By Paul Dale | The 3AM Method Series FOR PARENTS
| Understanding that your child’s ability to cope mentally under stress is limited by their age, changes how you respond to their results. |
In Part 1, we established the architecture: the Chimp, the Human, and the Computer — three brain systems with three different speeds, three different agendas, and three very different roles in competitive performance. In Part 2, we looked at what this means for parents watching their child from the sideline, and why the instinct to fix, coach, and motivate so often makes things worse.
Now we get to the work itself.
Because understanding the three-brain model is not the destination. It is the map. And the territory — the actual training, the deliberate sessions, the hard and unglamorous process of programming a better response into the Computer — is where the real coaching happens. That is what this article is about.
The Chimp will always speak. The question is not whether it speaks — it is what the Computer does in reply.
3AM Coaching Principle
Over the course of my career coaching at every level — from grassroots juniors to national teams across four countries — the most common mistake I have seen is not bad technique. It is well-intentioned coaching delivered in entirely the wrong conditions. Players who have been rehearsed brilliantly for calm, and not at all for chaos. Players whose computers have been beautifully programmed for practice and left completely blank for competition.
The result, predictably, is a player like Pablo.
THE PROBLEM WITH COMFORTABLE PRACTICE
Most junior training environments share a common design flaw: they are comfortable. The coach feeds balls from the basket in a predictable sequence. The rally partner cooperates. There is no score, no consequence, no pressure. Errors are met with gentle correction. The atmosphere is constructive and encouraging.
None of this is wrong, in isolation. There is an important place for calm, technical practice — particularly when introducing new patterns or rebuilding technique. But if all of the training happens in comfortable conditions, the Computer receives a very specific — and very incomplete — education.
It learns to play tennis when everything is fine.
The problem, of course, is that competitive matches are not fine. The opponent is trying to win. The score creates stakes. The Chimp activates early and often. And when the Computer goes looking for a programme to handle these conditions, it finds nothing there — because it was never taught what to do when the environment became uncomfortable.
The Core Diagnostic Question
Before designing any training session, every coach must be willing to answer this honestly: “Have we actually trained the Computer for this — or have we only ever practised it in calm conditions?” If the honest answer is “no,” then the solution is not more repetition of the same drill. The solution is a fundamentally different kind of practice.
The three-brain model tells us precisely why this matters. The Computer retrieves programmes automatically — at speed, without conscious deliberation. When the Chimp activates during a match, the Human brain (the seat of reason and decision-making) is already playing catch-up. The Computer fires first. And it will fire whatever programme is most deeply installed — regardless of whether that programme serves the player or sabotages them.
If the most deeply installed programme is “execute technique in calm conditions,” that is what the Computer will attempt to run. And it will fail in competition, not because the player lacks talent, but because the programme was never designed for the environment it is now being asked to operate in.
How the Computer Actually Learns
Before we can deliberately train the Computer, we need to understand how it installs new programmes. Professor Steve Peters describes the Computer as a storage-and-retrieval system — powerful, fast, and entirely dependent on what has been loaded into it. Understanding the conditions under which learning occurs is the foundation of intelligent training design.
The Computer does not learn from explanation alone. It does not install a programme because a player has been told the right thing to do, or because they intellectually understand the correct response. It installs programmes through a very specific combination of two elements.
This is why the 3AM methodology insists that pressure must be present in training, not as an occasional add-on, but as a deliberate and regular feature of the training environment. We are not trying to make practice harder for its own sake. We are matching the training conditions to the retrieval conditions — so the Computer can actually find the right programme at the right moment.
Gremlins and Autopilots: The Two Kinds of Software
Peters draws a critical distinction between two types of programmes the Computer can hold. Both are automatic. Both fire without conscious deliberation. But their effect on performance could not be more different.
🔴 Gremlins — Unhelpful Programmes
“I always choke when I’m serving for the set.”
🟢 Autopilots — Helpful Programmes
“When I’m tight on serve, I go back to my routine, bounce the ball, and trust my toss.”
🔴 Gremlin
“If I lose the first set, I’m probably going to lose this match.”
🟢 Autopilot
“One set tells me nothing about the next. Reset and compete.”
🔴 Gremlin
“I can’t change my tactics mid-match — I don’t know how.”
🟢 Autopilot
“Something isn’t working. What’s one adjustment I can make right now?”
🔴 Gremlin
“Mistakes are a sign I’m not good enough to be here.”
🟢 Autopilot
“Errors are data. What just happened, and what do I do next?”
Both Gremlins and Autopilots were installed through experience. Neither was consciously chosen. The child who spent years hearing “you always crack under pressure” did not decide to believe it — the Computer installed it from the repetition and emotional weight of the message. Equally, the player who has spent hundreds of sessions practising calm recovery from errors does not choose to reset — the Computer automatically retrieves the Autopilot.
This is the most important practical insight in the entire framework: Gremlins are not removed by understanding. They are replaced by a better-installed alternative.
You cannot talk a Gremlin out of the Computer. You cannot reason it away with a conversation, however insightful. The only way to genuinely displace a Gremlin is to build, through deliberate repetition and emotional charge, an Autopilot that is more deeply installed than the Gremlin it replaces. Understanding comes first — it shows the player what is running in there. But the actual replacement requires training.
The Four Programmes Every Competitive Player Must Have Installed
In Part 1, I posed the central coaching question: have we actually trained the Computer to deal with this? That question points to four specific areas where the Computer must hold well-installed Autopilots if a player is going to perform under competitive pressure. Let me address each of them directly — and practically.
Applied 3AM Analysis
What the Chimp Needs from the Coach
There is one more dimension of Computer training that is frequently overlooked: the Chimp itself needs to be trained, not just managed.
We established in Part 1 that the Chimp’s job is to warn the system and then step back. The problem for many competitive players is that their Chimp has never learned to step back — because it has never experienced a training environment where stepping back was the right response. The Chimp has been given plenty of evidence that its warnings matter (errors happen, matches are lost, the threat is real) and very little evidence that the Human and Computer can handle what comes next.
Every time a player practises staying in control after an error — every time they execute a reset routine and win the next point, every time they adapt mid-drill and find a solution — they are providing the Chimp with new evidence. The system can handle this. The warning was heard. Now something useful is happening. The Chimp, receiving this feedback repeatedly, begins to release its grip more readily.
This is not a metaphor. It is a functional description of how the three systems interact when training is designed well. The Chimp is designed to stand down when the Human and Computer demonstrate that they have things under control. The coach’s job is to provide the conditions in which that demonstration can happen — again and again, until the evidence is compelling enough to shift the Chimp’s default response.
Every time a player finds a solution under pressure, they are sending a message to their Chimp: you gave the alarm, and we handled it. Step back. We have this.
3AM Coaching Principle
What “Training the Computer” Looks Like in Practice — for Coaches
Translating these principles into a working training approach requires a shift in how coaches design sessions. This does not mean abandoning technical work. It means wrapping technical work in a framework that considers what the Computer is learning alongside the body.
Before the Session
Ask three questions: What is the technical focus today? What is the Computer being trained to do? And what pressure level is appropriate for where this player currently is? These three questions sit alongside each other, not in sequence. The Computer is always being trained, whether we design it intentionally or not. The only question is whether what it is learning is useful.
During the Session
Build in compression. Begin with technical work in lower pressure conditions. Progressively increase the stakes — introduce scoring, consequence, competitive format — until the player is executing the target pattern under genuine pressure. Then observe. What holds? What breaks? Which Gremlins appear? This is not a session going badly. This is diagnostic data being generated in real time.
Do not over-correct during high-pressure sequences. When the Chimp is active and the Computer is working hard to retrieve the right programme, constant coaching interruptions simply add more input to an already overloaded system. Name the error if needed. Then let the Computer continue to work. The correction belongs in the reflection after the drill — not in the middle of it.
After the Session
The reflection phase is where the Human brain does its most important work. Ask the player — not “what went wrong?” but “what did you notice?” and “what did your Computer try to do?” and “if that moment came again, what do you want to be available?” These questions invite the Human to participate in the reprogramming process. The Computer installs most reliably what the Human has understood, chosen, and committed to practising.
What Progress Looks Like — and Why Patience Is the Most Important Tool
Computer reprogramming is not a linear process. It does not happen because a player has understood something important, or because they had one brilliant session, or because a coach asked exactly the right question. It happens through accumulated experience — through the patient, unglamorous repetition of doing the right thing in the right conditions over months and years.
What realistic progress looks like for a player working through this process:
-
Early Weeks
The player can name the Gremlin after it has run. They notice it in retrospect — after the match, after the session. This is already significant progress. The Human is becoming aware of what the Computer is doing. Awareness precedes change.
-
Months 2–4
The player begins to catch the Gremlin mid-activation — noticing it as it happens, rather than only afterwards. They may not yet be able to interrupt it reliably, but they can name it in real time. The Chimp loop is becoming more transparent. The reset Autopilot is beginning to fire in some situations.
-
Months 4–8
Moments of Autopilot retrieval become regular. Not in every match — and under the most extreme pressure, the Gremlin cluster still fires — but in practice and in lower-stakes competition, the better programmes are running. The player begins to trust that the reset works. The Chimp gets the message more often.
-
Year Two and Beyond
The Autopilots are sufficiently embedded to hold up under high-pressure match conditions. Not perfectly, not always — the Computer is a dynamic system that will always reflect the most recent training as much as the most established. But the architecture has shifted. The Gremlin cluster that once dominated competitive performance is no longer the default. A better set of programmes is now the most deeply installed. The player competes from that foundation.
These timelines are not fixed. They vary with age, the depth of the Gremlin cluster, the quality and consistency of the training environment, and the developmental stage of the Human brain. A twelve-year-old with a very young prefrontal cortex will progress differently from a seventeen-year-old whose impulse control has had more years to develop. A player who has held a particular Gremlin since early childhood will need more repetition to displace it than one whose unhelpful programme was more recently installed.
What does not vary is the mechanism. Repetition plus emotional charge in condition-matched environments is how the Computer learns. There is no shortcut to that process — only the patience and skill to design it well.
The 3AM Methodology
The Chimp warns. The Human understands. The Computer acts.
The coach’s job is to make sure the Computer has been given something worth acting on.
Conclusion: Every Session Is an Installation
The most powerful shift a coach can make — in how they think about training, in how they design sessions, in how they observe and respond to what they see on court — is this one: every session is installing something into the Computer.
Not every coach has framed it that way. But every coach who has ever run a session has done it regardless. The player who practises their serve under comfortable conditions for six months has had something installed. The player who sits on the bench between sets and runs the same self-critical internal script every time an error happens has had something installed. The player whose coach shouted “you always do this!” in a crucial moment had something installed in that moment that may take years to dislodge.
The 3AM methodology does not change what is happening. It simply makes it conscious. It gives coaches, players, and parents the language to ask: what are we installing? Is it serving this player? And if not — what would we rather install instead?
Understanding the answers to those questions and designing the training to act on them is the most important coaching work there is. It is slower than the drilling technique. It is less immediately visible than improving a serve or fixing a backhand. But it is the work that determines whether the technique, the serve, and the backhand are actually available when the match is on the line.
Pablo — and every player like him — is not waiting for a better coach. He is waiting for better software. And software, unlike talent, can always be written.