was enough to make children with lower working memory measurably worse, while the same five helped children with higher working memory. Five was just what that study tested. I'd pick one thing for an activity to be about, and treat anything past a second point as probably not landing for a lot of the group.

How many players should be on each team in a youth practice? Fewer than you might think. In one study of ten-year-olds, 4v4 doubled the actions per player per minute compared to 7v7, and studies with players as young as under-7 show the same pattern: more touches each, less time standing in line. If a session feels slow, cutting team sizes and running more games at once is usually a faster fix than anything you say. Small-Sided Games for Young Players has more of the research.

Is game-based practice better than drills? For tactical skills, the comparison studies mostly favor game-based practice. For technique, most comparisons found no difference, and none found drills better. The research is thin, especially with young children, and a player who can't yet do a skill may need it taught directly before a game can teach them when to use it.

---

This is part one of a series on what actually shapes a youth practice. See The Fundamentals of Soccer Coaching for the broader idea it's built on, and continue with How Much Should You Actually Say?, which covers what to do once the session is running.