Conditioning and Recovery

Overview

Hockey asks your body for something unusual: repeated near-maximal efforts of roughly half a minute to a minute and a half, separated by two to five minutes sitting on a bench, for sixty minutes of clock time. Training for that is not the same as training for general fitness, and it is not the same as training for a marathon or for a one-rep-max squat.

This document covers what the demand actually is, what to train and why, how to reduce your injury risk, how to warm up, fuel and recover, and — because most readers have a job — what the highest-value ninety minutes a week looks like.

Two honest framings before we start. First, sports science is a contested field, and hockey-specific sports science is a small one. Where the evidence is good, it is called good. Where it is weak, mixed or absent, it is called that, rather than dressing coaching craft up as fact. Second, conditioning is not a substitute for skating technique. A player with an inefficient stride burns far more energy to cover the same ice. Read Skating alongside this.

---

What Hockey Actually Demands of Your Body

The shape of a shift

The best current summary of hockey physiology is a 2024 review in the Scandinavian Journal of Medicine and Science in Sports by Vigh-Larsen and Mohr. Its description of match play is worth taking literally:

• Shifts last roughly 30 to 80 seconds, separated by longer periods of largely passive recovery on the bench. ⚠️ Attribute this one carefully — the review is reporting it, not reporting it as its own. Its introduction reads "Studies have documented that each shift typically lasts 30–80 s, with subsequent recovery periods of ~2–5 min" and cites references 1–3: Lignell et al. (2018) — the same paper cited elsewhere in this guide for positional workload — Brocherie et al. (2018) and Bracko et al. (1998). Those three measured it; Vigh-Larsen and Mohr collected it. Anyone restating this figure should name the primaries, not the review. (The 2026 elite-hockey nutrition review puts a number on that bench time: shifts of "∼30-80 s" are "interspersed with passive recovery periods (2-5 min)".)
• That adds up to about 15 to 25 minutes of on-ice time per player across a game.
• That on-ice time comes in about twenty shifts. The review does not give a shift count, but the NHL's own statistics do: across the 2024-25 regular season, skaters who played at least half a season averaged 21.0 shifts per game — 19.4 for forwards and 24.0 for defencemen, at a mean of 47.7 seconds per shift (47.1 s for forwards, 48.8 s for defencemen). Individual regulars ranged from about 11 to 33 shifts a game depending on role.
• Nearly 50% of the distance covered is at high-intensity skating speeds — and the review attaches a positional half to that same sentence which is worth carrying: this comes "with an accentuated intense activity pattern in forwards compared to defensemen." Note what that does and does not say. It is a claim about intensity of activity pattern, not about total distance skated, and it separates forwards from defencemen — it says nothing about one forward position covering more ice than another. On total distance the measured direction is the opposite of the folklore: Lignell and colleagues, tracking 36 NHL players through a game, found "defensemen (D) covered 29% more (p ≤ 0.05) skating in total than forwards (F) and were on the ice 47% longer. However, F performed 54% more (p ≤ 0.05) high-intensity skating per minute than defensemen." So forwards work harder per minute; defencemen cover more ground. No published study measures skating distance at the level of centre versus winger, so the common assertion that "the centre covers more ice than anyone" is coaching craft — Center sets out what is actually established and carries the citation.
• Heart rate rises toward maximum during each shift.

A Swedish study of a single game (Lögdal and colleagues, 2022) — six elite junior players, three forwards and three defencemen, median age 17, so treat it as a detailed case series rather than a population estimate — measured them accumulating 11 minutes 18 seconds (± 5:04) above 90% of maximal heart rate across a game, with blood lactate — a marker of how hard the fast, oxygen-independent energy pathway has been working — ranging from 1.8 to 10.7 mmol/L, mean 5.5, and forwards significantly higher than defence.

Note the number in the review is 30–80 seconds, not the 30–50 seconds often quoted in coaching material. Both are true of different things. A well-managed shift in a well-coached team is nearer 30–45 seconds. Eighty seconds is what happens when you get stuck on the ice — an iced puck you can't change on, a long defensive-zone shift, a rec-league team with two and a half lines. The physiological point is that the top end of the range is where the wheels come off, which is why Game Management treats shift length as a tactical discipline and not just a fitness matter. The ~45-second central tendency used elsewhere in this guide is not a published figure: it is computed from the NHL's public shift-chart API across 8,325 shifts in an 11-game 2024-25 sample — mean 47 s, median 45 s — and it is set out in full in Playing Without the Puck.

The three energy systems, in plain English

Your muscles run on a molecule called ATP. You have three ways of making it, and they differ in how fast they deliver and how long they last.

1. The phosphocreatine (PCr) system. A small store of ready-made chemical energy sitting in the muscle. It is instantaneous and it is what powers your first three or four hard strides, a shot, a battle on the wall. It runs down in seconds and has to be rebuilt.
2. The glycolytic system. Breaking down carbohydrate rapidly without needing oxygen to arrive first. Fast, powerful, and the source of the burning-legs sensation late in a long shift. It provides the bulk of the extra energy for hard efforts lasting from roughly ten seconds to a couple of minutes.
3. The aerobic (oxidative) system. Burning carbohydrate and fat with oxygen. Slower to ramp up, effectively unlimited in duration, and — this is the part players miss — it is the system that rebuilds the other two.

The Vigh-Larsen and Mohr review describes hockey match play as significantly challenging both aerobic and anaerobic systems, "with great reliance on both glycolytic and phosphagen ATP provision."

A 2026 study by Gabrys and colleagues in Sports measured the energy contributions across a repeated high-intensity effort protocol in semi-professional players and found the total split to be 63.1% ± 2.6% aerobic, 29.8% ± 2.9% phosphagen, and 7.4% ± 1.5% glycolytic.

Read that number carefully, because it is easy to misuse. It is the split across a whole protocol including the recovery periods, not the split during one hard effort. During the first fifteen seconds of a hard shift, the phosphagen and glycolytic systems dominate. The aerobic share of the total climbs the more efforts you string together, precisely because the aerobic system is doing the rebuilding in between. Hockey is anaerobic within a shift and aerobic across a game. That single sentence is the whole physiology section.

Why the distance runner dies after two shifts — and the sprinter dies in the third period

This is the question that makes the physiology concrete.

The good distance runner who is exhausted after two shifts has a well-developed aerobic system and an undertrained anaerobic one. They can produce energy steadily for a long time. What they cannot do is produce a very large amount of it very quickly, tolerate the acidity that comes with doing so, or repeat it. On top of that, they are being asked to do it in an unfamiliar movement pattern — the skating stride loads the hips, groin and glutes in a way running does not — by muscles that have never been asked for that. Running fitness is genuinely useful in hockey, but it is fitness for running.

The pure sprinter or gym-strong player who fades in the third has the opposite problem. The first shift is superb. But phosphocreatine has to be rebuilt between shifts, and rebuilding it is an oxygen-dependent process — the rate at which muscle restores PCr after exercise is used in clinical research as an index of mitochondrial oxidative capacity. If your aerobic system is weak, you come back for shift two with a partly empty tank, shift three with less again, and by the third period you are a shadow. The Vigh-Larsen review points to exactly this mechanism, noting "pronounced PCr degradation, with potential inadequate resynthesis in a proportion of fast-twitch fibres in situations of repeated intense actions."

The practical translation: anaerobic power determines how good your best shift is. Aerobic fitness determines how similar your twentieth shift is to your first. You need both, and most amateur players are short on the second.

Fatigue across a game is also a fuel problem

The same review reports that hockey's high-intensity pattern "favours muscle glycogen as fuel, leading to pronounced reductions despite the relatively brief playing time, including severe depletion of a substantial proportion of individual fast- and slow-twitch fibres," and that player tracking shows the ability to perform high-intensity skating is compromised in the final stages of a game, corroborated by measured post-game reductions in repeated-sprint ability.

Glycogen is stored carbohydrate. Individual muscle fibres can run empty even while the muscle as a whole still has fuel — and an empty fibre is a fibre that cannot contract hard. This is the physiological basis of the fuelling section below, and it is why the review lists glycogen depletion, dehydration and rising body temperature as prime candidates for third-period fatigue.

One honest caveat about aerobic testing

You will read that VO₂max — the standard laboratory measure of aerobic capacity — predicts hockey performance. Be careful. The Gabrys study found "[n]o significant relationships were observed between maximal oxygen uptake (VO2max) and the RHIE [repeated high-intensity effort] performance parameters, energy system contributions or lactate responses" — quoted exactly as printed in the Europe PMC abstract — except a moderate relationship between absolute VO₂max and absolute aerobic work. Parameters measured at the anaerobic threshold related more consistently to metabolic work than VO₂max did.

That is one study in fourteen semi-professional players, so do not over-read it either. The reasonable position: the mechanism by which aerobic fitness aids between-shift recovery is well established, but a single lab VO₂max number is a poor predictor of who plays well in the third period. Train the quality; don't chase the test score.

---

What to Train, and Why

Five buckets. In rough order of how much most amateur players are neglecting them.

1. Aerobic base — the recovery engine

What it buys you: faster rebuilding of phosphocreatine between shifts, faster clearance of the by-products of hard glycolytic work, a lower heart rate at any given workload, and therefore a third period that resembles your first.

What it looks like: genuinely easy, conversational-pace continuous work — cycling, rowing, running, brisk uphill walking — for 30 to 60 minutes. The defining feature is that it should feel too easy. Most amateurs train their "easy" sessions too hard and their "hard" sessions too easy, and get the adaptation from neither.

Do it off the ice. Ice time is expensive and better spent on skills and on efforts that look like shifts.

2. Anaerobic capacity and repeat-sprint ability — the shift itself

What it buys you: a bigger, more repeatable hard effort, and better tolerance of the discomfort of a long shift.

What it looks like: intervals that mirror the demand. The obvious and correct starting point is to copy the game — hard efforts of roughly 30 to 45 seconds, with 2 to 4 minutes of recovery, repeated 6 to 12 times, on a bike, a rower, a hill or the ice. That work-to-rest ratio is the whole point; if you shorten the rest, you have built a different (and also useful) session, but you have stopped training the thing hockey asks for.

This prescription is coaching craft derived from the measured game demands above, not a directly tested protocol — treat the numbers as a sensible starting shape, not a validated dose.

Also useful: short repeat sprints (6–10 seconds, full effort, 30–60 seconds rest) for the phosphagen system specifically.

3. Lower-body strength and power — the stride and the battle

What it buys you: a harder push, a faster start, more stability on your edges, and more mass behind you in a puck battle. See Body Contact and Battles for what you do with it.

Does it transfer to skating? Partly, and honestly the evidence says partly rather than hugely. Keiner and colleagues (2024) in the Journal of Strength and Conditioning Research found that maximum strength and jump tests explained 1–35% of the variance in on-ice linear sprint performance in elite youth players, and 22–30% of the variance in in-game peak skating speed, concluding there was "a clear influence of 1RM in squatting and CMJ performance on on-ice linear sprint as well as in-game peak skating speed." Roczniok and colleagues (2024) found elite U16 players' skating sprint was less related to off-ice jump ability than sub-elite players' was — a reminder that as technique improves, technique becomes the limiter.

Read that as: strength is a real but partial contributor. Getting stronger will help. It will not fix a stride fault.

What it looks like: squats, deadlifts or trap-bar deadlifts, split squats and lunges (skating is a single-leg activity, so train single-leg), hip thrusts and glute work, calf work, and jumps. Lateral and rotational movement deserves specific attention because the stride is a lateral push, not a backward one.

Get coached on the lifts. A barbell loaded badly is one of the few ways an off-ice programme can injure you outright.

4. Core and rotational strength — shooting and holding your ground

What it buys you: shot power comes from the ground up through a rotating trunk, not from the arms (see Shooting); and every board battle is a test of whether your midsection can transmit force from your legs to your opponent without folding.

What it looks like: anti-rotation and anti-extension work (Pallof presses, planks, dead bugs, carries) plus deliberate rotational work (medicine-ball throws, cable rotations). Loaded carries are unglamorous and excellent for hockey because that is exactly what a battle feels like.

5. Mobility — hips and groin above all

What it buys you: the ability to actually get into a deep, wide, powerful stride position, and a hip and groin complex that is strong through range rather than only in the middle of it.

Why hips specifically: the skating stride is a repeated, forceful hip abduction (pushing the leg away from the midline) followed by adduction (bringing it back), in a deeply flexed hip position, thousands of times a game. Nothing else in daily life does this. The evidence on what that does to hockey hips is covered in the next section, and it is sobering.

What it looks like: hip flexor and adductor work through range, 90/90 hip rotations, deep squat holds, and — importantly — strength at the end of your range, not just passive stretching into it. Range you cannot control under load is not usable range.

---

Injury Prevention

Hockey's injury profile is unusual because it combines a repetitive, groin-loading movement pattern with high-speed collisions on a hard surface surrounded by boards.

A 2025 review in Sports on youth hockey reports injury rates ranging from 11.7 to 34.4 per 1000 athlete-hours. In women's collegiate hockey, a 2026 systematic review and meta-analysis (Laaksonen and colleagues, BMJ Open Sport and Exercise Medicine, 18 reports, 3,365 injuries across 719,010 athlete-exposures) reports 5.21 injuries per 1000 athlete-exposures, with games (10.10) more than three times as dangerous as practices (3.11). Concussion was the single most common injury type in that dataset, at 0.54 per 1000 exposures.

Groin and adductor strain — the classic hockey injury

This deserves the most space because it is both the most characteristic hockey injury and the one where the evidence is most interesting.

How common it is. In a 2025 American Journal of Sports Medicine study of NHL goaltenders, proximal adductor injuries accounted for the largest number of injuries (n = 371) and the highest total days missed (6,126) — more than any other injury category. Goaltenders are an extreme case because of the butterfly position (see Goaltender), but the pattern holds across positions.

The single best-known risk factor. Tyler and colleagues (2001, AJSM) measured pre-season hip strength in professional ice hockey players and found that a player was 17 times more likely to sustain an adductor strain if his adductor strength was less than 80% of his abductor strength. Pre-season adduction strength was 18% lower in players who went on to be injured. In uninjured players, adduction strength averaged 95% of abduction strength; in injured players, 78%.

That is a striking finding and it points somewhere useful: your groin is probably weak relative to the muscles that oppose it, because skating trains the pushing-away muscles far harder than the pulling-together ones.

Does strengthening it help? Here the evidence gets more nuanced, and you should know that.

• The same group's 2002 follow-up reported that a pre-season adductor strengthening programme coincided with adductor strains falling from 11 in the previous two seasons to 3 in the two seasons afterwards (0.71 versus 3.2 per 1000 player-game exposures). Caveat: that is a before-and-after comparison against a historical control, not a randomised trial, so it cannot rule out other changes over those seasons.
• The Copenhagen adduction exercise (a partner- or bench-supported side-lying adduction movement) is the most-studied specific exercise. A 2025 systematic review with meta-analysis in the Scandinavian Journal of Medicine and Science in Sports (Quintana-Cepedal and colleagues, 15 studies) reported increased strength — adduction strength SMD 0.72 (95% CI 0.41 to 1.0), abduction 0.92 (0.47 to 1.4) — and no statistically significant effect on seasonal groin injury prevalence: RR 0.83 (95% CI 0.41 to 1.68). Two heavy caveats, both of which are usually left off when this study is quoted. First, the review's own words: "the quality of evidence underpinning all findings was rated as very low." That certainty rating applies to the strength result as much as to the injury result. Second, the paper was retracted in April 2026 (Scand J Med Sci Sports 36(4):e70287). It is reported here because it is the source the numbers came from and because deleting it would leave the claim uncited, but it should not be relied on. With it withdrawn, the strength claim is currently unsupported for ice hockey specifically — the only remaining trial below is in rink hockey, a different sport, and cannot carry it. The recommendation to train the adductors therefore stands on the risk-factor finding instead (Tyler and colleagues, 2001/2002 — low adductor strength relative to abductor strength, measured in ice hockey players), not on demonstrated strength gains in this sport.
• A 2024 critically appraised topic by the same group concluded there is "Grade B evidence to suggest that inclusion of the CAE may not be associated with reduced injury rates." Note the population: that paper is titled "Can the Copenhagen Adduction Exercise Prevent Groin Injuries in Soccer Players?" — it is a football paper, flagged here as the 2022 cluster-randomised trial below is.
• A 2025 scoping review in Cureus found the exercise in 13 of 19 studies of groin-pain prevention and concluded that "as CAE alone shows limited preventive effects, developing new, multifaceted exercise approaches is essential."
• Not all trials are negative: a 2022 cluster-randomised trial in male high-school footballers found the exercise alone gave a relative risk of 0.42 (95% CI 0.19 to 0.90), and combined with the Nordic hamstring exercise, 0.19 (0.07 to 0.54). Note the population is footballers, not hockey players.
• Dose appears to matter: a 2024 three-arm randomised trial found a higher-volume programme (twice weekly) produced significantly greater adductor strength gains than a low-volume one (+24%). Note the population, because it is not what the citation looks like: the participants were thirty high-level rink hockey players aged 14 — rink hockey is played on quad skates on a hard floor, and involves none of the skating-stride adductor loading this whole section is about. It is a dose-response finding about an exercise, not a finding about ice hockey players.

The honest bottom line. Adductor strengthening appears to make your adductors stronger — the pooled effect sizes are large, but the certainty behind them was rated very low and the review reporting them has since been retracted, so "definitely" and "reliably" overstate it. With that review withdrawn, the strength claim is currently unsupported for ice hockey specifically: what is left is a single 30-player dose-response trial in a different sport (rink hockey), which cannot carry it. What is on firmer ground is the risk factor: low adductor strength relative to abductor strength is a documented risk factor measured in hockey players specifically (Tyler and colleagues, 2001). Whether adding one specific exercise prevents groin injuries is not settled and leans negative — the pooled estimate found no significant effect, individual trials say sometimes yes, most of the trials are in footballers, and the exercise is usually studied in isolation when in practice it would be one part of a programme. Given the Tyler risk-factor finding, deliberate adductor strengthening remains a sensible thing for a hockey player to do; just do not treat it as a guarantee, and if you have current or recurring groin pain, see a physiotherapist rather than self-prescribing an exercise from a document.

Hips — the long-term issue nobody warns you about

Repeated forceful movement in deep hip flexion appears to reshape hockey hips. Among former elite Swedish players in a study of early hip arthroplasty (Abrahamson and colleagues, Journal of Hip Preservation Surgery, 2024), cam morphology — an alpha angle of 60° or more, meaning extra bone at the femoral head–neck junction — was present in 81% of the players. Read that 81% with the sampling in mind: the cohort was players from Sweden's highest league "seeking orthopedic consultation for hip and groin pain with restricted hip joint range of motion", and the figure is from the 31 of them with usable radiographs. It is the prevalence among elite players who presented to a clinic because their hips hurt, not among elite players in general, so it is an upper bound rather than a population rate. A pair of 2026 cross-sectional studies of youth ice hockey and ringette athletes (Martin and colleagues, Clinical Journal of Sport Medicine) found femoroacetabular impingement syndrome in 61% of males and 67% of females, and hip-related groin pain in 63% of males and 57% of females, with athletes in pain showing up to five-fold increased odds of positive findings on impingement tests.

Important interpretation note: these morphological findings are extremely common in athletic populations and are not, on their own, a diagnosis or a reason to panic — plenty of players have cam morphology and no symptoms. What the numbers say is that hip and groin pain in hockey players is common, is not just "a tight groin", and warrants proper assessment rather than more stretching. If you have persistent groin or hip pain, deep-hip pinching on a wide stride, or pain that has lasted more than a couple of weeks, see a physiotherapist or sports physician.

Knee — MCL

The medial collateral ligament (on the inside of the knee) is the classic hockey knee injury, typically from a blow to the outside of the knee or a leg trapped in a collision or against the boards. It was the third most common injury among NHL goaltenders (n = 95) in the 2025 study cited above.

What reduces risk: general lower-body and single-leg strength, balance and landing/deceleration mechanics — the ingredients of neuromuscular training programmes, which have decent evidence (below). Some knee injuries are simply collisions and are not preventable by training.

Shoulder and AC joint

Upper-extremity injuries make up a large share of hockey injuries: Dinh and colleagues' analysis of an estimated 258,302 hockey injuries in players aged 5 to 24 (2006–2023) found 35.57% involved the upper extremity, with the shoulder region the most common site (40.73% of upper-extremity injuries). The fracture-specific study below points the same way — the shoulder is the single most fractured site in the game at 27.0%. The acromioclavicular joint — where the collarbone meets the shoulder blade — is the classic hockey shoulder injury, from landing on the point of the shoulder or being driven into the boards.

What reduces risk: correctly fitted shoulder pads (see Equipment); learning how to take and absorb a hit and how to approach the boards with your head up and hands in front (see Body Contact and Battles); and general upper-back, rotator-cuff and scapular strength. Most of the risk reduction here is skill and awareness, not gym work.

Wrist and hand

The same Dinh dataset puts the wrist at 19.55% and digits at 11.23% of upper-extremity injuries, so roughly three in ten upper-body hockey injuries land on a wrist or a finger. That is the case for taking this section seriously.

One number here is widely misquoted, so read it carefully. Reiad and colleagues' fracture study (players aged 14 to 23, US emergency-department data 2006–2023) found the most common fracture sites were the shoulder (27.0%), wrist (19.3%) and lower arm (11.3%) — shoulder, not hand. You will see that 27.0% attached to the hand; it does not belong to the hand, and it does not support anything in this section — it belongs to the shoulder section above. What the fracture data does support here is that the wrist is the second most fractured site in the game, at 19.3%, and that the same study found college-aged players sustaining hand fractures at nearly three times the rate of high-school players (odds ratio 2.82). The wrist-and-hand section stands on those two numbers and on the Dinh percentages, not on the 27.0%.

What reduces risk: gloves that actually cover the wrist gap (the space between glove cuff and elbow pad is where slashes land), well-fitted elbow pads, and not putting a bare hand out to break a fall.

Concussion risk and body checking

The strongest evidence in hockey injury prevention is not about training at all — it is about rules. The 2025 youth review reports that body checking is the most significant modifiable risk factor, associated with a roughly threefold increase in concussion incidence, and that policy interventions prohibiting body checking have produced 50–70% reductions in injury rates.

What actually reduces injury risk in general

The best-evidenced intervention across team sports is a structured neuromuscular warm-up programme — the FIFA 11+ being the most studied example.

The one attributable number is this: a 2025 systematic review and meta-analysis of multicomponent exercise injury-prevention programmes in adolescent team athletes aged 10–19 (Liu, Liu and Yin, Frontiers in Pediatrics, 16 randomised controlled trials) found total injuries reduced by 35% (IRR 0.65, 95% CI 0.54–0.77), lower-extremity injuries by 33% (IRR 0.67), ankle injuries by 38% and overuse injuries by 39%. Note two things: that review is about multicomponent programmes generally, not the FIFA 11+ specifically, and its population is adolescent team athletes across sports, overwhelmingly not hockey players.

The wider "roughly 30–46% injury reduction" range often quoted for warm-up programmes does have a single, identifiable source: Patel and Shah (2025), "The Impact of the FIFA 11+ Injury Prevention Program on Injury Incidence in Football Athletes: A Systematic Review of Randomized Controlled Trials", Cureus 17(12):e100463, which reports that across the included studies teams doing the FIFA 11+ showed "preventive effects ranging from approximately 30% to 46%." Carry its qualifications: it is a narrative synthesis of five controlled trials, not a meta-analysis and not a pooled estimate — the authors explicitly declined to pool "due to heterogeneity in injury definitions, adherence reporting, and exposure-hour documentation" — and it is football, not hockey. The 35% / IRR 0.65 figure above comes from a separate, larger meta-analytic source and is the more precise of the two.

The finding that matters most is about compliance. In hamstring-injury prevention, high adherence (≥75% of sessions) gave RR 0.36 (0.28–0.48), while low adherence gave RR 0.92 (0.68–1.23) — that is, essentially no benefit. Each 10% increase in adherence corresponded to roughly 17% lower injury risk.

Translation: a mediocre injury-prevention routine you actually do beats an excellent one you do twice. Almost all of these programmes were developed and tested in football, not hockey, so the transfer is an inference rather than a hockey-specific finding — flagged as such.

---

Concussion

Read this section as written. It is the one part of this document with no nuance in it.

A concussion is a brain injury caused by force transmitted to the head — from a hit to the head, a hit to the body that snaps the head, or a fall. The current international reference is the consensus statement from the 6th International Conference on Concussion in Sport (Amsterdam, October 2022), published in the British Journal of Sports Medicine in 2023.

You do not have to be knocked out

Loss of consciousness is not required for a concussion, and is not typical of one. The great majority of concussions involve no loss of consciousness at all. Waiting to see whether someone was "knocked out" before taking it seriously is the single most dangerous misconception in amateur sport.

Red flags — call an ambulance

Read this before anything else in this section. The Concussion Recognition Tool 6 (CRT6), published alongside the Amsterdam consensus for exactly this purpose — for non-medical people on a bench — lists a set of red flags that mean emergency services, immediately. Do not wait, do not drive them yourself, do not "see how they are after the game":

These are the ten red flags CRT6 lists, in full. Learn them as a set:

• Neck pain or tenderness
• Seizure, "fits", or convulsion
• Loss of vision or double vision
• Loss of consciousness
• Increased confusion or deteriorating conscious state — becoming less responsive, drowsy
• Weakness or numbness/tingling in more than one arm or leg
• Repeated vomiting
• Severe or increasing headache
• Increasingly restless, agitated or combative
• Visible deformity of the skull

Any one of these is an ambulance call. Do not attempt to move the athlete — other than as required for airway support — unless you are trained to do so, and do not remove the helmet or other equipment. CRT6 does not make this conditional on neck pain or on the player being unconscious: "Assume a possible spinal cord injury in all cases of head injury." Wait for whoever is trained. (For what to do at the scene of a suspected neck injury, see Body Contact and Battles.)

The ordinary signs

Signs and symptoms that should make you suspect a concussion, where none of them is severe and none is getting worse, include: headache, pressure in the head, dizziness, feeling dazed, stunned or "not right", blurred vision, sensitivity to light or noise, nausea, balance problems, slowed responses, confusion about the score or period, emotional or behavioural change, and memory gaps around the incident.

Any of these still means the player comes off and does not return that day. And re-read the note above: a headache or confusion that worsens stops being an ordinary sign and becomes an emergency.

CRT6 is free. Have it on your phone, and know where the red-flag list is on it before you need it.

The absolute rule

If a concussion is suspected, the player comes off immediately and does not return to play that day. Not that shift, not that period, not that game. "Suspected", not "confirmed" — you do not need certainty, and nobody at a rec-league rink is qualified to rule it out.

There is no version of this that involves the player deciding for themselves. Symptoms can be delayed, judgement is one of the things a concussion impairs, and a concussed player will tell you they are fine. If in doubt, sit them out.

Why returning early is dangerous

A brain that is still recovering is more vulnerable to a second injury, and repeat injury sustained during recovery is associated with worse and more prolonged outcomes. That risk is the reason for the same-day rule and the reason return is staged.

Return to play is a medical decision

The consensus framework uses a graded, stepwise return-to-sport progression: starting with symptom-limited daily activity, progressing through light aerobic exercise, then sport-specific exercise, then non-contact training drills, then — only after medical clearance — full-contact practice, and finally return to game play. Each step depends on tolerating the previous one, and any return of symptoms means stepping back.

This document is not going to give you the timings, and you should be suspicious of any hockey resource that does. The progression must be managed by a qualified medical professional who has assessed the actual player. Concussion recovery varies enormously between individuals, and children and adolescents are managed more conservatively than adults.

What this document will tell you: come off, tell someone, do not drive yourself home, see a doctor, and do not go back on the ice until a medical professional says you can. If you are a coach, captain or parent, that decision is yours to enforce and not the player's to overrule.

The first hours afterwards

These are CRT6's own instructions, and none of them is conditional on how the player says they feel. An athlete with a suspected concussion should not:

• Be left alone initially — at least for the first 3 hours. Worsening of symptoms should lead to immediate medical attention.
• Be sent home by themselves. They need to be with a responsible adult.
• Drink alcohol, use recreational drugs, or take drugs not prescribed by their healthcare professional.
• Drive a motor vehicle until cleared to do so by a healthcare professional. Note the wording: not "until they feel well enough", and not just the drive home from the rink — until a healthcare professional clears them.

This is the part most often skipped, because it happens after everyone has left the rink. Someone has to own it: whoever drives the player home is also the person who stays with them.

---

Warm-Up

What a warm-up actually does

Raising muscle temperature increases the speed of contraction and relaxation, increases nerve conduction speed and reduces muscle stiffness. Practically: you can produce force faster, you move through range more easily, and you are less likely to ask a cold muscle for a maximal effort it was not ready for. There is also a preparation-and-attention element — see Mental Game.

The structured-warm-up evidence above — a 35% reduction in total injuries (IRR 0.65) in adolescent team athletes, mostly outside hockey — is the strongest argument for bothering.

Dynamic, not static, before you play

Static stretching means holding a stretch still for a period. Dynamic stretching means moving actively through a range — leg swings, walking lunges, hip openers, arm circles.

The evidence favours dynamic work before performance:

• A 2025 study in Sports found static stretching had "a detrimental effect when compared to [dynamic stretching] and [no stretching]" for repeated-sprint performance. Effect sizes ES = 0.31–2.26 on mean and total sprint time and ES = 0.89–1.86 on best sprint time — i.e. a range running from small to very large, not uniformly large. The population is ten male Division II soccer players, not hockey players, which is why the practical advice below is hedged.
• A 2025 review in Sports Medicine Open found passive static stretching effective for increasing range of motion but noted it "may transiently reduce maximal isometric strength."

Two honest qualifications. First, the size of the performance decrement is modest in most studies and is associated mainly with long holds — a brief 10–15 second hold as part of a fuller warm-up is unlikely to ruin your game. Second, pre-activity stretching is not an injury-prevention strategy: a 2025 consensus-style paper in the Journal of Sport and Health Science concluded that stretch training "does not serve as an all-encompassing injury prevention strategy." Stretch for range of motion, at a separate time. Warm up with movement.

The off-ice warm-up most amateurs skip entirely

This is the single most commonly skipped thing in adult recreational hockey. Players arrive fifteen minutes before puck drop, get dressed, and step onto the ice cold to do a hard crossover lap.

Ten to fifteen minutes before you get dressed is enough:

1. Raise (3–4 min) — light jogging on the spot, skipping, or a fast walk around the concourse. Get warm enough to notice it.
2. Mobilise (3–4 min) — leg swings front-to-back and side-to-side (the side-to-side ones are your groin), walking lunges with a rotation, hip openers, deep squat holds, arm circles and shoulder rolls, thoracic rotations.
3. Activate (2–3 min) — glute bridges, lateral band walks, a few bodyweight squats. This wakes up the muscles that the stride depends on.
4. Potentiate (2–3 min) — a few short accelerations, jumps or hops, building to near-full effort. This is the part that gets your nervous system ready to produce force fast, and it is the part people leave out.

The structure above is a standard warm-up framework (commonly taught as RAMP), applied to hockey. It is coaching craft in its specifics; the evidence supports structured warm-ups in general rather than this exact ordering.

The on-ice warm-up

Most rec games give you three to five minutes. Spend it deliberately:

• Skate first, shoot later. Two or three laps building from easy to brisk, including backwards, crossovers both directions, and a few hard starts and stops. Your edges need waking up more than your wrist shot does.
• Then handle the puck — a few stickhandling reps and short passes to get your hands and your feel for the puck going.
• Then shoot, starting easy. Do not open with a full slap shot from the blue line; that is a cold trunk producing a maximal rotation.
• Give your goalie proper shots. Not the top corner from the slot on the first shot. Start with easy shots from distance and work in. Your goalie is warming up too, and they are the one player who cannot ease into the game. See Goaltender.
• A few hard efforts. One or two five-second full-speed accelerations near the end so your first shift is not your warm-up.

---

Fuelling and Hydration

Hockey has one property that makes fuelling matter more than the modest playing time suggests: as the Vigh-Larsen review notes, glycogen depletion is pronounced "despite the relatively brief playing time". A 2026 review of nutrition for elite male ice hockey in the International Journal of Sport Nutrition and Exercise Metabolism states that carbohydrate, primarily muscle glycogen, is the major fuel for training and match play, and that match performance has been shown to be associated with pre-game muscle glycogen content.

Before the game

• The main pre-game meal, 3 to 4 hours before, built around carbohydrate (rice, pasta, potatoes, bread), with some protein and low fat and fibre — fat and fibre slow digestion and are the usual culprits behind a heavy stomach.
• A small carbohydrate top-up 1 to 2 hours before if needed: a banana, toast, a cereal bar. Keep it small and familiar.
• Never experiment on game day. Test any new food or drink at practice.
• Start hydrated. In a study of para hockey players, 40% arrived at competition already dehydrated by urine specific gravity. Drinking a glass of water an hour before is worth more than drinking a litre at the rink.

During the game

Ice hockey players sweat more than the cold rink suggests. The elite-hockey nutrition review reports sweat losses of 2.02 ± 0.74 litres per hour. A 2025 study of 25 NCAA Division I players compared measurements taken the mornings of and the mornings after games, and found a total body water deficit of 0.63 litres (1.2%) still present the next morning — that is a shortfall that survived the whole post-game evening, not the loss during the game itself, which will have been larger.

• Drink on the bench, every shift or two. Small, regular amounts rather than one big drink between periods. (The measured sweat and body-water figures above are sourced; the little-and-often pattern is the conventional practitioner recommendation, not something those studies tested.)
• Water is fine for most rec games. For long, hot or high-intensity games, a drink with carbohydrate and sodium (a standard sports drink) has a real rationale: it replaces fluid, adds fuel and replaces salt.
• Between periods is your best window for a slightly larger drink and, in a long tournament day, a small carbohydrate snack.

The late-evening rec game problem

This is the specific case most adult players face, and it is genuinely awkward. A 22:30 puck drop creates three problems at once:

1. You have not eaten for hours. Dinner at 18:30 and a 22:30 game means playing on largely empty tanks. Eat a proper carbohydrate-containing meal 3 to 4 hours before, and take a small, easily digested top-up an hour or so before if you feel empty.
2. You are already tired. You have worked all day. Accept that your ceiling is lower and manage shift length accordingly — see Game Management.
3. You cannot sleep afterwards. You finish at midnight, adrenalised, dehydrated and hungry. This is the real cost. Practical mitigations: rehydrate and eat something during and immediately after the game rather than at 01:00; keep post-game screens and bright light down; keep the post-game beer modest, since alcohol both dehydrates you further and measurably degrades sleep quality; and give yourself a wind-down routine rather than going straight from the rink to bed.

Be realistic: if you play at 22:30 on a work night, the sleep loss — not the training — is your main recovery limiter, and it is worth planning the next morning around.

---

Recovery

Sleep is the highest-leverage thing on this list

Nothing else in this document has the effect size of sleep, and nothing else is as often ignored.

The evidence base is more mixed than sleep advocates suggest, but it points consistently in one direction. A 2025 multidimensional review in the Journal of Clinical Medicine describes optimal sleep duration and quality as a critical "regenerative window" for athletic performance. Work on sleep extension finds that pre-emptive extra sleep improves alertness and vigilance during subsequent sleep restriction. Research on sleep and injury has found shorter sleep duration significantly associated with higher injury risk (OR 1.34, 95% CI 1.08–1.66) — though the specific analysis quoted here was a preprint, so treat the precise number as provisional while the direction of the finding is consistent with the wider literature.

What to actually do:

• Aim for a consistent 7 to 9 hours — the general adult sleep recommendation, carried in here rather than derived from any of the hockey studies cited above — with the emphasis on consistent — a regular sleep and wake time is easier to sustain than a heroic occasional lie-in.
• Protect the night before a game more than the night after, if you have to choose.
• Bank sleep before a heavy stretch (a tournament, a run of late games) rather than trying to repay it afterwards.
• A 20–30 minute nap is a legitimate tool on a day with an evening game.

Post-game nutrition

The goals are to replace fluid, replace glycogen and provide protein for repair.

• Carbohydrate is the priority for refuelling, and one 2025 survey found athletes systematically underestimate this, over-emphasising protein — only 1.8% of participants correctly identified the carbohydrate content required for rapid glycogen resynthesis. The respondents were 113 amateur Hungarian endurance athletes — runners and triathletes, not hockey players — so read it as evidence that recreational athletes generally get this wrong, not as a hockey-specific measurement.
• Include protein in your post-game meal for repair and adaptation.
• Replace fluid, using how you feel and the colour of your urine as a rough guide.
• Timing matters less than total intake for most amateur players. The idea of a narrow "anabolic window" of 30 minutes is far weaker than it was once presented; if your next game or practice is more than a day away, getting a good meal in within a couple of hours is entirely sufficient. Rapid refuelling matters most when you have another game the same day or the next morning — a tournament weekend, not a Tuesday night league.

Active recovery

Easy movement the day after — a walk, an easy spin on a bike, easy skating — is low-risk, feels good, and has reasonable support for short-term restoration. One 2026 network meta-analysis concluded active recovery "may be preferable for short-term restoration of explosive performance." Do not make it hard enough to be a training session.

An honest audit of popular recovery methods

Sports science's recovery literature is large, noisy and full of small studies. Here is what the evidence actually looks like, method by method.

Sleep — strong. Best supported, cheapest, most neglected. Everything below is a rounding error next to it.

Nutrition and hydration — strong. Well-established mechanisms and direct relevance to a sport that measurably depletes glycogen and loses two litres of sweat an hour.

Cold water immersion / ice baths — mixed, and with a genuine catch.

• Multiple recent reviews find CWI reduces perceived soreness and improves some markers: a 2025 systematic review of recovery modalities in soccer players (Vatne and colleagues, Sports, 41 studies) found that cold-water immersion "consistently improved jump performance and perceptions of fatigue, soreness, and overall well-being"; a 2026 soccer-specific review found it "may enhance recovery of muscle strength, reduce muscle damage, and alleviate soreness... but does not seem to impact sprint performance."
• Other analyses are less impressed: a 2026 meta-analysis found effects "outcome-specific", reducing subjective soreness while the apparent creatine-kinase benefit "was not robust after adjustment for publication bias". The same analysis found a transient impairment of explosive performance — countermovement jump was worse immediately after immersion (g = −0.68 at 0 h) but the effect had gone by 24–48 hours, and the overall effect across time points was not significant.
• The catch: a 2026 network meta-analysis concluded that "passive recovery seems to be more desirable than CWI when it comes to resistance training aimed at increasing strength without disrupting adaptive responses," and a 2025 mechanistic study in Medicine and Science in Sports and Exercise found post-exercise cooling "greatly reduces muscle microvascular perfusion and blunts postprandial amino acid incorporation in muscle."
• Practical verdict: reasonable after a game or a tournament when the goal is to feel better and play again soon. Questionable after a strength session when the goal is to adapt. It is a recovery tool, not a training tool, and the difference matters.

Compression garments — contested; graded weak here, but honestly so. A 2026 study found compression garments "had no impact on repeated sprint performance or exercise-induced leg soreness", concluding "the efficacy of compression garments for recovery remains inconclusive." That study is twelve adolescent male footballers wearing one garment for one 90-minute protocol, so it is thin ground for a grade — and the 2023 Driller and Leabeater review cited elsewhere in this document places compression garments in its high tier of positive evidence, alongside foam rolling, which this document reports approvingly in the next paragraph. Take the disagreement at face value: a broad narrative review says the evidence base is favourable, a recent controlled trial found nothing, and this document leans on the trial because it is the more direct test. Harmless, comfortable, and genuinely unresolved.

Foam rolling and massage — modest, and mostly on how you feel. Foam rolling appears in recovery reviews with reasonably positive evidence, and massage shows benefit for early soreness relief. But be clear what the benefit is: short-term improvements in perceived soreness and range of motion. Evidence that foam rolling improves subsequent performance, or prevents injury, is weak. That is not nothing — feeling less sore makes you likelier to train — but it is not tissue repair, and it does not "break up adhesions".

Static stretching as recovery — weak. There is little good evidence that post-exercise stretching meaningfully reduces muscle soreness. Stretch because you want range of motion, not because you think it undoes a hard game.

A general warning about this literature. Most recovery studies are small, short, hard to blind, and measure surrogate outcomes (soreness scores, blood markers) rather than "did you play better next week". Where two reviews of the same method disagree — which happens constantly above — the honest reading is the effect is probably small.

---

In-Season versus Off-Season

The classic amateur pattern is to do nothing all summer, panic in September, train hard for six weeks, get injured or bored, and then coast through the season. The classic slightly-more-serious pattern is the mirror image: hammer the gym in-season until you are too sore to skate, then take four months off. Both waste the year.

Off-season — build

• Aim: the biggest changes. Aerobic base, maximum strength, power, and correcting mobility deficits and asymmetries. Higher volume, more fatigue tolerated, because there is no game on Thursday to be fresh for.
• This is also the window for addressing a nagging issue properly with a physiotherapist, rather than managing it around games.
• Do not take the whole off-season off. Detraining is real; the strength and aerobic fitness you built does not keep.
• Keep skating occasionally if you can. Skating-specific conditioning does not transfer perfectly from a bike.

Pre-season — sharpen

• Shift the emphasis from base-building toward hockey-shaped intervals (the 30–45 seconds hard, 2–4 minutes easy pattern), power, and getting on the ice.
• Reduce total volume so you arrive fresh rather than fried.

In-season — maintain, and prioritise being fresh

• The mistake to avoid is training in-season as if it were the off-season. In-season the games are the hard conditioning. Your gym work exists to preserve what you built, not to add to it.
• Two quality strength sessions a week is a reasonable maintenance dose, though the evidence for the specific number is thinner than it is usually made to sound. A 2026 meta-analysis found "low frequency (≤2 sessions per week) and a total of around 12 sessions appear to facilitate better athletic adaptations." Three caveats: it is 7 randomised trials and 199 male athletes; the authors rated the evidence quality as low; and it is specifically about flywheel resistance training compared with traditional resistance training, not about barbell training frequency in general. It supports "twice a week is enough", not "twice a week is proven optimal". A 2026 study in elite young female footballers found a 12-week in-season strength programme at moderate intensities improved maximal strength "without altering resting levels of muscle damage or inflammatory markers" — that is, you can lift in-season without wrecking yourself.
• Keep intensity, cut volume. Fewer sets, still challenging loads. That is the standard maintenance principle, and it fits the data above.
• Do not lift heavy legs the day before a game. Put your hardest session as far from game day as your schedule allows.
• Keep some easy aerobic work. It is the cheapest thing to maintain and the first thing people drop.

---

Shift Length: Where Conditioning Becomes Tactics

Everything above converges on one moment: the last ten seconds of a long shift.

Here is the chain. Your phosphocreatine store is largely spent in the first fifteen to twenty seconds of hard skating. From there you are increasingly running on the glycolytic system, and the by-products of that accumulate. Heart rate is at or near maximum. And critically — the thing that degrades first is not your legs, it is your decision-making.

A tired player does not usually fall over. A tired player:

• stops scanning, because scanning costs attention and attention is expensive when you are gassed (see Scanning and Anticipation);
• takes the easy pass instead of the right one;
• stops skating back and starts gliding, which is how odd-man rushes happen;
• reaches with the stick instead of moving the feet, which is how tripping and hooking penalties happen;
• and stays out for another twenty seconds, because deciding to change is itself a decision.

This is why conditioning is a hockey IQ issue and not just a fitness issue. Game Management covers the tactical side — short shifts, changing on the right side of the puck, when a long shift is unavoidable and how to survive one. The physical side is this: the fitter you are, the longer your judgement lasts, and judgement is what you are actually paid in.

The practical rule that follows: change before you need to, not when you need to. A 40-second shift at full quality beats a 70-second shift where the last 30 were passengers. If you find yourself gliding, you needed to change fifteen seconds ago.

---

Training for the Adult Recreational Player

You have a job, possibly children, and one or two skates a week. Here is the realistic version.

The principles

1. Consistency beats intensity. Three sessions a week for a year beats six sessions a week for six weeks followed by nothing. The adherence data in the injury-prevention section makes this a measured finding, not a platitude.
2. Your games are your interval training. If you skate twice a week you are already getting hockey-shaped anaerobic work. You do not need to add much more.
3. Therefore your off-ice time should mostly be the things games don't give you: strength, aerobic base, and hip and groin resilience.
4. Skating technique is a better investment than fitness if you are new. An efficient stride reduces the energy cost of everything.

The highest-value 90 minutes a week

Split as two sessions plus a bit. This is a reasoned allocation from the evidence above, not a validated programme — treat it as a sensible default, and get form coaching on the lifts.

Session 1 — Strength (45 minutes)

• 5 min warm-up (raise, mobilise, activate)
• A squat or hinge pattern — goblet squat, trap-bar deadlift or barbell squat. 3 sets of 5–8.
• A single-leg pattern — split squat, step-up or lunge. 3 sets of 8 each leg. Skating is single-leg; train it single-leg.
• A push and a pull — press and row. 3 sets of 8 each.
• Adductor work — Copenhagen adduction or a regressed version (short-lever, knee-supported), plus a squeeze isometric. Start very conservatively; this exercise makes people extremely sore if they dive in.
• Anti-rotation core — Pallof press or a loaded carry. 3 sets.

Session 2 — Conditioning (30 minutes) Alternate week to week:

• Week A — hockey intervals. 8 × 40 seconds hard on a bike or rower, 2–3 minutes easy between. This is the game, off the ice.
• Week B — easy aerobic. 30 minutes genuinely easy, conversational. This is the between-shift recovery engine and the session people skip because it feels pointless. It is not pointless.

The remaining 15 minutes — mobility and prehab, split into small doses

• Hip and groin mobility, glute activation, and a few minutes of shoulder and upper-back work. Three 5-minute doses across the week beat one 15-minute session, and you can do them in front of the television.

What to cut if you only have 45 minutes

Keep Session 1 (strength) and do your conditioning in your games. Strength and adductor work are the things your hockey does not provide. Aerobic fitness is the thing you will miss in the third period, so if you can add anything, add a single easy 30-minute aerobic session — it is the lowest-effort, highest-return addition available.

What not to do

• Do not add a hard conditioning session the day before a game. You will play worse and gain nothing.
• Do not go from zero to a full programme in one week. The most common way an adult amateur gets hurt is enthusiasm.
• Do not neglect the warm-up because you are short of time. It is the highest-evidence injury item here.

---

Youth Players

This section is deliberately brief, because youth physical development is a field with qualified specialists in it and this document is not one of them. If you coach or parent a young player, the single most useful thing in this section is the recommendation to work with qualified coaches and, where relevant, medical professionals.

Youth training is not scaled-down adult training. Young athletes are growing, their skeletal growth plates are vulnerable in ways adult bone is not, and the appropriate emphasis is on broad athletic development — coordination, agility, balance, movement skill, speed — rather than on maximising sport-specific fitness. A 2026 review notes that "growth-related weak points such as physes and traction apophyses predispose to unique injury patterns" and that injury risk in youth sport "is driven not by load alone but by the interaction between developmental biology and sporting exposure."

Avoid early specialisation. The consensus recommendations in the youth sports medicine literature are consistent:

• Avoid specialising in a single sport before around age 12.
• Avoid training more hours per week than the athlete's age in years, with a general cap around 16 hours per week.
• Play another sport in the off-season.

Note: the specific numeric recommendations above appear in a 2026 review in Sports Health restating widely cited youth sports medicine guidance — but note the sport: the paper is "Early Sport Specialization and Intense Training in Junior Tennis Players", and its tennis-specific recommendations (tournament counts, weekly on-court hours) are not reproduced here. The three figures quoted above are the general youth-sport ones it restates, not tennis-only numbers. They are consensus recommendations rather than experimentally derived thresholds, and are flagged as such.

Supporting evidence is consistent in direction: early specialisation has been associated with increased odds of chronic injury in college students, and multi-sport participation in high school with lower injury rates and greater career durability among NFL athletes. Burnout is a real and documented risk, not just an injury one.

On the hip finding specifically: the cam morphology and impingement data earlier in this document come from populations who played a great deal of hockey from a young age — and, in the case of the 81% cam figure, from players who had already presented to a clinic with hip and groin pain, so it overstates what a typical player's hip looks like. Even discounted for that, it is a reason for variety in a young player's athletic diet, and a reason to take a young player's persistent hip or groin pain seriously rather than treating it as growing pains.

Resistance training for youth is not inherently dangerous — supervised, well-coached, technique-focused strength training is generally regarded as safe and beneficial for young athletes. The risk lies in poor supervision and inappropriate loading, not in the concept. Get qualified coaching.

---

Common Mistakes

• Training only the aerobic system — long runs and nothing else — and then being blown out by the first hard shift.
• Training only the anaerobic system — all sprints, all lifting, no easy aerobic work — and then disappearing in the third period. This is the more common error among competitive amateurs.
• Judging your fitness by a lab number. VO₂max showed no significant relationship with repeated on-ice effort performance in the study cited above; train the quality, not the test.
• Skipping the off-ice warm-up entirely, arriving fifteen minutes before puck drop, and taking your first hard stride cold. The highest-evidence injury-prevention measure available, ignored for want of ten minutes.
• Static stretching hard immediately before playing, which has measured short-term decrements in strength and repeated-sprint performance and does not prevent injury.
• Ignoring the groin until it goes. Low adductor strength relative to abductor strength is a documented hockey risk factor, and the stride trains the opposite muscles.
• Assuming any single exercise is a guarantee. The Copenhagen adduction exercise appears to build strength — on very-low-certainty evidence — and the pooled estimate shows no significant reduction in groin injury prevalence on its own.
• Self-treating persistent hip or groin pain with more stretching, for months, instead of getting it assessed.
• Playing on after a head knock because you weren't knocked out. Loss of consciousness is not required for a concussion and is not typical of one.
• Letting the player decide whether they're concussed. Judgement is one of the things a concussion impairs.
• Sending them home alone, or letting them drive. CRT6 says not left alone for at least the first 3 hours, not sent home by themselves, and no driving until a healthcare professional clears them — none of it conditional on how they say they feel.
• Helping a downed player up after a head knock. Assume a possible spinal cord injury in all cases of head injury: do not move them unless you are trained to, and do not remove the helmet.
• Training hard in-season and doing nothing in the off-season — or the reverse. Both waste most of the year.
• Lifting heavy legs the day before a game and wondering why your first step is gone.
• Underestimating sweat loss because the rink is cold. Elite players have measured sweat losses of around two litres an hour.
• Not eating before a late game, then not eating or drinking after it, then not sleeping.
• Buying recovery. Compression garments, ice baths and massage guns are, at best, small effects. Sleep is free and larger than all of them combined.
• Ice-bathing straight after a strength session, where the goal is adaptation and cooling appears to blunt it.
• Staying out for 75 seconds. Conditioning is what keeps your decision-making alive in the last ten seconds of a shift; the cheapest way to protect it is to change earlier.

Key Takeaways

1. Hockey is anaerobic within a shift and aerobic across a game. Shifts of roughly 30–80 seconds at near-maximal heart rate, with longer passive recovery. Anaerobic power sets how good your best shift is; aerobic fitness sets how similar your twentieth is to your first.
2. The reason aerobic fitness matters is rebuilding, not endurance. Restoring phosphocreatine between shifts is an oxygen-dependent process. That is why a strong, unfit player fades and a fit, weak player never had a first shift worth fading from.
3. Train five things: aerobic base, hockey-shaped intervals (30–45 seconds hard, 2–4 minutes easy), lower-body strength and power, core and rotational strength, and hip and groin mobility with strength through range.
4. Your groin is probably weak relative to your abductors, and that is a documented hockey risk factor — a player with adductor strength below 80% of abductor strength was found to be 17 times more likely to strain an adductor. Strengthen it deliberately, but know that no single exercise is proven to prevent the injury, and that the pooled evidence on the Copenhagen exercise was rated very low certainty in a review that has since been retracted — leaving the strength claim currently unsupported for ice hockey specifically.
5. Learn all ten concussion red flags, and call an ambulance for any one of them: neck pain or tenderness · seizure, fits or convulsion · loss of vision or double vision · loss of consciousness · increased confusion or deteriorating conscious state · weakness or numbness/tingling in more than one arm or leg · repeated vomiting · severe or increasing headache · increasingly restless, agitated or combative · visible deformity of the skull. A headache or confusion that is getting worse is a red flag, not an ordinary symptom. Assume a possible spinal injury in any head injury: do not move them. Short of a red flag, suspected concussion still means off the ice, no return that day, no driving, not left alone for the first few hours, and a medical professional before you play again — you do not need to lose consciousness to have one.
6. The warm-up is the highest-evidence injury measure you control, and compliance beats programme design: high adherence cut injury risk by about 64% in the hamstring-prevention research (RR 0.36) while low adherence gave essentially no benefit. Ten to fifteen minutes off-ice — raise, mobilise, activate, potentiate — then skate before you shoot.
7. Sleep is the single highest-leverage recovery tool, and it is the main casualty of a 22:30 rec game. Plan around it, rehydrate and eat at the rink rather than at 01:00, and keep the post-game beer modest.
8. Be sceptical of recovery products. Ice baths help you feel better and may blunt strength adaptation; compression is inconclusive; foam rolling mostly changes how sore you feel. Carbohydrate, fluid, protein and sleep are where the evidence is.
9. In-season, maintain; off-season, build. Two quality strength sessions a week maintains what you built. Doing nothing all summer throws it away.
10. Conditioning is a decision-making asset. What fails at the end of a long shift is your judgement before your legs — so change before you need to, not when you need to. For youth, the rule is variety, not specialisation: no single-sport specialisation before about age 12, weekly training hours below the athlete's age in years, qualified coaches rather than adult programmes scaled down.

Sources — retrieved 27 July 2026

Hockey physiology and match demands: Vigh-Larsen JF, Mohr M — "The physiology of ice hockey performance: An update", Scand J Med Sci Sports, 2024 (PMID 36517860) · Gabrys T et al. — "Energy Structure of Repeated On-Ice Efforts and Its Dependence on the Aerobic Capacity of a Hockey Player", Sports, 2026 (PMID 41893607) · Lögdal N, Laaksonen MS, Andersson EP — "Individual Fluctuations in Blood Lactate During Ice Hockey", Int J Exerc Sci, 2022 (PMID 36159342) · Rago V et al. — "Game Demands of a Professional Ice Hockey Team", J Hum Kinet, 2022 (PMID 36457463) · Lignell E, Fransson D, Krustrup P & Mohr M — "Analysis of High-Intensity Skating in Top-Class Ice Hockey Match-Play in Relation to Training Status and Muscle Damage", J Strength Cond Res 32(5):1303–1310, 2018 (PMID 28557852) — the source for the forwards-versus-defencemen distance and intensity split above; abstract retrieved and quoted directly, 27 July 2026

Phosphocreatine resynthesis and oxygen: Haseler LJ, Hogan MC, Richardson RS — "Skeletal muscle phosphocreatine recovery in exercise-trained humans is dependent on O2 availability", J Appl Physiol, 1999;86(6):2013–18 (PMID 10368368) · Harris RC, Edwards RHT, Hultman E, Nordesjö LO, Nylind B, Sahlin K — "The time course of phosphorylcreatine resynthesis during recovery of the quadriceps muscle in man", Pflügers Arch, 1976;367(2):137–42 (PMID 1034909) · Tomlin DL, Wenger HA — "The relationship between aerobic fitness and recovery from high intensity intermittent exercise", Sports Med, 2001;31(1):1–11 (PMID 11219498)

Shifts per game (primary data): NHL public stats API — skater time-on-ice report, 2024-25 regular season; the shiftsPerGame and timeOnIcePerShift fields, aggregated here over the 587 skaters with 41 or more games played

Off-ice training and transfer to skating: Keiner M et al., J Strength Cond Res, 2024 (PMID 38335067) · Glaude-Roy J et al., Eur J Sport Sci, 2024 (PMID 39256936) · Roczniok R et al., J Hum Kinet, 2024 (PMID 39132412)

Groin and adductor: Tyler TF, Nicholas SJ, Campbell RJ, McHugh MP — "The association of hip strength and flexibility with the incidence of adductor muscle strains in professional ice hockey players", Am J Sports Med, 2001 (PMID 11292035) · Tyler TF et al. — "The effectiveness of a preseason exercise program to prevent adductor muscle strains in professional ice hockey players", Am J Sports Med, 2002 (PMID 12239001) · Quintana-Cepedal M et al. — "The Copenhagen Adduction Exercise Effect on Sport Performance and Injury Prevention: A Systematic Review With Meta-Analysis", Scand J Med Sci Sports, 2025;35(8) (PMID 40827942) — ⚠️ RETRACTED (retraction notice: Scand J Med Sci Sports. 2026 Apr;36(4):e70287). The review also rated "the quality of evidence underpinning all findings" as VERY LOW. Both facts are stated inline in the text; the numbers are reported only because they are the source of the figures quoted there · Quintana-Cepedal M, de la Calle O, Olmedillas H — "Can the Copenhagen Adduction Exercise Prevent Groin Injuries in Soccer Players? A Critically Appraised Topic", J Sport Rehabil, 2024 (PMID 37734743) — football population, flagged inline · Saito H et al., Cureus, 2025 (PMID 41583230) · Fujisaki K et al., Int J Sports Phys Ther, 2022 (PMID 35949380) · Quintana-Cepedal M et al. — dose-response of the Copenhagen adduction exercise, J Sports Sci, 2024 (PMID 39551738) — population is 30 high-level rink hockey players aged 14, not ice hockey; flagged inline

Hockey injury epidemiology: Keshen S et al. — "Musculoskeletal Injuries in National Hockey League Goaltenders", Am J Sports Med, 2025 (PMID 40653669) · Zheng Y et al. — "Research Progress on Common Sports Injuries Among Youth Ice Hockey Players", Sports, 2025 (PMID 41441433) · Abrahamson J, Lindman I, Jónasson P, Tegner Y — "High prevalence of former elite ice hockey players requiring early hip arthroplasty surgery", J Hip Preserv Surg, 2024 (PMID 39664208) — the 81% cam figure is from 31 players who sought orthopaedic consultation for hip and groin pain; a symptomatic clinic sample, flagged inline · Martin M, Bullock GS, Galarneau JM, Schneider G, Emery CA, Mohtadi NG — "Examining Symptoms, Clinical, and Radiographic Signs of Femoroacetabular Impingement Syndrome in Youth Ice Hockey and Ringette Athletes, Part 2", Clin J Sport Med, 2026;36(2) (PMID 41931369) · Martin M, Soligon C, Galarneau JM, et al. — "Evaluating the Prevalence of Signs and Symptoms Associated With Hip-Related Groin Pain in Youth Ice Hockey and Ringette, Part 1", Clin J Sport Med, 2026 (PMID 41947288) · Laaksonen J, Pakarinen O, Vaajala M, Liukkonen R, Kuitunen I — "Lower limb injuries and concussions dominate in collegiate women's ice hockey: a systematic review and meta-analysis", BMJ Open Sport Exerc Med, 2026;12(1):e002684 (PMID 41657371) · Dinh J, Reiad M, Wang J, et al. — "Breaking the Ice: Patterns and Prevalence of Upper Extremity Injuries in Youth Ice Hockey Players Over a 20-Year Study Period", Hand (NY), 2026;21(7):1197–1205 (PMID 40741926) · Reiad M, Dinh J, Khan A, et al. — "Epidemiology of acute ice-hockey-related orthopedic fractures in high school- and college-aged players in the United States from 2006-2023", Eur J Trauma Emerg Surg, 2026;52(1):117 (PMID 41925829) — shoulder 27.0%, wrist 19.3%, lower arm 11.3%; the 27.0% is frequently misquoted elsewhere as a hand figure

Concussion: Patricios JS et al. — "Consensus statement on concussion in sport: the 6th International Conference on Concussion in Sport — Amsterdam, October 2022", Br J Sports Med, 2023 (PMID 37316210, doi:10.1136/bjsports-2023-106898) · Concussion Recognition Tool 6 (CRT6), Echemendia RJ et al., Br J Sports Med 2023;57:692–694 (doi:10.1136/bjsports-2023-107021) — the ten red flags, the "Remember" box ("Do not attempt to move the athlete (other than required for airway support) unless trained to do so"; "Assume a possible spinal cord injury in all cases of head injury"), and the "Athletes with suspected concussion should NOT" list (not left alone for at least the first 3 hours; not sent home by themselves; no alcohol or recreational drugs; no driving until cleared by a healthcare professional). CRT6 is free to copy in its current form and belongs on your phone · SCAT6, published alongside the same consensus

Warm-up and injury prevention programmes: Zarei M et al. — adherence to FIFA 11+, Am J Sports Med, 2026 (PMID 41981827) · Gu J et al. — adherence and hamstring injury, Ann Med, 2026 (PMID 41945463) · Liu H, Liu X, Yin L — "Effects of multicomponent exercise injury prevention programs on adolescent team athletes (10-19 years old): a systematic review and meta-analysis", Front Pediatr, 2025;13:1561993 (PMID 41574355) — the source of the IRR 0.65 / 35% figure; adolescent team athletes across sports, not FIFA 11+ specifically and not hockey · Eser C et al. — FIFA 11+ and ankle injury, Muscles, 2025 (PMID 40843917) · Patel P, Shah M — "The Impact of the FIFA 11+ Injury Prevention Program on Injury Incidence in Football Athletes: A Systematic Review of Randomized Controlled Trials", Cureus, 2025;17(12):e100463 (PMID 41625863) — the source of the "30% to 46%" range; football players, narrative synthesis rather than meta-analysis, flagged inline

Stretching: Aouadi R et al. — static vs dynamic stretching and repeated sprint, Sports, 2025 (PMID 40863784) · Longo S et al., Sports Med Open, 2025 (PMID 41201748) · Warneke K et al. — stretch training and injury prevention, J Sport Health Sci, 2025 (PMID 40513717)

Nutrition and hydration: Nordstrøm A — "Nutrition for Elite Male Ice Hockey", Int J Sport Nutr Exerc Metab, 2026 (PMID 42362109) · Heesch AJ et al. — body water changes in collegiate ice hockey, J Strength Cond Res, 2025 (PMID 40267420) · Gavel EH, Rae S, Logan-Sprenger HM — fluid balance in para hockey, Front Sports Act Living, 2024 (PMID 39654844) · Csanaky M et al. — athlete knowledge of recovery nutrition, Nutrients, 2025 (PMID 41305679)

Recovery methods: Vatne E, Oliva-Lozano JM, Saenz C, Cost R, Hagen J — "Post-Exercise Recovery Modalities in Male and Female Soccer Players of All Ages and Competitive Levels: A Systematic Review", Sports (Basel), 2025;13(10):343 (PMID 41150478) — soccer population, flagged inline; source of the "consistently improved jump performance and perceptions of fatigue, soreness, and overall well-being" quotation · Yu T et al. — CWI protocol optimisation network meta-analysis, BMC Sports Sci Med Rehabil, 2026 (PMID 41845491) · Betz MW et al. — "Postexercise Cooling Lowers Skeletal Muscle Microvascular Perfusion and Blunts Amino Acid Incorporation into Muscle Tissue", Med Sci Sports Exerc, 2025 (PMID 40249909) · Veen J et al. — CWI in soccer players, Scand J Med Sci Sports, 2026 (PMID 41490103) · Zhu Y et al. — CWI outcome-specific effects, PeerJ, 2026 (PMID 42473449) · Engel F et al. — compression garments, Physiol Rep, 2026 (PMID 41693319) · Hou Y et al. — comparative recovery modalities, Healthcare, 2026 (PMID 42194413) · Driller M, Leabeater A — recovery strategies review, Sports, 2023 (PMID 37999430)

Sleep: Kaczmarek F et al. — "Sleep and Athletic Performance: A Multidimensional Review", J Clin Med, 2025 (PMID 41227002) · Juginović A, Rodman L — sleep banking, Clocks Sleep, 2026 (PMID 41892386) · Tawfeeq R, Malm CB — sleep duration and injury risk, preprint, 2025 (PPR1110297)

In-season and off-season training: Qu W et al. — in-season flywheel resistance training dose-response, BMC Sports Sci Med Rehabil, 2026 (PMID 41546061) · Bousselmi M et al. — 12 weeks of in-season strength training, Sports, 2026 (PMID 42043068)

Youth development: Thurber L, Kantrowitz DE, Wang KC, Jayanthi N, Colvin A — "Early Sport Specialization and Intense Training in Junior Tennis Players: A Sport-Specific Review", Sports Health, 2026 (PMID 41272920) — tennis population; the three numeric thresholds quoted are the general youth-sport recommendations it restates, flagged inline · Zinke TJ et al. — early specialisation and chronic injury, J Sport Rehabil, 2026 (PMID 42463116) · Chundi G et al. — specialisation and NFL career outcomes, Eur J Sport Sci, 2026 (PMID 41527183) · Li L et al. — physiological mechanisms in youth musculoskeletal injury, Front Physiol, 2026 (PMID 42266238)

The literature here is drawn from the Europe PMC REST API (https://www.ebi.ac.uk/europepmc/). Individual publisher pages (bjsm.bmj.com and others) are not publicly accessible, so abstracts and the quoted passages come from Europe PMC records rather than publisher full text.

Limits and open questions — flagged inline:

• Full text of the Amsterdam 2022 concussion consensus. The publisher's site (bjsm.bmj.com) is not publicly accessible, so the recognition, immediate-removal, no-same-day-return, graded-return and medical-clearance principles here rest on the Europe PMC record for PMID 37316210 rather than the full text; the exact wording of the stages and any stated minimum durations are not reproduced. Deliberately, no timings are given in this document — the consensus requires that progression be individually managed by a medical professional.
• Shifts per game. The NHL's public stats API publishes a shiftsPerGame field per skater. The figures in the text are computed from the NHL time-on-ice report for the 2024-25 regular season, restricted to the 587 skaters with 41 or more games played — forwards 19.4 shifts per game (median 19.6), defencemen 24.0 (median 24.4), all skaters 21.0, with a 10th-to-90th-percentile spread of 16.0 to 26.1 and mean shift lengths of 47.1 s for forwards and 48.8 s for defencemen. This is NHL regular-season data and should not be read across to junior, women's or rec hockey. The 30–80 second shift range and 15–25 minutes of on-ice time are sourced separately (Vigh-Larsen and Mohr 2024), and the bench-recovery figure is sourced too: the elite-hockey nutrition review (Nordstrøm 2026) describes shifts "interspersed with passive recovery periods (2-5 min)".
• The 8 × 40 seconds interval prescription and the 90-minute weekly plan. These are coaching craft, reasoned from the measured game demands, and are labelled as such in the text. They are not validated protocols.
• The RAMP warm-up structure. The general efficacy of structured neuromuscular warm-ups is well supported; the specific four-stage ordering used here is a widely taught coaching framework, not a tested hockey-specific protocol.
• Phosphocreatine resynthesis as an oxygen-dependent process. Two primary papers state it outright. Haseler, Hogan and Richardson (1999) put it in the title: "Skeletal muscle phosphocreatine recovery in exercise-trained humans is dependent on O2 availability", and manipulated inspired oxygen to show it — "the PCr recovery time constants were significantly different, being longer in hypoxia [33.5 ± 4.1 s (SE)] and shorter in hyperoxia (20.0 ± 1.8 s) than in normoxia (25.0 ± 2.7 s)", concluding that "PCr recovery, under normoxic conditions, is limited by O2 availability". Harris and colleagues (1976) showed the same thing mechanically: "Occlusion of the circulation of the quadriceps completely abolished the resynthesis of PC. Restoration of resynthesis occurred only after release of occlusion." Tomlin and Wenger's review adds the applied link — aerobic fitness aids recovery through "enhanced PCr regeneration" — while noting the distinction the text preserves: oxidative dependence is not the same as correlating with VO₂max, and "PCr resynthesis following single bouts of high intensity effort does not always correlate well with maximal oxygen consumption".
• Sleep and injury risk (OR 1.34). Taken from a 2025 preprint that had not, as far as could be established, completed peer review. Flagged in the text.
• Youth training-hour thresholds (age in years; 16 hours per week). These are consensus recommendations restated in a 2026 sport-specific review of junior tennis, not experimentally derived thresholds — and they are taken at second hand from that review rather than from the original consensus documents.
• Warm-up and injury-prevention effect sizes. Essentially all of this literature is in football and general adolescent team sport, not hockey. The transfer to hockey is an inference. The frequently quoted "30–46% injury reduction" range does trace to a source — Patel and Shah's 2025 systematic review of FIFA 11+ randomised trials (Cureus 17(12):e100463), which concludes that reductions were "typically ranging from 30% to 46%" across five controlled trials. Two reasons it is still not presented as a finding here: the population is football players (youth, collegiate and adult), not hockey players, and the review performed "a narrative synthesis" rather than a meta-analysis "due to heterogeneity in injury definitions, adherence reporting, and exposure-hour documentation", so the range is a span of individual trial results and not a pooled estimate. The attributable figure used in the text remains Liu et al. (2025), IRR 0.65.
• The Copenhagen adduction exercise meta-analysis (PMID 40827942) has been retracted (April 2026), and rated the certainty of all its own findings as very low. Its figures are retained in the text with both facts stated, rather than deleted, so that the numbers previously quoted here can be traced. Do not build a programme on them.
• Concussion incidence trend. A reported 290% increase in recorded concussions in male hockey players between 2004 and 2023 is deliberately not used in the text, because a rise in recorded cases over a period of rapidly increasing concussion awareness cannot be interpreted as a rise in actual incidence.