You have almost certainly been told to read Thinking, Fast and Slow by Daniel Kahneman. You have almost certainly not finished it. Fair enough, it is roughly 500 pages and a lot of it is about gambles, insurance and stock pickers. But somewhere in there is one sentence worth the whole book for a coach: confidence is not evidence. The feeling of being sure tells you how neat your story is, not how true it is. And if that is true, quite a lot of what happens on a Saturday afternoon needs a second look.
TL;DR
This is not a book review. It is a distillation: the handful of Daniel Kahneman's ideas that survive the translation onto a hockey pitch, what they mean for the way coaches judge players, plan seasons and talk to a team, and where they line up with things guest coaches have been saying in masterclasses on The Hockey Site for years. Every claim is tied to a chapter, so you can go straight to the five pages that matter to you and ignore the rest.
Sources used
Daniel Kahneman, Thinking, Fast and Slow (2011). Chapter references are given in the text.
Start with the honest bit: this is not a light read
Daniel Kahneman won a Nobel prize for work on how people actually make decisions, as opposed to how economists assumed they did. Thinking, Fast and Slow is his summary of a career's worth of that research, much of it done with Amos Tversky. It is brilliant, and it is hard going. It circles back on itself, and most of us will never finish it. That is fine. What follows is the result of reading it looking for what a coach could actually use, keeping only the ideas that change something you do on the pitch or on the bench.
Two systems, and why a team needs both of them trained
Kahneman's whole book hangs on one distinction. There are two modes of thinking, and he calls them System 1 and System 2.
System 1 operates automatically and quickly, with little or no effort and no sense of voluntary control.
System 2 allocates attention to the effortful mental activities that demand it, including complex computations.
That is chapter 1. System 1 is the player who receives on the turn and has already played the ball into the channel before anyone has consciously decided anything. System 2 is that same player on Monday, watching the clip back, working out why the second option was better.
The catch, and this is the part worth sitting with, is that System 2 is lazy. It tends to accept whatever System 1 hands it. A fluent, fast impression gets rubber-stamped as fact. That applies as much to the coach on the sideline as to the player on the pitch.
For a session, the practical split is clean. Fast reads are System 1 and they are built by repetition in game-like pictures until they are quick and reliable. Tin Matkovic's masterclass on pre-scanning is exactly this: scanning as mapping rather than glancing, timed to the arrival of the ball so that the picture is already there before the first touch. Robert Noall then adds the bit coaches often skip, which is what happens after the scan, when the pre-planned option closes and the player has to change their mind mid-action. His rule in the workshop on on-ball decision making, "don't marry your prescan", describes a System 1 skill that has to be trained on its own.
Planning, video review and honest self-assessment are the other system. They are effortful, they tire quickly, and they should be protected from rush and fatigue. Doing squad selection in the car park twenty minutes after a bad loss is not a System 2 decision, whatever it feels like.
Which of your regular coaching decisions are you actually making on the slow system, and which ones just feel considered because you made them out loud?
Confidence is not evidence
If there is one chapter to screenshot for a coaching group, it is chapter 7. Kahneman gives the idea a deliberately ugly name so it sticks.
...I will use a cumbersome abbreviation for it: WYSIATI, which stands for what you see is all there is.
The mind builds the best available story out of whatever information happens to be in front of it, and it does not register what is missing. Confidence then tracks the coherence of that story rather than the quality of the evidence behind it. A tidy explanation of why a player keeps losing the ball in the middle third may simply be the tidiest thing you can build from the two clips you happened to watch.
And here's the thing. A cluster of the book's other findings are really just WYSIATI wearing different shirts, and each one has an obvious hockey version.
The halo effect (chapter 19): one salient impression bleeds into everything else. A player with a beautiful aerial gets credited with a defensive reading of the game that nobody has actually checked.
The availability heuristic (chapter 12): we judge how common something is by how easily examples come to mind. The vivid error in the last five minutes of the fourth quarter weighs more in selection than sixty steady minutes from three weeks ago.
The law of small numbers (chapter 10): small samples throw up extreme results far more often than large ones, and we read causes into them anyway. Two good trial sessions is not evidence, it is a small sample with a story attached.
David Passmore has made a related argument from the talent development side in his masterclass on a research based approach, that selection should move towards repeated performance behaviours rather than one-off impressions, precisely because judging early and judging once tends to reward the player who grew first rather than the player who will still be there at senior level. Kahneman gives you the mechanism behind why that bias is so hard to feel from the inside.
The practical move is small and slightly uncomfortable. Before a confident call, ask what information you do not have, rather than reviewing what you do. Go and find the missing angle, or the second opinion, instead of enjoying how well the first story hangs together.
Think of the last player you described in one word to another coach. How much of that word came from evidence, and how much from the one moment you happened to be watching?
Your gut is right about some things and badly wrong about others
This is the part of the book that is fairest to experienced coaches, and it gets misquoted constantly. Kahneman does not say intuition is worthless. He says it is skilled under two conditions, both of which have to hold: an environment regular enough to be predictable, and enough practice in it with good feedback to have learned those regularities (chapter 22).
Run a coaching career through that filter and it sorts itself quickly. Reading a developing attack, sensing that a press is about to break, knowing from the body language of a defensive line that the next overhead is coming: regular situations, seen thousands of times, with feedback arriving within seconds. Trust that. Predicting which fourteen-year-old will be an international in eight years: an irregular environment, feedback that arrives years later if at all, and no realistic way to have learned the pattern. That gut feeling deserves far less weight than it usually gets, no matter how experienced the person holding it.
Henk Verschuur's masterclass on the cognitive process of coaching lands in similar territory, in that the gap at the top is less often tactical knowledge and more often the coach's ability to manage their own mind under pressure, starting with knowing their own triggers. Kahneman's two conditions give a coach a usable test for the same problem: not "am I experienced", but "is this the kind of question my experience could have taught me to answer?"
Where does your feedback arrive within seconds, and where does it arrive years later? Which of your strongest convictions sit in the second group?
The statistical illusion that makes shouting look like it works
Let's get real for a second, because this one quietly shapes coaching behaviour everywhere.
Extreme performances tend to be followed by more average ones. Not because of character or feedback or a hard word at half time, but because luck contributes to extremes and luck does not repeat (chapter 17). Kahneman tells the story of flight instructors who had worked out, as they saw it, that "occasions on which he praised a performance were likely to be followed by a disappointing performance, and punishments were typically followed by an improvement."
Read that in a hockey context. A coach who tears into a player after their worst game sees improvement next week. A coach who praises a player after their best game sees a drop. Both draw the obvious conclusion, and both are wrong. They have watched regression to the mean and credited it to their own method. Kahneman's own position on the evidence is blunt, that "rewards for improved performance work better than punishment of mistakes."
Two things follow. Judge a coaching method across many repetitions, never by what happens immediately after an extreme performance. And notice that this is the statistical foundation under a lot of what coaches already believe on values grounds. Adam Commens has argued in his masterclass on values based coaching that the difference at the top is a culture where athletes are empowered to bring their best in pressure moments rather than a superior tactical brain. Kahneman does not prove that culture wins matches, but he does remove one of the main pieces of "evidence" people use against it.
The last time a hard word seemed to work, what would you have expected to happen anyway?
The season plan is almost certainly too optimistic
Kahneman is savage about planning, and coaches are not exempt. The planning fallacy (chapter 23) is the finding that plans and forecasts cluster around a best-case path, ignoring how similar efforts have actually gone. He pairs it with optimistic bias (chapter 24), which he rates as the most consequential of the cognitive biases, while being fair enough to note that optimism also fuels persistence and courage.
His fix is the outside view. Instead of forecasting from the specifics of your own plan, put your case in a reference class of similar cases and look at what usually happens. So the question is not "how long will this player's return to full sharpness take, given how well rehab is going", it is "how long has it taken the last ten players in this squad in this situation". The inside view always feels more informative. It is usually less accurate.
The second tool is the premortem, borrowed from Gary Klein and endorsed by Kahneman (chapter 24). Before locking in a plan, gather the staff and say: it is a year from now, this has failed badly. Please take 5 to 10 minutes to write a brief history of that disaster.
It works because it legitimises doubt. Loyalty and group optimism silence the assistant who can see the flaw; assuming failure has already happened gives them permission to speak. Run it on a new game model, a change of press, a season structure, or a promise made to a player about their role. Run it while the decision is still open, not after commitment has hardened.
If your season plan fails, what is the most likely reason, and who on your staff already suspects it?
Losses loom larger, which is why teams play not to lose
Part four of the book is prospect theory, the work with Amos Tversky that the Nobel prize was really about, and one line carries most of the coaching value (chapter 26).
When directly compared or weighted against each other, losses loom larger than gains.
People do not evaluate outcomes in absolute terms. They measure them against a reference point, and a loss hurts more than an equivalent gain pleases. Put eleven players in a tight match with that wiring and you get behaviour every coach recognises: the safe square ball instead of the line-breaking pass, the extra touch, the team that stops playing once it is a goal up.
Two levers come out of this. The first is the reference point, because it is partly yours to set. The same one-goal defeat away from home is a disaster or a step forward depending entirely on what the group was told to expect. The second is framing (chapter 34), where logically identical descriptions of the same situation produce different decisions, because people respond to the wording rather than the underlying reality. "Don't lose it in there" and "take the option that breaks the line" can describe the same instruction and will not produce the same hockey.
This is where Kahneman and the coaches converge most neatly. Noall's rule of praising the decision rather than the outcome, "great decision, unlucky this time", is a deliberate re-framing that pulls the reference point off the result and onto the choice. Andreu Enrich makes a similar point from the player's side in Tips from intelligent players, that the typical spiral starts with a mistake that hurts, and that the way back is simple successful actions and support from teammates rather than a lecture. Loss aversion explains why the spiral is so predictable in the first place.
Listen back to your own touchline language for one quarter. How much of it names what you want, and how much of it names what you are afraid of?
Attention is a budget, not a character trait
One last thread, and it is the one most directly useful for session design. Kahneman treats mental effort as a genuinely limited resource, measurable in something as physical as pupil dilation, and he describes the law of least effort: the mind drifts to the least demanding way of doing things (chapter 2). As a skill becomes automatic it costs less attention, which frees capacity for everything else.
So when a player defaults to the easy option late in a session, that is not always a character problem. It can be a budget problem. Kahneman also argues that self-control draws on the same limited budget as hard thinking (chapter 3), although the specific "ego depletion" research behind that claim has held up poorly in later replication work, so it is worth treating as a plausible idea rather than a settled fact. Either way, scheduling your most demanding decision-making block at the end of a ninety-minute session is asking a lot.
He also has a lovely description of flow as effortless, absorbed concentration in a task well matched to a player's skill, where the task holds the attention rather than the player forcing it. That is a design brief for a drill: optimally challenging, not chaotic, not trivial. It is also the argument Tin Matkovic makes about scanning, that a player who understands why they are looking will keep doing it, while a player following an instruction will not. And it sits behind Andreu Enrich's emphasis on players regaining confidence through simple successful actions rather than being talked at, because the simple successful action is what puts attention back within budget.
Look at your last session plan. Where did you put the hardest thinking, and was anyone still capable of it by then?
So what was all that?
So where does this leave you? A big, difficult book, reduced to five things a coach can actually act on.
First, two systems. Train fast reads by repetition in game-like pictures, and protect slow thinking from fatigue and hurry, because a tired System 2 simply endorses whatever the gut offered.
Second, confidence is not evidence. Before a firm judgement on a player, ask what is missing rather than admiring how well the story fits. Halo, availability and small samples are all the same error in different clothes.
Third, trust intuition only where the environment is regular and the feedback has been good. Reading a press, yes. Predicting a fourteen-year-old's ceiling, no.
Fourth, watch out for regression to the mean, because it will happily persuade a coach that harshness works and encouragement backfires when neither conclusion is earned.
Fifth, plan from the outside view, run a premortem before committing, and remember that losses loom larger than gains, so the way a target or a setback is framed will shape the hockey you get.
None of this makes anyone a better coach on its own. What it does is slow down the moment before a confident decision, which, reading the book with coaches in mind, seems to be the whole point.
Try this one thing
Of everything above, the premortem is the only tool you can run this week with no pitch, no kit and no budget. It takes ten minutes and it works best on the decision you are closest to committing to, whether that is a change of press, a season structure, or a promise you are about to make to a player about their role.
Name the decision out loud, exactly as it stands now.
Tell your staff it is a year from now and the decision has gone badly wrong.
Give everyone five minutes, in silence, to write the history of that failure.
Collect the stories and look for the reasons that appear more than once.
Fix the two most plausible risks before you commit, not after.
The reason it works is not the list. It is that assuming failure gives the assistant who could already see the flaw permission to say it. If you run it, we would genuinely like to know what came out of the room that would not have come out otherwise.
