How Players Learn to See the Game Faster: The Science of Decision-Making, and How to Train It on an African Court

Every coach has said it. "He just sees the game." "She's always a step ahead." "He reads it before it happens." We say these things about our best players as if we are describing a gift — something they were born with, something the others simply don't have.

We are wrong about that. Not completely wrong, but wrong in the way that matters for our job. Seeing the game faster is not mostly a gift. It is mostly a trained skill — and once you understand the science of how it is trained, you will realize that a great deal of what we do in practice actively prevents our players from developing it.

This is the longest piece I have written for this blog, because it is the most important one. The BAL series I just finished ended on a single idea: the team that won the championship was the team whose players could read what a game needed and change what they were doing to match it. Adaptable players. Players who could see it and solve it in real time. The obvious question — the one every African coach reading that series should have asked — is how do you actually train that? How do you build a player who sees the game faster, on a concrete court, with forty players and six balls and no technology?

That is what this piece answers. It is going to name the actual science, because you deserve to know the names and the ideas, not just the drills. But every piece of it lands on the floor of a real African gym. Let's go.

Part 1: What "seeing the game" actually is

Start with a question that sounds simple and isn't. When a great point guard makes a brilliant pass, what exactly is the skill?

The intuitive answer is that they have great vision — they see more of the court. But that's not quite it. The research on what separates expert players from novices is one of the oldest and most robust bodies of work in sport science, going back to the 1960s, and it does not say experts have better eyesight. A player with 20/20 vision and a player who is an elite decision-maker are looking at the same court. The difference is in what they extract from what they see.

Three findings from that literature matter for us, and they are remarkably consistent across sports — they have been demonstrated in squash, soccer, volleyball, basketball, and more.

First: experts read cues earlier. The foundational work here was done by Bruce Abernethy in the late 1980s and early 1990s, studying squash and badminton players. He used a clever method: he showed players film of an opponent striking the ball, but cut the film off at different moments — before contact, at contact, just after — and asked them to predict where the ball was going. Experts could predict the shot accurately from the opponent's body before the ball was even struck. Novices had to wait until they could see the ball's flight. The expert was reading the hips, the shoulders, the racquet preparation — the advance cues — while the novice was waiting for information that arrived too late to act on. Abernethy called this advance cue utilization, and it is one of the most replicated findings in the field.

Translate that to basketball. A great help defender starts moving on the drive before the ball-handler has committed, because they are reading the hips and the shoulders and the angle of the first step. A great passer throws to the open man before he is open, because they read the rotation of the defense a beat before it happens. They are not faster of foot. They are earlier in their reading.

Second: experts see patterns, not pieces. A separate line of research — going back to chess studies and brought into sport by researchers like Paul Ward and Mark Williams — shows that experts recognize and recall structured patterns of play far better than novices, but only when the patterns are real. Show an expert a structured game situation for a few seconds and they can reconstruct where everyone was. Show them random, meaningless positions and their advantage mostly disappears. The expert isn't remembering twelve individual players. They are seeing two or three meaningful shapes — "that's a high pick-and-roll with the weak side loading up" — the way you read a word instead of spelling out the letters. This is pattern recognition, and it is why an experienced player can glance at a defense and instantly know what is available, while a young player sees ten separate moving bodies and freezes.

Third: experts look in better places. Research on visual search — including work using eye-tracking — found that experts use fewer, longer fixations than novices, and often use a visual anchor point to take in several things at once with their peripheral vision rather than darting their eyes around (Mann and colleagues established this clearly in a 2007 meta-analysis). The novice's eyes jump frantically from the ball to a man to another man. The expert holds a stable gaze — often somewhere around the opponent's chest or the space between players — and reads the whole picture from there. They also use situational probabilities: they know, from experience, what is likely to happen in this situation, so they don't waste attention on the unlikely.

Now, here is the crucial point, and the one that determines everything about how we should coach. For decades, the assumption was that because experts have these perceptual skills, the way to develop players is to train the perception directly — show them film, run them through recognition drills, train their "vision" in isolation. Some of that has value. But the deeper and more recent understanding of the science says something different and more demanding: these perceptual skills are not separate from action. They are built by acting. You cannot fully learn to read a defense by watching defenses. You learn to read a defense by playing against live defenses, thousands of times, while making real decisions with real consequences.

To understand why, we need the single most important idea in modern skill acquisition. It has a name, and the name is worth knowing.

Part 2: Perception and action are one loop, not two steps

The idea is called perception-action coupling, and it comes out of a school of thought called ecological psychology, founded by a researcher named James Gibson in the mid-20th century. I am going to explain it carefully because once you see it, you cannot unsee it, and it changes how you watch your own practices.

The traditional, common-sense model of how a player makes a decision goes like this: the player perceives the situation (sees the floor), then thinks about it (processes the options, decides), then acts (executes the pass). Perceive, then think, then do. Three separate steps, happening in order, with the brain in the middle doing the deciding. Almost all traditional coaching is built on this model — which is why we stop play, explain the right decision, and expect the explanation to transfer to the game.

Gibson's insight — and the decades of sport research built on it — is that this model is wrong, or at least deeply misleading, for fast and open skills. In the real game, perception and action are not separate steps. They are a single continuous loop. You perceive in order to act, and you act in order to perceive better. The defender shifts, which changes what you see, which changes how you move, which changes what the defender does, which changes what you see — continuously, in real time, far too fast for a "stop and think and then decide" process to keep up.

Gibson gave us a second idea that makes this concrete, and it is my favorite concept in all of this: the affordance. An affordance is an opportunity for action that the environment offers you, specifically, right now. A passing lane is an affordance. A gap to drive is an affordance. A closeout that's a half-step too long is an affordance to attack. Crucially, affordances are relationships between the player and the situation — they depend on who you are. A passing lane that is "open" for a 6'8" guard with a high release is not open for a 5'9" guard, because the affordance depends on the body perceiving it. A gap that affords a drive to a quick player does not afford a drive to a slow one.

This is why great decision-making cannot be a list of rules. The right basketball decision is not "when X happens, do Y." It is "perceive what this specific situation affords you, with your body and your abilities, against these specific defenders, in this specific moment — and act on it." The skill of seeing the game is the skill of perceiving affordances — and you can only learn to perceive the affordances that are real for you by acting on them yourself and finding out which ones are really there.

Here is the coaching consequence, and it is severe. When you remove the defender from a drill, you remove the affordances. A player doing a cone dribbling drill is not perceiving any opportunities for action, because a cone does not defend, does not shift, does not offer or take away a passing lane. The player is just executing a movement. They may get better at the movement. But they are not developing the perception-action coupling that the actual game requires, because there is nothing to perceive and nothing to read. You have taken the single most important ingredient — the live, reading-and-responding relationship between the player and a real opponent — and thrown it in the bin, and kept only the empty shell of the movement.

This is the scientific reason that the cone drills I have been arguing against for this entire blog's existence do not transfer to games. It is not just that they are unrealistic. It is that they decouple perception from action, and perception-action coupling is the actual thing being learned. A decoupled drill teaches a decoupled skill, and there is no decoupled skill in a real basketball game.

Part 3: The performer, the task, and the environment

So if we don't develop decision-making by explaining decisions and running decoupled drills, how do we develop it? This is where the science becomes intensely practical, and where it hands us a tool we can use on any court in Africa, free of charge.

The tool comes from a framework called ecological dynamics, which combines Gibson's ecological psychology with the science of how complex systems organize themselves. Its practical coaching method is called the constraints-led approach, and it rests on a model proposed by a researcher named Karl Newell in 1986. Newell said that every movement a player produces — every decision, every action — emerges from the interaction of three things. Learn these three. They are the entire toolkit.

One: the individual (or "organismic") constraints. This is the player themselves — their height, their strength, their speed, their skill level, but also their confidence, their fatigue, their emotional state. A tired player perceives the game differently than a fresh one. A fearful player sees fewer affordances than a confident one.

Two: the task constraints. This is the structure of the activity — the rules, the goal, the number of players, the size of the court, the equipment, the scoring system. These are the things you, the coach, control directly. You decide it's 3-on-3. You decide the court is half-size. You decide a score only counts if everyone touched the ball. You decide there are no dribbles allowed.

Three: the environmental constraints. This is everything outside the task — the physical conditions (a concrete court, harsh sun, an uneven surface), but also the social and cultural surroundings (whether players feel safe to take risks, the expectations they carry, the noise).

Newell's insight is that movement and decision-making emerge from the interaction of these three — they are not installed by instruction. And this gives the coach a completely different job description. Your job is not primarily to tell players what to do. Your job is to design the constraints so that the right kind of decisions naturally emerge from the players solving the problem themselves. The phrase the researchers use is that the coach becomes a "designer of learning environments." You set up the situation so that the game itself teaches the lesson.

Let me make this concrete, because this is the part you can use tomorrow.

Suppose you want to teach players to move the ball quickly and stop them from over-dribbling — a near-universal problem in youth basketball, and one I see constantly on African courts where every young player wants to be the one with the ball. The traditional approach: stop play, lecture them about ball movement, tell them to pass more, run a passing drill with no defense. It rarely works, because the moment the real game returns, the habit returns.

The constraints-led approach: change the task constraint. Play a small game — 3-on-3, 4-on-4 — with one rule added: you may not dribble at all, or you may take only one dribble. Now say nothing. Just let them play. Within minutes, the players are forced to move without the ball, to cut, to find angles, to read where the pass is available — because the constraint has made over-dribbling impossible and made ball movement the only way to succeed. They are not following your instruction to pass more. They are discovering, themselves, the affordances that ball movement creates, because you have designed a world in which that is the way to win. The lesson installs itself, through their own perception and action, which is exactly why it sticks when the constraint is removed.

That is the whole method. You identify the behavior you want, and instead of instructing it, you design a constraint that makes it emerge. Want players to attack the rim instead of settling for jumpers? Make baskets in the paint worth more. Want them to learn to use both hands? Constrain the strong hand. Want them to communicate on defense? Play a game where they're allowed to talk but the offense plays in silence. Want quicker decisions? Shrink the court or shrink the shot clock so there's less time to hold the ball. The constraint does the teaching.

Part 4: Why the practice that looks worse is often working better

Now I have to tell you something that will feel wrong, because it contradicts one of the deepest instincts we have as coaches. It is one of the most established findings in the science of motor learning, and almost every coach gets it backwards.

We instinctively believe that practice should look good. We want crisp reps, clean execution, smooth drills, visible improvement within the session. When a drill produces lots of mistakes and messy execution, we feel like it isn't working, and we often abandon it. This instinct is leading us astray, and the science explaining why is among the most important things a coach can understand.

The key distinction comes from research by Robert Bjork and others, and it is the difference between performance and learning. Performance is how well someone executes during practice. Learning is how much they have actually durably acquired — what remains days later, and whether it transfers to new situations. These sound like the same thing. They are not. And the crucial discovery is that the practice conditions that produce the best performance during the session often produce the worst learning, and the conditions that make performance look worse during the session often produce the best learning. Bjork called the helpful kind of difficulty "desirable difficulties."

The clearest example is something called the contextual interference effect, demonstrated in a famous 1979 study by Shea and Morgan and replicated many times since. Take two ways of organizing the same practice. In blocked practice, you work on one thing at a time, repeated: shot, shot, shot, shot, then move on to layups, layup, layup, layup, then passing. In random (or interleaved) practice, you mix them up unpredictably: shot, layup, pass, layup, shot, pass, in no fixed order. Blocked practice looks far better during the session — the player grooves the shot, the reps get smoother, everyone feels good. Random practice looks messier — more misses, less rhythm, more visible struggle.

And yet, on a test days later, and especially on transfer to new situations, the players who did the messy, random practice consistently outperform the ones who did the smooth, blocked practice. Every time. The struggle was the point. Because in random practice, the player has to reconstruct the solution fresh on every rep — they can't just repeat the last motion on autopilot — and that reconstruction is the actual learning process. The blocked player grooving the same shot ten times in a row is, after the first couple, barely learning at all. They are just repeating a motion their body has already temporarily solved.

Why does this matter so much for African coaches specifically? Because a real basketball game is the most random, interleaved, contextually-interfering environment there is. No two possessions are the same. The game never lets you do the same thing ten times in a row on autopilot. So practice that is random, varied, messy, and game-like is practice that prepares players for the actual demands of the game — while practice that is blocked, grooved, and clean prepares them for a tidy world that does not exist once the whistle blows. The drill that looks impressive in your session is often the one building the least transferable skill. The chaotic small-sided game that makes you wince is often the one building players who can actually play.

This connects directly to everything above. Live small-sided games are random and interleaved by their nature. They couple perception and action by their nature. They present real affordances by their nature. They are, almost by definition, the most efficient decision-training environment available — and they require no equipment beyond a ball and some bodies, which makes them the perfect tool for our context.

There's a related idea worth naming, because the research on it has grown in the last fifteen years: variability is not noise to be eliminated — it is part of the learning. A researcher named Wolfgang Schöllhorn developed an approach sometimes called differential learning, built on the finding that adding variation to movement — never repeating a rep exactly the same way, deliberately introducing fluctuations — can produce better learning than trying to groove one "correct" technique. The deeper principle, which goes back to a Russian scientist named Nikolai Bernstein, is that there is no single perfect movement to be copied; skilled players are ones who can adapt their movement to each unique situation. A jump shot is never thrown twice from exactly the same place against exactly the same closeout. So a player who has only ever grooved one identical shooting motion is less prepared than one who has explored many variations and learned to adapt. Repetition without repetition, Bernstein called it: repeating the problem (put the ball in the basket) without repeating the solution (the exact motion) — because every real attempt is a slightly different problem demanding a slightly different solution.

Part 5: The honest part — what does NOT work, and where instruction still matters

This blog's standard is that I tell you what the evidence says, including when the evidence inconveniences a fashionable idea. There are two places I have to do that here.

First: be very skeptical of "brain training" gadgets. There is a large and growing commercial industry selling tools that claim to train the athletic brain in general — stroboscopic glasses that flicker to "train your visual system," computerized cognitive drills, neuro-tracking software, lights-on-a-wall reaction trainers. They are seductive, they look scientific, and increasingly they are being marketed across Africa as the modern, professional way to develop players. Here is what the evidence actually says, and it was stated clearly in a 2024 review in the journal Sports Medicine by Job Fransen: these generic tools can produce near transfer — you get better at the specific task the tool trains, and sometimes at similar lab tests — but there is no good evidence of far transfer to actual sport performance. Training your reaction time on a flashing-light board makes you better at flashing-light boards. It does not reliably make you a better basketball player, because the skill of basketball is not generic reaction speed — it is reading this game, perceiving basketball affordances, coupled to basketball actions. The perceptual skill of an expert is exquisitely specific to their sport. A gadget that trains perception detached from the sport trains a detached skill, for the same reason a cone drill does.

This matters enormously for us, because we do not have money to waste. A club in Luanda or Lagos or Nairobi tempted to spend scarce funds on flashing lights and strobe glasses should know that the science does not support it — and that the most powerful decision-training tool in existence, the small-sided game, costs nothing. Do not let anyone sell you an expensive solution to a problem that a 3-on-3 game solves better and for free.

Second: the constraints-led approach does NOT mean "never coach anything" or "just roll the ball out." This is the most common misunderstanding of everything I've written above, and I need to correct it before someone takes this article and uses it to justify lazy coaching. Saying that decision-making emerges from designed constraints does not mean the coach is silent and passive, or that technique never gets taught, or that you just throw the players a ball and watch. That is not the constraints-led approach; that is abdication.

The skilled constraints-based coach is intensely active — just active in a different way. They are constantly observing what is emerging, adjusting the constraints when they're not producing the right behavior, asking players good questions that direct their attention to the right information ("what did you see there?" "where was the space?"), and yes, sometimes giving direct instruction or demonstrating technique when that is what the moment needs. The research does not say instruction is useless. It says instruction is one tool among several, that it is badly overused in traditional coaching, and that it works best when it is brief, well-timed, and points the player toward information they can then use in live play — rather than a long lecture that replaces the playing. A young player who literally does not know how to perform a jump stop may need to be shown. The point is not to never teach the movement; the point is that the decision of when to use it can only be learned in live, coupled, game-like play. Teach technique when technique is the bottleneck. Design constraints when decision-making is the bottleneck. Know which is which. That judgment is the art of coaching, and no framework removes the need for it.

Part 6: Putting it on your court tomorrow

Let me bring this down to specific things you can do, in our conditions, with what we have. None of this requires money, technology, or more than a ball and a group of players.

Play more, drill less — but design the play. The single biggest change most African youth programs could make is to shift the balance of practice away from decoupled line drills and toward small-sided games with designed constraints. Not unstructured scrimmage — designed games, where you've set a constraint to draw out a specific behavior. 1-on-1, 2-on-2, 3-on-3, 4-on-4, in various configurations, are your primary decision-training tools. This is also a gift for our reality of large numbers: instead of forty players standing in four long lines waiting for one turn at a cone drill, you can run five small-sided games at once across the width of a court and have everyone playing, reading, and deciding constantly. More players active, more decisions per player, better learning — all at the same time.

Use numerical imbalance deliberately. One of the most studied and useful constraints in basketball is playing with uneven numbers. Research on small-sided games — including work specifically on basketball with youth players — shows that manipulating the number of attackers versus defenders sharply changes the decisions players have to make. Play 3-on-2 or 4-on-3 and the offense is forced to read advantages, move the ball ahead of the defense, and make quick decisions about when to attack the extra man. Play 3-on-4, putting the offense at a disadvantage, and you force patient, precise decision-making under pressure. Shifting these numbers is like turning a dial on the difficulty and the specific reads you want to train.

Manipulate space and time. Shrink the court to force quicker reads and tighter handles. Enlarge it to train spacing and pace. Shorten the shot clock to remove the option of over-dribbling and force fast decisions. Each of these is a constraint that changes what the game demands without you saying a word.

Change the scoring to change the behavior. This is one of the most powerful and underused tools. What you reward is what they will learn to perceive. Make paint baskets worth three and jump shots worth one, and watch players suddenly start reading driving lanes they used to ignore. Award a point for a defensive stop, not just a basket, and watch defensive engagement transform. Require a certain number of passes, or that the ball touch both sides of the floor, before a shot counts, and watch ball movement and court-reading emerge on their own.

Ask, don't just tell. When you do stop play, lead with a question that points attention at the right information, rather than a statement that hands over the answer. "What did you see?" "Where was the help coming from?" "What was open that you didn't take?" This keeps the player in the perceiving-and-deciding loop rather than turning them into someone who waits to be told. A player who is always told what to do learns to wait for instructions; a player who is always asked what they saw learns to see.

Keep the variability in. Resist the urge to make practice tidy. Let it be messy, varied, unpredictable. Mix skills rather than blocking them. Don't panic when a game-like drill produces mistakes — the mistakes are often the sound of learning happening. The smooth, grooved, beautiful-looking drill is frequently the one teaching the least.

Protect the players' freedom to explore — which in our context is a cultural act. Players only develop the ability to read and solve the game if they are allowed to try things, fail, and adjust. In many African basketball cultures — and I say this as someone who works inside one — there is a strong coaching tradition of authority, correction, and fear, where a player who tries something creative and fails gets pulled and scolded. The science is clear that this suppresses exactly the exploration that builds decision-making. A player terrified of making a mistake stops perceiving affordances and starts perceiving only the bench. If you want players who see the game, you have to build an environment where it is safe to look. That is not softness. That is how the skill is built.

Why this is the whole game

Let me close by connecting this back to where the BAL series ended, because it is all one argument.

The team that won Season 6 won because its players could read what the game demanded and adapt — play fast when fast was available, defend and grind when the championship required it. Bilingual players. Players who could see it and solve it in real time. That adaptability, that game-sight, that capacity to perceive what a situation affords and respond to it under the highest pressure — that is the most valuable thing a basketball player can possess, and it is the hardest thing to develop.

The deep message of the science is that this capacity is not a gift reserved for a lucky few, and it is not built by the methods most of us were taught to use. It is built by playing the game, in designed and varied and game-realistic forms, thousands and thousands of times, while making real decisions with real consequences against real opponents, in an environment where it is safe to explore and the constraints are arranged so the right reads emerge. It is built by coupling perception to action and never tearing them apart. It is not built by cones, by lines, by lectures, by grooved decoupled drills, or by expensive gadgets that train detached skills.

And here is the part that should make every African coach sit up: none of what actually works requires money. The single most powerful decision-development tool in all of sport science is the small-sided game, and it costs nothing. A concrete court, a ball, some bodies, and a coach who understands how to design the constraints — that is a world-class decision-training environment. We do not need to wait for facilities or technology or budgets to develop players who see the game. We need to understand the science and rearrange what we already do.

The player who "just sees the game" was, almost always, made — in a gym, by a process, that we can understand and reproduce. That is the most hopeful thing I know about coaching. The gift was never the point. The process is the point. And the process is available to all of us, right now, on the courts we already have.

Go design something. The next time you run practice, pick one behavior you want to see more of, and instead of telling your players to do it, build a small game where it's the only way to win. Then stand back, and watch them teach themselves. That is the whole science, on your court, for free.

A note on the science behind this piece: the ideas here come from established bodies of research in motor learning and skill acquisition — the ecological psychology of James Gibson; the expert-performance and anticipation work of Bruce Abernethy, Paul Ward, and Mark Williams; the ecological-dynamics and constraints-led framework developed by Keith Davids, Duarte Araújo, Ian Renshaw, Ross Pinder and colleagues, building on Karl Newell's 1986 constraints model; the performance-versus-learning and "desirable difficulties" work of Robert Bjork; the contextual-interference research of Shea and Morgan; the movement-variability ideas of Nikolai Bernstein and Wolfgang Schöllhorn; and recent reviews on the limits of generic perceptual-cognitive training, including Job Fransen's 2024 paper in Sports Medicine. I've translated their work into coaching language; any oversimplifications in that translation are mine.

Liked this piece?

I'm finishing an ebook called The African Skills Coach — Game-Like Player Development in Tough Environments. Everything in this article — designing constraints, building decision-making, developing players who see the game, all adapted to courts with limited equipment and large numbers — is what the ebook is about, in practical, session-by-session detail. The ebook is coming soon. Click here to join the waitlist.

Coach Sérgio Benitez Cristóvão
Head of Youth Basketball, Atlético Petro de Luanda
Modern Hoops Africa · coachsergiobcristovao.blog