The Mental Game

Overview

This document is about the part of hockey that happens between your ears: resetting after a mistake, playing when you are nervous, staying confident through a bad run, and being useful to your team on a night when nothing works. It is written for a player, not for a coach or a therapist, and every technique in it is something you can do on a Tuesday night beer-league bench.

It is deliberately conservative about what it claims. Sports psychology has real findings and it also has a large amount of confident-sounding folklore, and the two are hard to tell apart from the outside. Where the research supports something, this document says so and names the work. Where the evidence is thin or contested, it says that too. Where an idea is coaching craft — widely taught, sensible, untested — it is labelled as coaching craft and you should treat it as a suggestion rather than a finding.

---

How claims are labelled in this document

Three phrases are used consistently, and they mean specific things.

• "The research supports this." There is published, peer-reviewed work — usually a meta-analysis or a well-replicated finding — behind the claim. The named study is in the Sources section.
• "The evidence here is thin" or "contested." The idea is popular, and it may well be true, but the studies are few, small, mostly done outside sport, or the field disagrees. Use it, but don't build your season on it.
• "This is coaching craft." It is what experienced coaches and players do. Nobody has tested it properly. That is not a reason to ignore it — most of hockey is coaching craft — but it is a reason not to present it to a teammate as a scientific fact.

No statistics in this document are invented. Where a number appears, it comes from a named study and is cited.

---

1. Why the mental side is not optional

Hockey is unusually unforgiving of a wandering mind

Three features of the game make attention more load-bearing here than in most sports.

It is fast and it never stops. There is no set piece, no serve, no huddle. Between whistles the situation changes completely every two or three seconds, and there is no moment where the game waits for you to catch up. If you spend four seconds thinking about the pass you just missed, you have missed roughly two more decisions.

Your mistakes are public and immediate. A misread in football might cost a yard. A blown coverage in hockey is a puck in your net thirty seconds later, with your name attached to it in front of everyone in the building including your own bench. There is nowhere to hide, and the feedback arrives before you have finished processing the error.

You are tired when it matters. A shift is short — the measured range in the research literature is roughly 30 to 80 seconds, with a well-managed shift nearer 30 to 45 (Vigh-Larsen and Mohr's 2024 physiology review; see Conditioning and Recovery for the full numbers) — precisely because you cannot maintain the required output for longer. But the last ten seconds of a shift, when your legs are gone, is also when your decision-making gets tested hardest. Mental discipline is what stands in for judgment when you are too tired to think clearly. See Conditioning and Recovery for the physical side of that.

The honest version of "it's all mental"

You will hear that the gap between your best game and your worst game is entirely mental. That is an overstatement, and stating it as fact does the idea a disservice. Here is the defensible version.

Your physical capacities — top speed, shot velocity, edge quality, strength — are close to fixed over the span of a week. They improve over months of training and they degrade with fatigue and injury, but they do not swing much between Saturday and Sunday. Your attention, your decision speed, your willingness to take a hit to make a play, and your composure after a turnover can swing enormously between Saturday and Sunday. So when you play badly on a night when you are healthy and rested, the explanation is more likely to be in the second list than the first. That is a reasonable inference, not a measured finding, and it is why mental skills are worth training.

What the mental side cannot do. It will not make you faster, it will not fix a bad first step, and it will not turn a wrist shot with no weight transfer into a good one. Anyone selling mental training as a substitute for skill work is selling you something. The mental game closes the gap between your worst performance and your best. It does not raise your best.

---

2. The shift reset

This is the single most valuable mental skill in hockey, and the one most worth practising deliberately.

The problem

You make a mistake. You lose a battle on the wall, you fan on a one-timer, you get beaten wide. The play does not stop. Within two seconds the puck is somewhere else and you have a new job. If you are still holding the last two seconds, you are not doing the new job.

Then the shift ends and you sit down, and this is where the real cost lands. Rumination on the bench is not a neutral activity. The mechanism here is a theory, not a measured finding, and none of it was tested on hockey players — treat it as a plausible account: attentional control theory, stated primarily by Eysenck, Derakshan, Santos and Calvo (Emotion, 2007), holds that "adverse effects of anxiety on processing efficiency depend on two central executive functions involving attentional control: inhibition and shifting." In plainer words, anxiety consumes the central executive's capacity, specifically degrading the inhibition function (blocking task-irrelevant material) and the shifting function (moving attention where it needs to go). A brain busy re-running a turnover has less capacity available for the things the bench is for — watching what the opposition's defence pair does under pressure, hearing the coach's adjustment, tracking who you are coming out against. So one bad shift becomes two.

The reset itself

The technique below is coaching craft in its specific form — no study has tested this exact four-step sequence — but each component has independent support and the whole thing is close to universally taught. Build it as a fixed sequence so it runs without deliberation.

One: a physical trigger. A single, specific, repeatable movement. Tap both shin pads with your stick. Squeeze the top of the shaft hard once and release. Adjust your right glove. It does not matter what it is; it matters that it is always the same thing and that it is physical. The point is to give the reset a definite start, so that resetting is an action you perform rather than a mood you wait for.

Two: one breath out, long. Not a deep breath in — a long breath out. Slow exhalation is the part that increases cardiac vagal activity, which is the physiological signature of coming down from a spike. The research supports the physiological effect: Laborde and colleagues' systematic review and meta-analysis (Neuroscience & Biobehavioral Reviews, 2022), pooling 223 studies, found that voluntary slow breathing increases vagally-mediated heart rate variability — during the breathing, immediately after a single session, and after multi-session interventions. Note what that does and does not cover: it is HRV in general populations, not state anxiety, and not athletes specifically. The anxiety claim sits a tier lower — the evidence here is thinner than the physiology: the athlete-specific studies cited below are individual trials, not a meta-analysis. And the evidence for a direct performance benefit is thin — reviews are explicit that no definitive data establish that breathing techniques improve sport performance itself. So use it to change how you feel, not because it will make your next pass better.

Three: a cue word. One or two words, said internally, that point at what you are about to do — not at what you just did, and not at how you should feel. "Next." "Feet." "Inside." "Wall." The research supports self-talk in general: the standard meta-analysis (Hatzigeorgiadis and colleagues, 2011, covering 32 studies and 62 effect sizes) found a positive moderate overall effect, ES = 0.48.

Two of its moderator findings matter for how you use this, and the first is routinely misreported.

• It supports a matching hypothesis, not blanket instructional superiority. Instructional self-talk beat motivational self-talk only for tasks with relatively fine motor demands. The paper reports that advantage for fine tasks only; it reports no corresponding advantage for gross motor tasks, and it does not report which way that comparison leaned. So the honest statement is that instructional superiority is unestablished for gross tasks, not that motivational self-talk wins there. Self-talk interventions overall also worked better on fine tasks than gross ones. So match the cue to the job: instructional for the precise things — a shot placement, a faceoff hand position, a saucer pass through a lane — and motivational for the gross, effortful things — a hard forecheck, a puck battle, the last twenty seconds of a shift. "Get the stick on the ice" is an instructional cue attached to a gross-motor task, and the meta-analysis does not show it beating "come on, work."
• The effect was larger for novel tasks than for well-learned ones. Neither kind of cue will do much for a skill you have already automated.

Four: name your next action, not your last one. Before your skates hit the ice again, you should be able to say what you are doing on this shift in one concrete sentence: "First forechecker, take the D's inside shoulder." "Get above the puck." "Hard on the wall, don't get beaten to the middle." This is the whole reset in one move; the previous three exist to make it possible.

Do not try to suppress the thought

The instinctive move — deciding not to think about the mistake — tends to be counterproductive. Deliberately suppressing a thought is effortful, the effort itself consumes attentional capacity, and the thought often rebounds. The acceptance-based alternative is to notice it, let it be there, and move your attention to the next action anyway. This is the core of the Mindfulness-Acceptance-Commitment (MAC) approach.

The evidence for MAC is partial, and the part that supports "acceptance" specifically is the weakest part — which is worth stating plainly, because MAC is usually sold as a package. A robust Bayesian meta-analysis of eleven controlled trials (Int J Sport Exerc Psychol, 2023) found:

• Mindfulness: g = 0.64 (95% CI 0.00–0.91), with strong evidence. MAC does what it says on developing mindfulness.
• Sport performance: g = 0.37 (95% CI 0.00–0.79), with moderate evidence. A small effect, and the authors put it no higher than "might be effective."
• Experiential acceptance: g = 0.36 (95% CI 0.00–0.92), with only anecdotal evidence.

Note the confidence intervals: every one of them has a lower bound of 0.00. The point estimates are the headline; the intervals say that on eleven trials none of these effects is separated from zero.

• Psychological flexibility: not supported. The authors' own conclusion is that "Currently available trials, however, do not provide sufficient evidence supporting the MAC protocol's effectiveness in developing psychological flexibility, experiential acceptance, and promoting well-being among athletes."

So: the acceptance stance below is worth using, and it is better than suppression, but file it under "the evidence here is thin" rather than "the research supports this." A systematic review in the British Journal of Sports Medicine has separately examined mindfulness-based programmes for elite athlete mental health, which is a different question again.

The practical translation: "I made a bad read, and I am now going to win this wall battle" works better than "stop thinking about the bad read."

What to do on the bench

You have three jobs on the bench and none of them is post-mortem.

1. Recover. Get your breathing down. See Conditioning and Recovery.
2. Watch. The bench is the best seat in the rink for reading the game — which defenceman panics under a forecheck, whether their centre is cheating for offence, which of their forwards has stopped backchecking.
3. Plan one thing. Decide the single thing you are doing on your next shift, in the concrete form above.

Analysis of the mistake is a post-game job. It is genuinely valuable then, and it is genuinely harmful during. If you cannot let something go, park it explicitly: "I'll look at that after." Some players keep a short note on their phone in the room afterwards. That is coaching craft, but it is cheap and it works for a lot of people because it converts an intrusive thought into a scheduled task.

---

3. Playing to a process rather than an outcome

The distinction

An outcome goal is about a result you do not fully control: score tonight, win, get a point, don't get scored on. A process goal is about an action you fully control: keep the gap through the neutral zone that Defending the Rush sets out — two to three stick lengths, tightening to a stick and a half at the red line — get your first touch away from pressure, shoot within half a second of receiving in the slot, finish every check.

You cannot control whether the puck goes in. Between the moment the puck leaves your blade and the moment it hits the goalie's chest, there is a screen you did not place, a deflection you did not cause, and a save that is somebody else's skill. You can control your route to the puck, your gap, your first touch, your shot selection and whether your feet are moving.

Why outcome focus makes you more inconsistent

Three proposed mechanisms. The first is plain reasoning, the second has a reasonable evidence base, and the third — flagged in place — has recently been undermined.

Outcomes are noisy, so outcome-based confidence is noisy. If your self-assessment is "did I score," then a night where you generated four good looks and hit two posts registers as a failure, and your confidence takes damage it has not earned. Process-based self-assessment is stable because the input is stable.

Outcome focus invites self-monitoring, and self-monitoring breaks automated skills. The research supports this: the leading account of choking under pressure is the self-focus or explicit-monitoring theory — pressure raises self-consciousness about executing correctly, which pushes attention onto processes that normally run outside working memory, turning a fluent skill into a slow, deliberate one. The primary experimental support is Beilock and Carr (Journal of Experimental Psychology: General, 2001), who showed that choking occurred in golf putting — a proceduralised skill — but not in a non-proceduralised arithmetic task, and that it was eliminated by self-consciousness training while dual-task training did nothing. They describe their results as supporting "explicit monitoring theories of choking and the popular but infrequently tested belief that attending to proceduralized skills hurts performance." Note their own phrase, infrequently tested: this is a well-regarded account with a real experiment behind it, not a large replicated literature. Worrying about the result makes you watch yourself shoot, and watching yourself shoot is how you shoot badly.

Outcome focus points your attention inward, and inward is the wrong direction. This one is contested, and it recently became much more contested. The claim is that an external focus of attention — on the intended effect of the movement — produces better motor performance and learning than an internal focus on the body part doing it. It rests on a large body of work reviewed by Gabriele Wulf, and on a 2021 meta-analysis in Psychological Bulletin (Chua and colleagues) that supported it. Note what that meta-analysis pooled: 73 performance studies and 40 learning studies drawn from clinical and non-clinical populations and from children as well as adults, with health status, age group and skill level tested as moderators and none of them found to moderate the effect. So it is not an athlete-specific result, and the breadth of the pool is part of why its precision looks better than the underlying hockey-relevant evidence really is.

But in 2024 McKay and colleagues published a robust Bayesian re-analysis of the same literature, also in Psychological Bulletin, and found "moderate to strong evidence of publication bias for all analyses." Once they corrected for that bias, the estimated effects collapsed to near zero — g = 0.01 for performance, 0.15 for retention, 0.09 for transfer — with the Bayes factors favouring the null in every case. Their conclusion was that focus of attention has varied effects that "on average" are "small to nil."

How to treat this. External cueing is free, it costs you nothing to phrase a cue as "drive the ice away" rather than "extend your leg", and there is no evidence it hurts. But it belongs in the contested tier, not the top one, and you should stop hearing it presented as settled. If an internal cue is the only way you can feel a correction, use the internal cue.

Wording your cues externally

Same instruction, two wordings. The right-hand column is the one conventionally recommended — with the caveat above that the evidence for it is weaker than it is usually presented as being.

The pattern: point the cue at the puck, the ice, the target or the opponent, not at your own limbs.

Building your own process list

Pick three process goals for the season, no more, and make them specific to your role.

• Defenceman: maintain gap through the neutral zone; first touch on retrievals goes to the wall, not the middle; get shots through from the point rather than looking for the perfect lane.
• Winger (low zone collapse / 2-1-2): arrive at the net front on every entry; support the wall battle rather than waiting at the blue line; first pass out of your own zone is up the boards.
• Centre (low zone collapse): be the lowest forward in your own zone; win the puck-side draw or tie it up; get above the puck on every turnover.

Score yourself on those three after the game — but do not throw the scoresheet away entirely, and the reason is in the research below.

What the evidence actually says — and it is not "process goals only". Two studies bear on this directly.

Process goals do have measured support. Kingston and Hardy (1997, The Sport Psychologist 11:277–293) ran two goal-setting training programmes on 37 club golfers, one process-oriented and one performance-oriented, and measured them at three competitions. They report significant Group × Test interactions (p < .05) for self-efficacy, cognitive anxiety control and concentration, which they say "provide further evidence for the positive impact of process goals in competitive situations." Note the comparison, because it is narrower than the claim it is often recruited for: this is process against performance goals, not process against outcome goals.

The study that did test the full trichotomy points somewhere more interesting. Filby, Maynard and Graydon (1999, Journal of Applied Sport Psychology 11:230–246) split 40 participants into five ability-matched groups on a soccer task — four using different combinations of outcome, performance and process goals, one control — over a five-week training period and then a competition. Group differences were significant (p < .05) in both. Their conclusion: "The superior performance of the groups using multiple-goal strategies provided evidence to support the efficacy of maintaining a balance between the use of outcome, performance, and process goals." Their opening premise is worth reading as well — sport psychologists have pushed athletes toward "only self-referenced performance and process goals", while studies of successful performers find they "do in fact make effective use of outcome goals."

So the practical position, adjusted to fit that: process goals are the ones to score yourself on, because they are stable, controllable and usable, and because they are what improves under training. But the evidence does not support deleting outcome goals — it supports keeping them in the mix and not letting them be the thing you judge a night by. Both studies are small (37 golfers, 40 participants on a wall-volley task), neither is in hockey, and neither measures a season.

One citation nearby that does not support any of this: Kyllo and Landers' meta-analysis of goal setting in sport and exercise (1995, Journal of Sport and Exercise Psychology 17:117–137) is the standard reference — 36 studies, an overall benefit of goal setting of 0.34 of a standard deviation — but the moderators it headlines are goal difficulty (moderate goals did best), absolute versus relative goals, and time frame (combined short- and long-term goals did best). It does not report a process-versus-outcome comparison at all. Do not cite it for one.

---

4. Confidence

Where it actually comes from

The most useful model is Albert Bandura's four sources of self-efficacy, and its ranking matters more than the list.

1. Mastery experiences — by a distance the strongest source. Having actually done the thing. Not having imagined it, not having been told you could; having done it.
2. Vicarious experience. Watching someone comparable do it. Weaker than direct experience, but genuinely useful when you are very unsure of yourself — which is exactly when a demonstration from a teammate at your level beats one from the best player on the ice.
3. Verbal persuasion. Being told. Weakest of the three, and note the asymmetry Bandura's work identifies: discouragement lowers self-efficacy more effectively than encouragement raises it. Which is why one dismissive comment from a coach outweighs five compliments, and why what you say to a struggling teammate matters less than what you avoid saying.
4. Physiological state — and specifically how you interpret it, not the state itself. Your heart rate before a big game does not determine your confidence; whether you read it as "I'm ready" or "I'm falling apart" does. This is the hook for the reframing technique in section 5.

The practical consequence is blunt: confidence is downstream of preparation and evidence. Self-talk helps (ES = 0.48, moderate), but self-talk with no bank of mastery experiences behind it is a thin layer over nothing. If you want to feel confident taking a puck off the wall under pressure, the reliable route is to have taken two hundred pucks off the wall under pressure in practice. See Practice and Development.

How confidence collapses

Confidence rarely erodes evenly. What usually happens is that it collapses in one narrow area and then leaks outward. You fan on a shot, so you stop shooting. You stop shooting, so you handle the puck less in the offensive zone. You handle it less, so your touch goes. Now you are not just a player who missed a shot; you are a player who is bad at hockey.

Two things make this worse. Fatigue — a tired brain rates everything more negatively. And comparison — watching the player who took your spot in the lineup is a reliable way to feel worse without learning anything.

The hockey version: the scoring slump

This is the classic, and it is worth spelling out because the trap is invisible from inside it.

A forward stops scoring. The response feels obvious: be more careful, wait for a better look, don't waste chances. So they shoot less. Shooting less means fewer chances, fewer rebounds, fewer deflections and fewer of the ugly goals that break slumps. It also means the sample gets smaller, so the drought lasts longer in games even if the underlying rate has not changed. And it means they are now carrying the puck into contact looking for the perfect play instead of releasing it, which produces turnovers, which further damages confidence.

The direction of this loop is coaching craft rather than a measured finding — no published study establishes that slumping players shoot less, and there is no honest number to attach to it. But the reasoning is sound, the pattern is universally described by coaches and sport psychologists working with hockey players, and the prescription follows either way.

Rebuilding after a bad run

Shoot more, not less. Set a process target — five shot attempts a game, whatever suits your role — and hit it regardless of whether anything goes in. You are deliberately rebuilding the sample. See Shooting.

Change the unit of success. If "goal" is the only success available to you, you will get almost none of them. Add units you can actually collect: a puck retrieved with pressure on you, a clean zone entry, a shift that ends in the offensive zone, a battle won on the wall. This is the same move as section 3, applied under duress.

Contribute in the other currencies. A forward who is not scoring can be the best forechecker on the team tonight, can be first back, can win draws, can be physical. This is genuinely valuable to your team, and it also rebuilds mastery experiences in areas where the feedback is immediate — which is the fastest route back, given Bandura's ranking.

Rebuild the evidence bank. Confidence responds to evidence, so manufacture some. In practice, do reps of the exact thing you have lost — not general shooting, the specific release you fanned on — until you have twenty good ones. That is mastery experience in the technical sense and it is the strongest lever you have.

Do not add more analysis. The instinct when nothing works is to think harder about it. Given the self-focus account of choking, that is precisely the wrong direction: more explicit monitoring of a skill you have already automated makes it worse. Do the technical work in practice, and play on instinct in games.

---

5. Nerves and arousal

What the inverted-U actually says — and what it does not

You will be told, confidently, that performance follows an inverted U against arousal: too little and you are flat, too much and you fall apart, somewhere in the middle is your peak. The picture is useful. The way it is usually presented is not accurate, and it is worth being precise about why.

Where it comes from. Yerkes and Dodson published in 1908. The subjects were mice — a Japanese species nicknamed the "dancing mouse" — being trained on a discrimination task with electric shock. The paper was cited roughly ten times in the following fifty years, and several of those citations were already calling it a law.

What the original actually distinguishes, and almost nobody repeats. The relationship depends on task difficulty. Simple, well-learned tasks tend to improve roughly linearly with arousal. Only complex or unfamiliar tasks show the inverted U. The widely reproduced textbook curve is a later, Hebbian version that drops this distinction, and it is described in the literature as frequently miscited.

What is genuinely contested. The correlation is real enough. The causal mechanism has not been established. Flow theory and reversal theory both offer competing accounts. And the sharpest criticism for our purposes is that the same physiological state gets interpreted completely differently by different athletes, so a single universal curve is too generalised to be actionable.

What this means for you on a hockey bench. Do not think "there is an optimal arousal level and I should be in the middle." Think instead:

• Your optimal state is individual, and it may not be calm. This is the core of Yuri Hanin's Individual Zones of Optimal Functioning (IZOF) model, developed over roughly four decades of research, which replaces the universal curve with individually determined optimal emotion zones. Some players genuinely play best furious. Some play best serene. Find yours empirically — see below.
• It also depends on what you are about to do. Simple, physical, automated tasks tolerate high arousal well. A hard forecheck, a shot block, a puck battle: crank it up. Fine, complex, novel tasks do not. A shootout attempt, a first-time power-play look you installed on Tuesday, a delicate saucer pass through a lane: you want to be lower.

The catastrophe model — the version that best matches how hockey actually falls apart

The catastrophe model — Fazey and Hardy (1988), The Inverted-U Hypothesis: A Catastrophe for Sport Psychology, British Association of Sports Sciences Monograph No. 1 — is a better description of a bad period than the inverted U. The evidence for it is thin, though, and it is routinely overstated. The study usually cited is Hardy and Parfitt (1991), British Journal of Psychology 82:163–178, which manipulated cognitive anxiety and physiological arousal independently and found the pattern the model predicts. Its sample was eight basketball players. By this document's own rule — the top tier is reserved for a meta-analysis or a well-replicated finding — eight participants does not qualify. Use the model as a useful mental picture of how a period falls apart, not as an established mechanism.

Two claims matter.

Under high cognitive anxiety — worry, specifically, as distinct from physical arousal — performance does not decline gently past the optimum. It falls off a cliff. In the high-anxiety condition, maximum performance was higher, minimum performance was lower, and the drops were far larger. That is the period where your team concedes three in six minutes.

Recovery requires more than backing off slightly. The model involves hysteresis: once you have gone over the edge, getting back does not happen by returning to the arousal level you were at when you fell. You have to come down well below it. Practically, this is the argument for the thirty-second timeout — under NHL Rule 87.1, each team is permitted one thirty-second timeout per game — and for a coach deliberately shortening the bench, changing lines, or having the goalie freeze a puck. It is also the argument for you personally taking a full, deliberate reset rather than "calming down a bit."

Nerves versus anxiety

These are not the same thing and conflating them is unhelpful.

Nerves are the physiological ramp-up before something that matters: heart rate up, breathing quick, hands unsettled, stomach hollow, needing the toilet. They arrive roughly on schedule, they are proportionate, and they subside once play starts — usually inside the first shift. They are the body preparing to work hard. They are a good sign, and top players describe them across every sport.

Anxiety in the performance sense is the cognitive component: worry, catastrophising, rehearsing failure, "what if I'm terrible tonight." It does not subside at puck drop. It is what the catastrophe model treats as the dangerous variable, and it is what attentional control theory says eats your capacity to inhibit distraction and shift focus.

Two consequences. Nerves need managing, not eliminating — a player with no nerves before a big game is often flat. Anxiety needs addressing. And persistent anxiety — the kind that is there on Wednesday about Saturday, or that has stopped being about hockey — is section 13.

Practical regulation

Reframe the arousal rather than fighting it. The research supports this and it is one of the more robust findings here: arousal reappraisal — being told that the physical signs of stress are your body mobilising resources to help you, not evidence that you are failing — improves performance and produces a healthier physiological stress profile. Jamieson and colleagues have shown this repeatedly, though largely in academic-testing contexts rather than sport, so the transfer to a rink is an inference. It also fits Bandura's fourth source exactly: it is the interpretation of the physiological state that moves self-efficacy, not the state.

The move is small and concrete. Standing on the blue line for the anthem with your heart going: "That's my body getting ready" rather than "I'm nervous." Same sensation, different label.

Breathe out longer than you breathe in. A workable pattern is in for four, out for six, six to ten cycles. The long exhale is the active ingredient. As noted above, the anxiety-reduction and vagal effects are supported; a direct performance effect is not established.

Use the warm-up to set your level, in either direction. Most players think of a warm-up as raising arousal. It works in both directions and you should know which one you need. If you are flat: short hard bursts, quick feet, loud communication, hard shots. If you are over the top: longer smooth skating, easy passes, deliberately slow breathing, less talking.

Learn your own zone empirically. This is the practical form of IZOF and it takes about eight games. After each one, note in two lines how you felt beforehand — flat, edgy, calm, angry, distracted — and how you actually played. Patterns show up fast, and they are often surprising. Some players discover they play best mildly irritated. Some discover their "focused and calm" nights are their worst.

---

6. The pre-game routine

What a routine is for

A routine exists to make your state consistent, so that the only variable left is the game. If you arrive in a different frame of mind every week, your performance will vary for reasons that have nothing to do with the opposition.

The honest evidence position: routines work; what nobody can tell you is which bits of them are doing the work. Those are two different claims, and they are very easy to run together.

The research supports the effect. Rupprecht, Tran and Gröpel's meta-analysis (International Review of Sport and Exercise Psychology, published online 2021, in print 2024), pooling 112 effect sizes, found significant positive effects of pre-performance routines on sport performance: SMC = 0.31 (95% CI 0.18–0.44) in pre–post designs, and a moderate-to-large Hedges' g = 0.64 (95% CI 0.45–0.83) in experimental designs under low pressure, with a comparable g = 0.70 under pressure — though that last one carries a much wider interval, 95% CI 0.24–1.16, so the pressurised estimate is far less precise than the other two. On skill level, be careful how you read what the paper says: the effects were "not moderated by the type of PPR, age, gender, skill level, or intervention characteristics." That is a null moderator result — no difference was detected — not a positive demonstration that novices and elite athletes benefit equally. It is a reason not to assume routines are only for experts, not evidence that they work identically for everyone. The main effect is a genuine finding and you should build a routine on the strength of it.

What is genuinely unsettled is the anatomy. Cotterill's earlier 2010 review in the same journal concluded that understanding of routines "falls somewhat short of that required to recommend with confidence," and the mechanism question — which psychological components matter, and how a routine should be taught — is still open. So: build one, and do not believe anyone who tells you exactly which elements are essential.

Routine versus superstition — one test

They look identical from outside. The test is simple: if you cannot do it, does it wreck you?

A routine is a sequence of actions you perform because of what they do — a warm-up that raises your body temperature, breathing that settles you, imagery that primes the patterns. If the room is different or you arrive late, you compress it and you are fine.

A superstition is an action you perform because of what you believe will happen if you don't. Same socks, same tape job, same order of dressing, tapping the goalie's pads a fixed number of times. Harmless in itself — but you have made your composure dependent on something outside your control, and one day the sock will be in the wash.

The practical rule: keep the routine, prune anything that would break you if it were unavailable. If you play away from home, in a rink with a different room layout, on an unusual schedule, your routine must still run.

A workable timeline

Times are relative to puck drop. The specific numbers are coaching craft; adapt them to your league and your travel.

The night before

• Kit packed and checked. Skates sharpened if needed. See Equipment.
• Sleep. This has a stronger evidence base behind it than anything else in this section — see Conditioning and Recovery.
• No tactical cramming. Reviewing systems at 11pm the night before does not install them.

Three hours out

• Last full meal.
• If you use imagery, this is a good slot for the longer version.

Sixty to ninety minutes out — arrival

• Arrive at the same point in the timeline every week if you can. Rushing is the single most common self-inflicted arousal problem.
• Dress in the same order. Not superstition — one less decision.

Twenty to thirty minutes out — off-ice warm-up

• Raise body temperature. Dynamic movement, not static stretching.
• Hands: puck on a stick, passing off a wall, stickhandling.
• This is where you deliberately set your arousal up or down per section 5.

On-ice warm-up

• Skate to actual game speed at least twice — a warm-up entirely at 60% teaches your body that tonight is a 60% night.
• Goalies: track pucks properly from the first shot. See Goaltender.

The last sixty seconds

• Two or three slow exhales.
• One thought about the game, and make it a process cue. Not "we need this one." Something like "first pass up the wall" or "gap early."

Between periods

• Same principle, compressed: recover, one adjustment, one process cue.
• Do not spend the intermission on the referee.

Imagery, if you use it

The research supports imagery, and the best-supported framework for making it work is the PETTLEP model (Holmes and Collins, 2001), which has been examined in systematic reviews and has now had twenty years of development. The framework's core insight is that imagery works better the more closely it matches the real thing: do it in your kit, in a realistic posture, in real time rather than slow motion, from your own eyes rather than a television angle, including how the movement feels and not just how it looks, and with the emotion of a real game attached.

Keep it short and specific — two or three minutes on situations you will actually face tonight. "Winning a puck off the left wall against a hard forecheck" is useful. "Scoring a hat-trick" is daydreaming.

---

7. Focus and attention

Four kinds of focus, and hockey needs all of them

Robert Nideffer's model, which remains the standard framework, splits attention along two dimensions: width (broad versus narrow) and direction (external versus internal). That gives four styles, and hockey demands all four within a single shift.

The skill is not being focused. The skill is switching. A defender who stays narrow-external on the puck carrier through a three-on-two has stopped seeing the trailer. A forward who stays broad-external as the pass arrives will not catch it. The transitions are the game.

Attentional cues

A cue is a very short instruction you give yourself to force the right focus at the right moment. Rules for making them work, following the research above:

• Short. One or two words. Anything longer arrives too late.
• External, as a default. Point at the ice, the puck, the target or the opponent rather than at your own body. Wulf's work argues this strongly; McKay and colleagues' 2024 bias-corrected re-analysis argues it barely matters (see section 3). It is free, so do it — but don't treat an internal cue as an error.
• Match the cue to the task. For fine, precise actions — placing a shot, a faceoff grip, a delicate pass — use an instructional cue: "far post", "stick down". For gross, effortful actions — a forecheck, a battle, a shift block — a motivational cue is at least as good: "go", "work". This is what the self-talk meta-analysis actually found; instructional superiority holds only for fine tasks.
• Tied to a trigger, so you do not have to remember to use it: over the boards, at a whistle, as the puck goes behind your net.

You cannot hold high focus for sixty minutes and nobody does. What good players have is reliable trigger points — moments where focus snaps back up — rather than continuous concentration.

Puck-watching is a focus failure, not a laziness failure

Puck-watching is standing still, eyes locked on the puck, with an unmarked opponent behind you. It is the defensive error named most often at every level below professional — a coaching observation rather than a counted ranking — and it is almost always diagnosed as a want-to problem. It is not. It is an attentional one, and understanding that is what lets you fix it.

The puck is the brightest object in the environment. It moves fast, it is the thing everyone is shouting about, and it is where the danger appears to be. Your attention system is doing exactly what it evolved to do by locking onto it. The failure is that hockey requires you to voluntarily override that pull several times a shift and check the space you cannot see.

This maps directly onto the model above: puck-watching is getting stuck in narrow-external when the situation requires broad-external. And note that anxiety makes it worse — attentional control theory holds that anxiety specifically degrades the shifting function, which is exactly the function you need to break your gaze off the puck. A nervous player puck-watches more.

The fix is a trained habit of shoulder checks tied to specific triggers, not an instruction to "concentrate." That habit, and how to build it, is the subject of Scanning and Anticipation, which you should read alongside this section. See also Playing Without the Puck.

---

8. Fear

This section is written seriously because fear is common in hockey and almost never discussed, and the culture of the sport makes admitting to it difficult. No prevalence figure is given here, because there is no published estimate — and you should distrust the ones you hear, including the common claim that "something like half" of an adult rec dressing room has a version of it. What is safe to say is the qualitative version: coaches and players describe it constantly, and it is not rare. If you are afraid of something on the ice, you are not unusual, you are not soft, and it is not a character defect. It is a normal response to a fast game with hard surfaces, and it is workable.

Four fears, four different problems

Fear of contact — of being hit, or of hitting. Very common in players who came to hockey late, in players returning after time away, and in anyone who has just moved up a level. Note that in most adult rec and beer leagues body checking is not permitted at all, but body contact — battles, pins, incidental collisions — is unavoidable in every version of the game.

Fear of injury — usually specific and usually earned. Someone who has had a bad shoulder is not afraid of hockey; they are afraid of that shoulder. It is often at its worst on the first few returns after an injury has actually healed.

Fear of the puck — flinching on shot blocks, turning away from a point shot, closing your eyes on a deflection, or simply not wanting to be in the slot when the puck is coming there. Common in skaters, and its goalie version is its own topic.

Fear of making a mistake — the most damaging of the four, and the least visible, because it does not look like fear at all. It looks like a careful, quiet, tidy player who never does anything wrong.

Why fear of mistakes is self-defeating

The loop is worth understanding because it is completely counter-intuitive from inside it.

You are afraid of making a mistake, so you play safe. Playing safe means chipping the puck out rather than making a play, refusing the pass in a tight spot, hanging back rather than pinching, taking the extra half-second to be sure. Now: hesitation is itself a mistake, and it is the mistake hockey punishes hardest. The extra half-second is exactly how long a forechecker needs to arrive. The refused pass leaves your teammate with no options. The safe chip goes straight back into your zone. So the passive player makes more turnovers, not fewer, and each one confirms the fear.

There is a second cost that shows up over months rather than shifts. A player playing not to lose the puck stops attempting the plays that would improve them, so their development stalls, and their gap to the rest of the team widens, which feeds the fear again.

The prescription is not "be brave." It is to convert the vague fear of "making a mistake" into a specific process target you can succeed at — see section 3 — and to accept a defined error rate as the price of playing properly. Coaches worth playing for want mistakes of commission, not omission. If yours does not, that is useful information about your coach, not about you.

Working on physical fear: graded exposure

The general approach that works for fear — in and out of sport — is graded exposure: repeated, controlled, progressively harder contact with the feared thing, at a level where you stay in control, until it stops producing the response. The clinical evidence base for graded exposure in anxiety generally is strong; its specific application to hockey contact fear is coaching craft and is not cited to a study anywhere in this document. It does map onto the progression-based checking curricula that Hockey Canada and USA Hockey both teach, and both of those can be quoted directly. Hockey Canada's Checking Resource Guide is explicit that "checking is a 4-step progression that begins the first time a young player steps on the ice" — Step 1 positioning and angling, Step 2 stick checks, Step 3 contact confidence and body contact, Step 4 body checking — and that "body checking is the fourth and final step of a four step teaching progression." USA Hockey's Introduction To Body Contact uses the same shape as a pyramid, with angling as the "1st line of Body Play", stick checks second, body blocks third and body checks fourth, and adds an age-group progression that only introduces body checking at 12-and-Under. That is the same logic as graded exposure — start where you stay in control, and move up only when the level below is boring — but note what it is: a governing body's teaching sequence for a skill, not a tested treatment for fear.

For contact. Work up in deliberate steps rather than throwing yourself into a scrimmage. Stationary leaning and pushing against a partner. Then angling and steering along the boards at half speed. Then competitive one-on-one wall battles. Then the same at game pace. Each level should feel manageable before you move up, and you should stay at each one long enough to be bored by it. Details of the technique itself are in Body Contact and Battles.

Technique reduces fear more than courage does. A large part of fear of contact is a well-founded fear of being hurt because you do not know how to take a hit. Learning to see it coming, get your weight low, keep your feet moving, and — going into the wall — get your skates parallel to the boards and take it on the forearm and hip genuinely lowers the risk. (Note the distinction, because getting it backwards is dangerous: you turn your shoulder toward the checker, never into the boards — USA Hockey's manual says never to hit the boards with the point of your shoulder. Full treatment in Body Contact and Battles, which owns this.) The fear then drops for a good reason.

For the puck. Same principle, escalating: tennis balls, then pucks at reduced speed, then real shots from distance, then closer. Practise shot blocking properly, and note that "properly" is close to the opposite of what instinct suggests: square to the shooter, on your feet, close, presenting your shin pads, hands tucked with the backs of the gloves out — never sideways, never sliding, never lifting a leg, because all three expose the parts of you that have no armour and turn you away from the puck. Body Contact and Battles owns the technique and the equally important list of when not to block. A player who has never been taught to block a shot is right to be afraid of blocking one.

Check your equipment. This is unglamorous and it resolves a meaningful share of physical fear. Ill-fitting or inadequate protection makes you flinch, correctly. Someone who flinches at point shots while wearing thin shin pads is not being irrational. See Equipment.

Fear as information

One caveat, and it matters. Not all fear should be overridden. Fear that says "this specific situation is genuinely dangerous" — a player who is out of control, a hit from behind toward the boards, going into a corner with your head down while someone is arriving — is doing its job. The goal is not to switch fear off. It is to stop it operating in situations that are actually safe. See Risk Management.

Coming back from injury

The physical timeline and the psychological timeline are not the same, and the psychological one is longer. It is entirely normal to be cleared to play and still be protecting the injury three weeks later. Two things help: staged return through the same graded progression above rather than straight into a game, and knowing that the first genuine contact on the healed area is the moment that resolves most of it. Note also the IOC consensus statement's finding that mental health symptoms and physical health are not separable — symptoms increase injury risk and delay recovery — so a lingering fear is not merely uncomfortable, it is a physical risk factor.

---

9. Goaltender psychology

Goaltending has a reasonable claim to being the most psychologically demanding position in team sport, and writing on goalkeepers across sports consistently identifies it as a distinct psychological problem rather than a variation on the outfield one.

There is one peer-reviewed review in the neighbourhood, and it is worth citing precisely because of what it found. De Castro, de Castro and Coutts (2021) ran a scoping review of psychological interventions applied with football goalkeepers: they screened 178 articles and found nine that qualified, most of them about the penalty kick, and — the striking part — only one study in which psychological techniques were actually applied to goalkeepers, and it detected no improvement. Their conclusion is that goalkeepers "may benefit" from position-specific psychological techniques, reasoning by analogy from other sports rather than from goalkeeper evidence.

So the honest position is not "nobody has looked" but something more useful: somebody has looked, in the better-researched sport, and found almost nothing there. That is a football finding and this is a hockey document, which makes the gap wider still. Read this section as structured coaching craft with a plausible rationale, not as research findings, and discount the confident phrasing accordingly.

If you play out, read this section anyway — most of it is about how to treat your goalie.

Why the position is different

Your mistakes are the scoreboard. A forward's bad giveaway may cost nothing. A defenceman's blown coverage may be saved by the goalie. The goalie has nobody behind them. Every error converts directly into a goal, and the goal is announced, replayed and recorded against your name.

The failure is public, still, and prolonged. After a goal, play stops. Everyone looks at you. You stand there in silence for thirty to sixty seconds with no task, which is the psychologically worst possible arrangement — full attention on the error and no action available to reset with.

Rate of work is nothing like the rest of the team. You can be entirely uninvolved for eleven minutes and then be required to make an elite reaction save. Skaters have continuous work to occupy attention; you have to manufacture engagement across long empty stretches and then produce peak output instantly.

You are structurally isolated. You train separately, you often warm up separately, and you frequently have no positional peer on the roster except the person competing with you for the net.

Blame lands on you by default. The applied goalkeeper writing cited for this section is explicit about the disproportionate share of blame goalkeepers absorb for team defeats, and the guilt and self-criticism that come with being the last line — though, per the warning above, that is practitioner description rather than measurement.

The short memory

The one non-negotiable goaltending skill. It is trainable and it is a routine, not a personality trait.

Build a fixed post-goal routine and run it every single time, including on goals that were not your fault. If you only run it after bad goals, it becomes a marker of failure. Suggested shape, and this is coaching craft in its specifics:

1. Get the puck out of the net yourself, or don't — but pick one and always do it.
2. Skate a short, deliberate loop out to the top of the crease or to the corner. Move. Movement gives the reset a physical form and breaks the frozen posture.
3. Water bottle, one long exhale, wipe the mask.
4. One factual sentence about the goal — "screen, didn't find it" — and then stop. No evaluation, no blame, no "that was terrible."
5. Return to the crease, tap both posts, cue word, set for the faceoff.

Steps 4 and 5 are the whole thing. The single sentence is what stops your brain from re-running the goal all period; giving it one clean description satisfies the urge to process. Everything else is scaffolding.

The better-supported components here — controlled breathing to reduce state anxiety, a short self-talk cue, an acceptance rather than suppression stance toward the thought — are the same ones from earlier in this document, and carry the same caveats: the acceptance evidence is thin, and a cue word for a gross, effortful reset need not be instructional. The goalie-specific packaging is craft.

Confidence and the goalie

Bandura's ranking applies with particular force. Goalie confidence is built on saves made, and it is why a goalie coming off a bad night should face a high volume of makeable shots in the next practice before facing hard ones. Manufacturing mastery experiences is the fastest route back, and volume matters more than difficulty.

A related and useful reframe, again from applied practice rather than from a study: high-volume nights are often easier for goalies, not harder, because engagement is continuous and rhythm builds. The dangerous game is fifteen shots with a soft one in the second period.

How a team should treat its goalie

Most teams get this wrong in one of two directions.

Do not pile on. Obvious, and yet: visible frustration from a defenceman after a goal — the head shake, the stick slam, the skate-away — is read by the goalie as blame whether or not it was meant that way. Given the emotional contagion findings in section 11, that gesture does not stay between the two of you either.

Do not over-console either. Skating over after every goal to tell your goalie it wasn't their fault marks each goal as an event requiring management, which is precisely what the short-memory routine is trying to avoid. The best default is normal: a tap on the pads on the way past, one word, and back to your job. Treat the goal as unremarkable and it becomes easier for them to.

Break the isolation off the ice. The isolation is structural, so it takes deliberate effort to counter. Include the goalie in the room, in the pre-game, and in the social side. Costs nothing.

Talk to them during play. The goalie is the best-placed communicator on the ice and giving them a job in the flow of the game keeps them engaged through quiet spells. See On-Ice Communication.

For the technical and positional side, see Goaltender.

---

10. Coaches, benchings and criticism

Separate the message from the delivery

The most useful habit here is to treat feedback as two separate things: the content — is this true and can I use it? — and the delivery — was it fair, kind, well-timed, public?

These are independent. Bad delivery does not make content false. A coach who shouts "you're standing still!" across the bench has delivered it badly and may be entirely right. Discarding the message because you resent the messenger costs you the information.

The reverse also holds and matters more than people admit. Pleasant delivery does not make content correct. Plenty of coaching advice is confidently wrong, and plenty of it is a coaching preference presented as a rule of hockey — the failure mode this guide warns against throughout. "Wingers never go below the goal line" is a system choice, not a law. Ask which system, and ask why.

What the research says about how it is delivered. Coaching style does have measurable effects. Controlling coaching behaviours — public criticism, punitive responses, pressure — are associated with greater psychological need frustration, lower engagement, and more oppositional defiance and anger; autonomy-supportive coaching, which involves choice and collaboration, is associated with better motivation and resilience. Be precise about what was measured: the Delrue et al. vignette study cited below measured anticipated need satisfaction and frustration, engagement, oppositional defiance and anger — it did not measure stress. Burnout is a well-attested outcome in this literature generally but is not from that paper either. The direction of the effect is not in doubt; the specific outcome you attribute to it should match the study you are leaning on. One further idea is worth knowing but has no source behind it: that a closer coach-athlete bond can amplify the harm from a controlling style, because the athlete complies without internally accepting the demand. None of the coaching references below states it — so treat "he's hard on us but we love him is not automatically protective" as a reasonable caution rather than a finding.

And there is a line. Demanding is not the same as abusive. Criticism of your play is coaching. Personal degradation, humiliation in front of the group as a routine method, or anything that leaves you dreading the rink is not, and it is not something to absorb quietly. Escalate it — to a club official, a league, or a governing body.

When you are not playing

Start with an observation — and it is an observation, not a measured finding, because there is no study behind it: most players believe they deserve more ice time than they get. Whether or not the proportion is "most", it cannot be true of everyone simultaneously. Your self-assessment is running on a biased sample — you remember your good shifts vividly, you have never watched yourself play, and you cannot see the ones you were beaten on because you were behind the play. This is not a claim that your coach is right. It is a claim that you are not a reliable judge of your own case, and neither is anyone else about theirs.

Given that, here is what actually helps.

Ask for specifics, and ask at the right time. Not on the bench, not immediately after the game, not in front of the group. A day or two later, in private, with a specific question: "What do I need to do to get on the second power play?" A vague question — "why am I not playing more?" — gets a vague answer and reads as a complaint. A specific one is hard to dodge and gives you something to work on.

Then actually work on the answer, and make it visible. If the answer is "you're getting beaten back," the coach needs to see you backchecking hard in practice, not just in the two shifts you get.

Be excellent in the minutes you do get. A player who is conspicuously good in four minutes gets six. A player who sulks through four minutes gets two. This is coaching craft, but it is close to universal.

Stay physically ready. A cold player who goes over the boards after eight minutes on the bench and immediately gets beaten has confirmed the coach's decision. Keep moving, keep your legs alive, stay warm.

Recognise the limits of your control. Some benchings are political, some coaches have favourites, and some rosters are simply deeper than your current level. If you have asked the specific question, done the work, and nothing has changed over a real stretch of time, the problem may not be solvable from where you are sitting, and moving teams is a legitimate answer rather than an admission of failure.

Frustration has a price on the scoreboard

Two rules are worth knowing precisely, because losing your temper is expensive in a literal sense.

NHL Rule 39 — Abuse of Officials covers this in detail. A minor penalty is assessed for "any player who challenges or disputes the ruling of an official," for abusive language or gestures directed at any official, and — the one players routinely forget — for "any player or players who bang the boards with their sticks or other objects at any time, or who, in any manner show disrespect for an official's decision." Banging the glass in protest gets the same treatment, and persisting escalates to a misconduct.

NHL Rule 75 — Unsportsmanlike Conduct covers the same energy directed at anyone other than an official: a minor for obscene, profane or abusive language or gestures "directed at any person," and a minor for "any player who throws any object onto the ice from the players' or penalty bench."

The practical point is not that you should be nice. It is that your frustration, expressed, hands the other team a power play, and the goal they score off it will be attributed to a bad call rather than to you. IIHF and USA Hockey rules cover the same conduct in comparable terms.

---

11. Team dynamics, and your contribution to them

Your emotional state is not private

The evidence here is real but a tier below the label used elsewhere in this document: the source is Tamminen and colleagues' scoping review (International Review of Sport and Exercise Psychology, 2022), which maps a literature rather than pooling effects — there is no meta-analysis behind this claim. What that literature describes: emotions and behaviours spread between athletes, coaches' emotional expressions are associated with players' emotional experiences and with team performance, and interpersonal emotion regulation between teammates affects anxiety and results. The effect is moderated by things like the closeness of the group and the status of the person expressing the emotion — which means a senior player's body language carries further than a fourth-liner's. Treat the direction as well documented and the size as unquantified.

For you, the practical consequence is direct: when you slump on the bench, throw your head back, or bang your stick, you are not privately processing a bad shift. You are broadcasting. Your linemates read it, in about half a second, as one of three things — the coach is wrong, my linemate let me down, or this game is gone. None of those helps them on the next shift.

This is not an argument for fake enthusiasm, which nobody believes anyway. It is an argument for a neutral floor: whatever you are feeling, your visible baseline is upright posture, eyes on the ice, engaged. Feel what you feel; don't project it.

Being a good teammate on a bad night

The nights when you are playing badly are when your contribution to the group is most in play, because it is the easiest time to become a net negative.

Say the thing that helps, which is usually about the next shift. "Get above me next time and I'll hit you" is useful. "You've got to be there" is not.

Absorb, don't transmit. If a teammate makes an error, the useful response is nothing, or one word of support. The head shake costs you more than the error did.

Take responsibility out loud, once, and then stop. "That's on me" from the player who turned it over ends the conversation and keeps the group moving. Repeated self-flagellation makes everyone manage your feelings instead of playing.

Understand that your bad night is not the most important thing happening. There are seventeen other people who came to play. Your job on a night when you have nothing is to not make it their problem.

The wider subject — roles, standards, how groups set their own norms — is Team Play and Culture.

---

12. Motivation over a season

Improvement is not linear, and expecting it to be is what breaks people

The typical pattern is a fast early climb — everything is new, so everything improves — followed by long flat stretches punctuated by sudden jumps. Those flat stretches are the normal shape of learning, not evidence of failure, but they arrive exactly when the initial enthusiasm has worn off.

Two things help you sit through them.

Understand what a plateau usually is. Often you are consolidating: the skill is becoming automatic, which does not feel like progress but is what makes it usable under pressure. Often you are also being asked to do more with the same skill — the same skating in tighter space against better players — so your measured output is flat while your actual capability has risen. And sometimes it is a genuine ceiling on your current technique, which needs a change of method rather than more repetitions of the same one. See Practice and Development.

Measure something that moves. Wins, points and ice time are lagging, noisy indicators that can sit still for months. Process markers move weekly. This is section 3 applied across a season.

About "10,000 hours"

This one is genuinely misrepresented and worth correcting, because it does real damage to how players think about their own progress. The popular version — that 10,000 hours of practice is what separates experts from everyone else — overstates the research substantially. Macnamara, Moreau and Hambrick's sport-specific meta-analysis (2016) found that deliberate practice accounted for 18% of the variance in sports performance, leaving roughly four-fifths to starting age, physical attributes, coaching quality, the standard of players around you, genetics, injury history, and how you spend the hours.

For the motivational point, that cuts both ways on purpose. Practice is not pointless — 18% is a large effect for a single factor and it is the one you control. But hours alone are not the mechanism, so two players with identical ice time can improve at wildly different rates. The full argument, including the disputed 1%-among-elites figure and Ericsson's reply to the meta-analysis, is in Practice and Development; it is not repeated here.

The grind

Mid-season, in a long league, with early mornings and cold rinks, motivation stops being about ambition and becomes about habit.

• Rely on structure, not feeling. Turning up because it is Tuesday is more reliable than turning up because you feel like it.
• Have an off-season. Playing hockey twelve months a year is a good way to stop wanting to. Deliberate breaks are a recovery tool, not a lapse.
• Keep one thing you are actively working on. A player with a current technical project has something to be interested in on a night when the game is bad. A player just playing games has only the results, which are outside their control.
• Watch the drift from wanting to play to feeling obliged to play. That drift is the early signature of burnout, and burnout is one of the outcomes the coaching-climate research links to controlling environments. If it persists, it is worth taking seriously — see the next section.

---

13. When to seek real help

This document is about performance. It is not a substitute for care, and some of what looks like a mental-game problem is not one.

The IOC's 2019 consensus statement on mental health in elite athletes — the standard reference in this area — makes several points that matter here. Mental health symptoms and disorders are common among athletes and can impair performance. For depression and anxiety specifically, prevalence in elite athletes is broadly comparable to the general population — worth saying plainly because both the opposite claims circulate. That comparability does not generalise to every condition: the statement flags eating disorders and disordered eating as more common in athletes, particularly in weight-sensitive, aesthetic and endurance sports. Hockey is not usually a weight-making sport, but making a roster weight or a body-composition target happens, and the exception matters. Mental health cannot be separated from physical health: symptoms increase the risk of physical injury and delay recovery from it. And the statement is explicit about the need to distinguish between the character traits typical of committed competitors and genuine psychological maladaptation requiring treatment.

Signs it is time to talk to a qualified professional rather than reading more about mental skills:

• Anxiety that is present well before game day, or that does not settle once play starts.
• Low mood, loss of interest, or hopelessness that persists for weeks.
• Sleep that is consistently disrupted, or appetite that has changed markedly.
• Dread of going to the rink, in someone who used to enjoy it.
• Performance problems that have spread into work, study, relationships or family life — this is the clearest signal in the list, because a genuine hockey problem stays in the rink.
• Using alcohol or anything else to manage how you feel about playing.
• Any thought of harming yourself or of suicide. This one is not a threshold to weigh up.

What to do for anything less acute than that. Speak to your GP or family doctor, or to a qualified clinical or sport psychologist — in many countries sport psychologists are regulated and you can check registration. Most national federations and many leagues now have mental health pathways; the IOC statement's whole thrust is that these should exist and be usable. Note that a sport psychologist and a clinical psychologist are not the same profession, and if the problem has left the rink you probably want the latter.

None of this means you are failing at hockey. It means you have a problem that hockey advice is not the right tool for.

---

Common Mistakes

• Treating the mental game as something you either have or don't. Almost everything in this document is a trainable routine, not a personality trait. Composure is a sequence of actions, not a temperament.
• Analysing a mistake during the game. Analysis is a post-game activity. Doing it on the bench costs you the information you should be gathering there, and by the self-focus account of choking, thinking harder about an automated skill degrades it.
• Trying to suppress the bad thought instead of accepting it and moving on. Suppression takes effort, and the effort is the resource you needed for the next shift.
• Judging yourself on outcomes you don't control. Two posts and four good looks is a good game. If your self-assessment says otherwise, your self-assessment is measuring the wrong thing.
• Shooting less during a scoring slump. The one hockey-specific trap that is worth memorising — though that players actually do shoot less in a slump is coaching craft, not a measured finding, with no published study behind it. The arithmetic behind the advice stands either way: fewer attempts means fewer chances for the ugly goal that ends it, and a smaller sample that makes the drought last longer in games.
• Believing the textbook inverted-U. The relationship depends on task complexity, the optimal level is individual, and for simple well-learned tasks arousal is closer to linearly helpful. Learn your own zone empirically instead.
• Trying to be calm before every game. Some players play best angry, some best serene. Copying a teammate's pre-game state is copying the wrong thing.
• Confusing a routine with a superstition. If missing it would wreck you, it is a superstition, and you have made your composure depend on something you don't control.
• Mismatching the cue to the task. Instructional cues ("far post") are for fine, precise actions; motivational ones ("go", "work") are at least as good for gross, effortful ones. The self-talk meta-analysis supports a matching hypothesis, not blanket instructional superiority.
• Treating external-focus cueing as settled science. "Chest over your toes" is a reasonable default over "bend your knees", but the bias-corrected re-analysis of that literature found the effect close to zero. Use whichever wording actually makes you feel the correction.
• Puck-watching and calling it a work-rate problem. It is an attention problem, it gets worse when you are anxious, and it is fixed with triggered shoulder checks rather than with effort.
• Playing safe to avoid mistakes. Hesitation is the mistake hockey punishes hardest. Passive play produces more turnovers, not fewer.
• Over-consoling your goalie. Skating over after every goal marks each one as an event requiring management. A tap on the pads and back to work is better.
• Arguing with officials. Under NHL Rule 39, disputing a ruling, abusive gestures, or banging your stick on the boards in protest are all minor penalties. Your frustration becomes their power play.
• Assuming you deserve more ice time than you get. That players commonly believe this is an observation with no source behind it — but whatever the true proportion, it cannot be right of everyone at once, and you are not a reliable judge of your own case. Ask a specific question in private, then do the work the answer implies.
• Treating flat stretches as failure. Improvement is non-linear and plateaus are usually consolidation. Also: deliberate practice explains roughly 18% of the variance in sport performance, so hours alone were never the whole story.
• Managing a clinical problem with performance techniques. If it has spread beyond the rink, breathing exercises are the wrong tool — and if there is any thought of harming yourself, none of this document is the right tool. See §13, When to seek real help for the signals and the crisis numbers.

---

Key Takeaways

1. The mental game closes the gap between your worst performance and your best; it does not raise your best. Your physical capacities barely move between Saturday and Sunday. Your attention and composure move enormously. That is where the variance lives.
2. Build a four-step shift reset and run it every time: physical trigger, one long exhale, a cue word, then name your next action. The last step is the whole point. The four-step sequence itself is coaching craft — no study has tested this exact ordering — but each of its components has independent support, which is why it is worth running. Match the cue to the task — instructional for precise actions, motivational for effortful ones.
3. Ruminating on the bench costs you the next shift too. Recover, watch, plan one thing. Analysis of the mistake is a post-game job, and doing it during the game costs you the information the bench is for.
4. Judge yourself on process, not outcome — but keep the outcome goal on the list. You cannot control whether the puck goes in; you can control your gap, your route, your first touch and your feet. Pick three process goals for your role — and check which system your team plays before you write them, because the role changes with the system. The one study that tested outcome, performance and process goals against each other (Filby et al., 1999) found the combinations won, not process alone.
5. Confidence comes from mastery experiences above all — having actually done the thing. Self-talk works, at a moderate effect size (ES = 0.48), but it is a multiplier on evidence rather than a substitute for it. In a slump, shoot more: shooting less shrinks the sample and removes the ugly goals that end droughts.
6. The textbook inverted-U is not what the research says. Complex tasks show the curve; simple well-learned ones benefit more linearly from arousal. Your optimal state is individual — track it over eight games and find out what yours actually is.
7. Nerves are the body preparing; anxiety is the worry. Manage the first, address the second. Reframing arousal as readiness rather than as failure has real evidence behind it and costs nothing.
8. Build a pre-game routine — the meta-analytic evidence for routines is genuinely positive (SMC = 0.31 pre–post, g = 0.64 in experimental designs, for novices and elites alike). What nobody can tell you is which elements matter, so prune anything that would wreck you if it were unavailable.
9. Fear of making a mistake produces passive play, and passive play produces more mistakes. Hesitation is the error hockey punishes hardest. Physical fears respond to graded exposure and to better technique — and better technique lowers the actual risk, so the fear drops for a good reason.
10. Your emotional state is not private, and neither is your goalie's. Hold a neutral floor on the bench; give your goalie a fixed post-goal routine and a team that treats goals as unremarkable. And if it has spread beyond the rink, it is not a hockey problem — that warrants a qualified professional, and seeking one is not a failure of toughness. One signal is not a threshold to weigh up at all: any thought of harming yourself or of suicide means get help now, tonight. The full list of signals and the crisis numbers are in §13, When to seek real help.

Sources — retrieved 27 July 2026

Rules: NHL Official Rules 2025-2026 (PDF) — Rule 39 (Abuse of Officials), Rule 75 (Unsportsmanlike Conduct), Rule 87.1 (Time-outs), verified against the official rulebook text.

Physiology (cited in section 1 for the shift-length range): Vigh-Larsen JF & Mohr M (2024), "The physiology of ice hockey performance: An update", Scandinavian Journal of Medicine & Science in Sports 34(1):e14284, doi:10.1111/sms.14284, PMID 36517860 — source of the 30–80 second measured shift range; the full set of figures is owned by Conditioning and Recovery.

Arousal and anxiety: Yerkes–Dodson law — overview and criticisms (tertiary source, used only for the history of the 1908 study and documented criticisms) · Fazey JA & Hardy L (1988), The Inverted-U Hypothesis: A Catastrophe for Sport Psychology, British Association of Sports Sciences Monograph No. 1, National Coaching Foundation, Leeds — the originating statement of the catastrophe model named in section 5; a monograph, not a peer-reviewed empirical paper, and not retrievable online · Hanin, The Individual Zones of Optimal Functioning (IZOF) model, 1978–2014 · Zones of optimal functioning research: a review and critique · Hardy L & Parfitt G (1991), "A catastrophe model of anxiety and performance", British Journal of Psychology 82:163–178 — n = 8 basketball players; the only direct experimental test cited here

Attention and choking: Wulf (2013), Attentional focus and motor learning: a review of 15 years · Chua et al. (2021), Superiority of external attentional focus for motor performance and learning, Psychological Bulletin · the challenge to both: McKay B, Corson A, Seedu J, et al. (2024), "Reporting bias, not external focus: A robust Bayesian meta-analysis and systematic review of the external focus of attention literature", Psychological Bulletin (PMID 39480294) — found "moderate to strong evidence of publication bias for all analyses"; bias-corrected effects g = 0.01 (performance), 0.15 (retention), 0.09 (transfer) · Eysenck MW, Derakshan N, Santos R & Calvo MG (2007), "Anxiety and cognitive performance: attentional control theory", Emotion 7(2):336–353, PMID 17516812 — the primary statement of attentional control theory, source of the inhibition-and-shifting claim quoted in section 2; note it is a general cognitive-psychology theory, not a hockey or even a sport finding · Attentional control theory (tertiary source — an overview only; the claim in the text rests on Eysenck and colleagues) · Beilock SL & Carr TH (2001), "On the fragility of skilled performance: what governs choking under pressure?", Journal of Experimental Psychology: General 130(4):701–725, PMID 11757876 — the primary experimental source for the self-focus / explicit-monitoring account used in sections 3, 5 and Common Mistakes; choking occurred in golf putting but not in a non-proceduralised arithmetic task, was eliminated by self-consciousness training and unchanged by dual-task training. The authors themselves call the underlying belief "popular but infrequently tested" · Choking under pressure — self-focus and distraction accounts (tertiary source — an overview of the competing distraction account; the self-focus claim in the text rests on Beilock and Carr, not on this page) · Nideffer's model of attentional focus (Sport NZ concentration model) · NSCA, Attentional Styles

Self-talk, confidence, goal setting and imagery: Hatzigeorgiadis A, Zourbanos N, Galanis E & Theodorakis Y (2011), "Self-Talk and Sports Performance: A Meta-Analysis", Perspectives on Psychological Science 6(4):348–356, doi:10.1177/1745691611413136 — 32 studies, 62 effect sizes, ES = 0.48; supports a matching hypothesis: instructional self-talk beat motivational only for fine motor tasks, with no significant difference for gross tasks · Kyllo LB & Landers DM (1995), "Goal setting in sport and exercise: A research synthesis to resolve the controversy", Journal of Sport and Exercise Psychology 17:117–137 — 36 studies, overall goal-setting effect 0.34 SD; moderators reported are goal difficulty, absolute versus relative goals and time frame. It does not test a process-versus-outcome contrast, and must not be cited for one · Kingston KM & Hardy L (1997), "Effects of Different Types of Goals on Processes That Support Performance", The Sport Psychologist 11(3):277–293, doi:10.1123/tsp.11.3.277 — 37 club golfers, process-oriented against performance-oriented goal-setting programmes; significant Group × Test interactions (p < .05) for self-efficacy, cognitive anxiety control and concentration, giving "further evidence for the positive impact of process goals in competitive situations" · Filby WCD, Maynard IW & Graydon JK (1999), "The effect of multiple-goal strategies on performance outcomes in training and competition", Journal of Applied Sport Psychology 11(2):230–246, doi:10.1080/10413209908404202 — the study that tests the full outcome/performance/process trichotomy; 40 participants, five ability-matched groups on a soccer task, five weeks of training plus a competition; the multiple-goal groups performed best, supporting "maintaining a balance between the use of outcome, performance, and process goals" rather than process goals alone · Bandura's four sources of self-efficacy · Holmes PS & Collins DJ (2001), "The PETTLEP Approach to Motor Imagery: A Functional Equivalence Model for Sport Psychologists", Journal of Applied Sport Psychology 13(1):60–83, doi:10.1080/10413200109339004 — the originating paper for the PETTLEP model named in section 7 · Motor Imagery and Sport Performance: A Systematic Review on the PETTLEP Model · Twenty years of PETTLEP imagery: an update and new direction

Routines, breathing and reappraisal: Cotterill (2010), Pre-performance routines in sport, International Review of Sport and Exercise Psychology · Rupprecht AGO, Tran USS & Gröpel P, "The effectiveness of pre-performance routines in sports: a meta-analysis", International Review of Sport and Exercise Psychology (online 2021; print 2024, 17(1):39–64), doi:10.1080/1750984X.2021.1944271 — 112 effect sizes; significant positive effects, SMC = 0.31 (95% CI 0.18–0.44) pre–post, Hedges' g = 0.64 (0.45–0.83) in experimental low-pressure designs and g = 0.70 (0.24–1.16) under pressure; "these effects were not moderated by the type of PPR, age, gender, skill level, or intervention characteristics" — a null moderator result, not a finding of equal benefit · Laborde S, Allen MS, Borges U, et al. (2022), "Effects of voluntary slow breathing on heart rate and heart rate variability: A systematic review and a meta-analysis", Neuroscience & Biobehavioral Reviews 138:104711, doi:10.1016/j.neubiorev.2022.104711, PMID 35623448 — 223 studies; increases in vagally-mediated HRV during breathing, immediately after a single session, and after multi-session interventions. Covers HRV, not state anxiety, and is not athlete-specific — flagged in section 2 · Influence of a single slow-paced breathing session on cardiac vagal activity in athletes (a single study) · Slow paced breathing and power posing: a pre-competitive ritual during an American football season (note the sport is American football, not hockey) · Jamieson et al., Reappraising stress arousal improves performance and reduces evaluation anxiety · Optimizing stress responses with reappraisal and mindset interventions

Mindfulness and acceptance: "Effectiveness of the mindfulness–acceptance–commitment approach: a meta-analysis", International Journal of Sport and Exercise Psychology 22(5), doi:10.1080/1612197X.2023.2180070 — robust Bayesian meta-analysis of eleven controlled trials; mindfulness g = 0.64 (95% CI 0.00–0.91, strong evidence), sport performance g = 0.37 (95% CI 0.00–0.79, moderate), experiential acceptance g = 0.36 (95% CI 0.00–0.92, anecdotal) — all three intervals have a lower bound of 0.00 — and explicitly "Currently available trials, however, do not provide sufficient evidence supporting the MAC protocol's effectiveness in developing psychological flexibility, experiential acceptance, and promoting well-being among athletes." (Every figure and the quotation above are taken verbatim from the publisher's abstract page; the full text sits behind "Get Access" and has not been read.) · Mindfulness-based programmes and elite athlete mental health, British Journal of Sports Medicine 57(2):99

Practice and development: Macnamara, Moreau & Hambrick (2016), "The Relationship Between Deliberate Practice and Performance in Sports: A Meta-Analysis", Perspectives on Psychological Science 11(3):333–350, doi:10.1177/1745691616635591 — the sport-specific meta-analysis, and the source of both the ~18% of variance figure and the disputed 1%-among-elites figure referred to above · Ericsson (2016), "Summing Up Hours of Any Type of Practice Versus Identifying Optimal Practice Activities", Perspectives on Psychological Science 11(3):351–354, doi:10.1177/1745691616635600 — his reply to that meta-analysis · Macnamara, Hambrick & Oswald (2014), "Deliberate Practice and Performance in Music, Games, Sports, Education, and Professions: A Meta-Analysis", Psychological Science — the earlier, cross-domain meta-analysis. The full argument is worked through in Practice and Development, which cites all three.

Teams and coaching: Tamminen KA, Wolf SA, Dunn R & Bissett JE (2022), "A review of the interpersonal experience, expression, and regulation of emotions in sport", International Review of Sport and Exercise Psychology, doi:10.1080/1750984X.2022.2132526 — a scoping review, not a meta-analysis; it maps the literature rather than pooling effect sizes, and section 11 is labelled accordingly · Delrue J, Soenens B, Morbée S, Vansteenkiste M & Haerens L (2019), "Do athletes' responses to coach autonomy support and control depend on the situation and athletes' personal motivation?", Psychology of Sport and Exercise 43:321–332, doi:10.1016/j.psychsport.2019.04.003 — a vignette experiment on 101 judokas measuring anticipated need satisfaction/frustration, engagement, oppositional defiance and anger — not stress, and not the coach-athlete-bond moderation described in section 10 · University of Surrey, Controlling coaching tactics shatter athletes' wellbeing

Goaltending — the weakest source base in this document; section 9 flags this. No meta-analysis exists. The one peer-reviewed review in the area is about football goalkeepers, not ice hockey goaltenders, and reports a near-empty evidence base: de Castro MF, de Castro VOL & Coutts R (2021), "Psychological interventions applied with football goalkeepers: A scoping review", International Journal of Physiology, Nutrition and Physical Education 6(1):17–22, ISSN 2456-0057 — 178 articles screened, 9 included, most concerning the penalty kick; only one applied-intervention study, which detected no improvement. Hockey-specific sources, all non-peer-reviewed: AASP, The Last Line of Defense: Unravelling the Mental Strain of Goalkeepers (practitioner blog post) · Hockey Psychology, Staying Focused Under Fire (commercial coaching site) · Speidel ND (2022), "Ice Hockey Goaltender Performance: Examining The Role Of Basic Psychological Needs And Self-Determined Motivation", University of North Dakota (an unpublished PhD dissertation, not peer-reviewed)

Checking progressions (cited in section 8 for the staged introduction of contact): Hockey Canada — Checking Resource Guide: Teaching Checking, A Progressive Approach (PDF) — "checking is a 4-step progression"; Step 1 Positioning and Angling, Step 2 Stick Checks, Step 3 Body Contact, Step 4 Body Checking · USA Hockey Coaching Education Program, Introduction To Body Contact — the "Body Play Pyramid" (angling 1st, stick checks 2nd, body blocks 3rd, body checks 4th) and the age-group progression introducing body checking at 12-and-Under.

Athlete mental health: Reardon et al., Mental health in elite athletes: International Olympic Committee consensus statement (2019), British Journal of Sports Medicine 53:667 · PubMed record

Limitations worth knowing:

• No published study establishes that players in scoring slumps take fewer shots; the mechanism is described as coaching craft.
• The specific four-step shift-reset and post-goal goalie sequences are coaching craft with no direct experimental test.
• The application of graded exposure specifically to hockey contact fear is craft, though the clinical evidence for graded exposure in anxiety generally is strong. Hockey Canada's four-step guide and USA Hockey's Introduction To Body Contact are quoted in section 8, but what remains uncited is the claim that following them reduces fear — they are teaching sequences for a skill, not tested anxiety interventions.
• Goal setting. Kingston and Hardy (1997) measure a positive effect of process goals on self-efficacy, cognitive anxiety control and concentration, but against performance goals rather than outcome goals. Filby, Maynard and Graydon (1999) test all three goal types and find multiple-goal strategies best, supporting a balance that keeps outcome goals in rather than removing them. Both samples are small and neither is in hockey. Kyllo and Landers (1995) tests no process-versus-outcome contrast and is not cited for one.
• Attentional control theory is cited here to its primary statement (Eysenck, Derakshan, Santos & Calvo, Emotion, 2007). It is a theoretical account tested outside hockey, not a measured hockey finding, and is labelled as such in section 2.
• "Most players believe they deserve more ice time than they get" — an observation with no source behind it; flagged inline in section 10.
• Prevalence of fear in adult recreational hockey. No prevalence estimate is given anywhere in this document, because there is no published one. Be sceptical of the figures you will hear quoted — "about half of any adult rec dressing room" and similar — none of them trace to a study.
• All of section 9 (goaltender psychology) rests on a blog, a commercial site and an unpublished doctoral dissertation — see the source warning at the head of that section.