Evidence ledger
Every claim on the site, where it comes from, and how well it has been checked. A claim is only marked Checked once the exact passage has been read in the full source and recorded with its limits.
The four grades
- Checked
- The exact supporting passage was read in the full source and recorded with its limits.
- Abstract only
- Only the summary of the source has been read so far.
- Not yet checked
- The source is named, but no passage has been read and recorded yet.
- Model
- A simplified model or a physics rule, labelled as such. It shows how something works, not a measurement.
Your first run, with run-walk
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| S1 | Beginner plans such as the NHS Couch to 5K start with short runs between walks and build to about 30 minutes of running over nine weeks, with three runs a week. | NHS, Couch to 5K running plan (nhs.uk) | Not yet checked |
| S2 | Pooling 14 studies from six cohorts (232,149 people), any amount of running was linked with a lower risk of death from any cause during follow-up than no running, including amounts under 50 minutes a week. A link, not proof of cause. | Pedisic et al. 2020, Br J Sports Med 54:898-905 | Not yet checked |
How often and how far
4 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| S3 | The World Health Organization's 2020 guidelines advise adults to get 150 to 300 minutes of moderate activity, or 75 to 150 minutes of vigorous activity, each week. | Bull et al. 2020, Br J Sports Med 54:1451-1462 (WHO 2020 guidelines) | Not yet checked |
| S4 | In 532 novice runners training for a 4-mile race, a slower 13-week programme built on the "10% rule" had about as many injuries as a standard 8-week programme: roughly one runner in five in each group. | Buist et al. 2008, Am J Sports Med 36:33-39 | Not yet checked |
| S5 | In 873 novice runners, those whose weekly distance rose by more than 30% had more of some overuse injuries than those whose distance rose by less than 10%. A link, not proof of cause. | Nielsen et al. 2014, J Orthop Sports Phys Ther 44:739-747 | Not yet checked |
| S6 | Pooling studies, novice runners had about 17.8 injuries per 1,000 hours of running, against about 7.7 for recreational runners. | Videbæk et al. 2015, Sports Med 45:1017-1026 | Not yet checked |
How easy is easy: the talk test
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| S7 | In lab tests, the effort at which talking stops being comfortable sat close to the ventilatory threshold, a marker between moderate and hard exercise. | Persinger et al. 2004, Med Sci Sports Exerc 36:1632-1636 | Not yet checked |
| S8 | Borg's scale of perceived exertion rates effort from 6 (no exertion at all) to 20 (maximal exertion). | Borg 1982, Med Sci Sports Exerc 14:377-381 | Not yet checked |
| S9 | Well-trained endurance athletes in several sports do about 80% of their training sessions at low intensity. | Seiler 2010, Int J Sports Physiol Perform 5:276-291 | Not yet checked |
Shoes and kit that matter
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| S10 | A review of running shoe research argued that matching shoes to impact or to foot rolling has little support, and proposed choosing the shoe that feels most comfortable (a "comfort filter"). | Nigg et al. 2015, Br J Sports Med 49:1290-1294 | Not yet checked |
| S11 | In 927 novice runners given the same neutral shoe and followed for a year, runners whose feet rolled inwards (pronated) did not have more injuries than runners with neutral feet. | Nielsen et al. 2014, Br J Sports Med 48:440-447 | Not yet checked |
| S12 | Across trials in army, air force and marine recruits, assigning running shoes by the height of the foot arch did not reduce injuries. | Knapik et al. 2014, J Orthop Sports Phys Ther 44:805-812 | Not yet checked |
| ID | Claim | Source | Grade |
|---|---|---|---|
| S13 | For trained runners at moderate to fast speeds, setting the treadmill to a 1% incline best matched the energy cost of running outdoors. | Jones and Doust 1996, J Sports Sci 14:321-327 | Not yet checked |
| S14 | A review of studies comparing the two found running on a motorised treadmill broadly similar to running overground in how people move, with some differences. | Van Hooren et al. 2020, Sports Med 50:785-813 | Not yet checked |
| S15 | Runners adjust the stiffness of their legs to the surface, softer on hard ground and stiffer on soft ground, so the body bounces about the same on both. | Ferris, Louie and Farley 1998, Proc R Soc Lond B 265:989-994 | Not yet checked |
Staying motivated
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| S22 | In a study of 96 volunteers each building one new daily habit, it took a median of 66 days for the habit to feel automatic, with a range of 18 to 254 days. | Lally et al. 2010, Eur J Soc Psychol 40:998-1009 | Not yet checked |
| S27 | parkrun is a free, weekly, timed 5 km event held in parks in many countries, open to walkers, joggers and runners. | parkrun (parkrun.com) | Not yet checked |
| S23 | A meta-analysis of 94 studies found that "if-then" plans, which name when and where you will act, had a medium-to-large effect on reaching goals. | Gollwitzer and Sheeran 2006, Adv Exp Soc Psychol 38:69-119 | Not yet checked |
The running stride
9 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| R1 | Cadence, step length, contact time and bounce in the lab follow a simple model of a typical recreational runner. | Model tuned to typical published ranges | Model |
| R2 | Over a whole step the ground pushes up with an average of one body weight, so a shorter contact means a higher peak. | Newton's second law (physics) | Model |
| R3 | Heel landings usually show a sharp early impact peak; forefoot landings usually do not. | Cavanagh and Lafortune 1980, J Biomech 13:397-406; Lieberman et al. 2010, Nature 463:531-535 | Not yet checked |
| R4 | Timing of the working muscles through the stride, including activity just before landing. | Novacheck 1998, Gait & Posture 7:77-95 | Not yet checked |
| R5 | The Achilles tendon and the arch of the foot store energy on landing and return part of it. | Ker et al. 1987, Nature 325:147-149 | Not yet checked |
| R6 | 19 female forefoot strikers had lower knee loading but higher Achilles tendon force than 19 matched heel strikers, at 4 m/s. | Kulmala et al. 2013, Med Sci Sports Exerc 45:2306-2313 | Not yet checked |
| R7 | A review of 53 studies could not tell whether strike pattern changes injury risk; switching to a forefoot strike lowered economy in the short term. | Anderson et al. 2020, Sports Med 50:885-917 | Not yet checked |
| R8 | Raising step rate by 5 to 10% at the same speed reduced loading at the knee and hip. | Heiderscheit et al. 2011, Med Sci Sports Exerc 43:296-302 | Not yet checked |
| R9 | In one road race, about 89% of 936 runners filmed at the 10 km mark landed heel first. | Larson et al. 2011, J Sports Sci 29:1665-1673 | Not yet checked |
Springs and tendons
4 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H1 | The Spring lab treats the runner as a body bouncing on one leg spring: the spring-mass model of running and hopping. | Blickhan 1989, J Biomech 22:1217-1227; McMahon and Cheng 1990, J Biomech 23 Suppl 1:65-78 | Model |
| H2 | The lab works out peak force, how far the body drops while the foot is down, and the spring's stiffness from contact time, flight time, speed, leg length and body mass, with the sine-wave method of Morin and colleagues. | Morin et al. 2005, J Appl Biomech 21:167-180 | Model |
| H3 | Tendons tested in the lab return most of the energy used to stretch them when they recoil; only a small share is lost as heat. | Bennett et al. 1986, J Zool 209:537-548 | Not yet checked |
| H4 | A review proposed that more leg stiffness helps performance, and that there may be a best range for avoiding injury: some evidence links high stiffness with bone injuries and low stiffness with soft-tissue injuries. The authors called for studies that follow runners forward in time. | Butler, Crowell and Davis 2003, Clin Biomech 18:511-517 | Not yet checked |
Foot strike
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H5 | Of 283 runners filmed at the 15 km point of an elite-level half marathon, about 75% landed heel first, 24% on the midfoot and 1% on the forefoot. Midfoot landings were more common among the faster runners. | Hasegawa, Yamauchi and Kraemer 2007, J Strength Cond Res 21:888-893 | Not yet checked |
| H6 | In 52 runners from one college cross-country team, those who mostly landed heel first had about twice the rate of repetitive stress injuries of those who mostly landed on the forefoot. A small study looking back: a link, not proof. | Daoud et al. 2012, Med Sci Sports Exerc 44:1325-1334 | Not yet checked |
Cadence and step length
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H7 | In 10 recreational runners on a treadmill, the stride length each chose freely was close to the one that used the least oxygen; much longer or shorter strides cost more. | Cavanagh and Williams 1982, Med Sci Sports Exerc 14:30-35 | Not yet checked |
| H8 | A systematic review found that a higher step rate at the same speed went with less vertical bounce, lower ground forces and less energy absorbed at the knee. | Schubert, Kempf and Heiderscheit 2014, Sports Health 6:210-217 | Not yet checked |
Forces and impact
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H9 | If the force under the foot rises and falls as a smooth half-sine, its peak is about π/2 times the step time divided by the contact time, in body weights. | Physics: the impulse over one step (as in Morin et al. 2005, J Appl Biomech 21:167-180) | Model |
| H10 | Faster top running speeds come from pushing on the ground with more force, not from swinging the legs through the air more quickly. | Weyand et al. 2000, J Appl Physiol 89:1991-1999 | Not yet checked |
| H11 | A review of 18 studies found no difference in peak vertical ground forces between injured runners and controls. Loading rate was higher in runners with a history of stress fractures, mostly in studies looking back. | van der Worp et al. 2016, Br J Sports Med 50:450-457 | Not yet checked |
Arms, posture and trunk
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H12 | Running with the arms held behind the back, across the chest or on top of the head took about 3%, 9% and 13% more energy than running with a normal arm swing. Without arm swing, runners twisted their shoulders and pelvis more. | Arellano and Kram 2014, J Exp Biol 217:2456-2461 | Not yet checked |
| H13 | Running with the trunk leaning further forward lowered the stress at the joint behind the kneecap and shifted work towards the hip. | Teng and Powers 2014, J Orthop Sports Phys Ther 44:785-792 | Not yet checked |
Running economy
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H14 | In 12 well-trained men with similar VO2max, all among the first 19 finishers of a 10 km race, running economy explained about 65% of the differences in their race times. | Conley and Krahenbuhl 1980, Med Sci Sports Exerc 12:357-360 | Not yet checked |
| H15 | A meta-analysis found that strength training programmes improved running economy in middle- and long-distance runners. | Balsalobre-Fernández, Santos-Concejero and Grivas 2016, J Strength Cond Res 30:2361-2368 | Not yet checked |
| H16 | Running economy in trained runners depends on many things at once: physiology, biomechanics, training and the environment. | Saunders et al. 2004, Sports Med 34:465-485 | Not yet checked |
Uphill and downhill
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H17 | The hill widget uses a curve fitted to the measured energy cost of running on treadmill slopes from −45% to +45%. On it, cost rises steeply uphill and is lowest on downhills of roughly 10 to 20%. The coefficients are as widely quoted and have not yet been checked against the paper. | Minetti et al. 2002, J Appl Physiol 93:1039-1046 | Model |
| H24 | Exercise in which muscles work while lengthening, such as running downhill, causes the most delayed muscle soreness. | Cheung, Hume and Maxwell 2003, Sports Med 33:145-164 | Not yet checked |
| H25 | After one bout of unaccustomed exercise, repeating it causes less soreness and muscle damage: the "repeated bout effect". | McHugh 2003, Scand J Med Sci Sports 13:88-97 | Not yet checked |
Shoe science: foam, drop, plates
6 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| H18 | In 18 high-calibre runners, a prototype racing shoe with a curved carbon plate and a new foam lowered the energy cost of running by about 4% on average against two established racing shoes, and lowered it in all 18. | Hoogkamer et al. 2018, Sports Med 48:1009-1019 | Not yet checked |
| H19 | In 553 leisure-time runners given shoes with a 10, 6 or 0 mm drop and followed for 6 months, overall injury risk did not differ by drop. In an exploratory split, low-drop shoes went with fewer injuries in occasional runners and more in regular runners. | Malisoux et al. 2016, Am J Sports Med 44:2933-2940 | Not yet checked |
| H20 | In 848 recreational runners given a softer or a harder version of the same shoe, the harder shoe went with a higher injury risk. Split by body mass, the difference showed only in the lighter runners. | Malisoux et al. 2020, Am J Sports Med 48:473-480 | Not yet checked |
| H21 | In 264 recreational runners followed for 22 weeks, those who used more than one pair of running shoes in turn had a lower injury risk (hazard ratio 0.61). A link, not proof. | Malisoux et al. 2015, Scand J Med Sci Sports 25:110-115 | Not yet checked |
| H22 | For road races, World Athletics rules limit the sole of a shoe to 40 mm thick and allow no more than one rigid plate inside it. | World Athletics Technical Rules, Rule 5 (athletic shoes) | Not yet checked |
| H23 | The Shoe cut-away is a generic shoe built for the lab, not any brand or model. Its foam squash is an illustration, not a measurement. | Illustration | Model |
Energy systems
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E1 | Muscles remake ATP in three ways: from phosphocreatine, by breaking down glucose without oxygen (glycolysis), and with oxygen in the mitochondria. All three work at once in almost every effort, rather than one after another. | Gastin 2001, Sports Med 31:725-741 | Not yet checked |
| E2 | In an all-out effort, the aerobic and anaerobic systems supply equal shares of the energy at somewhere between 1 and 2 minutes, most probably at about 75 seconds. | Gastin 2001, Sports Med 31:725-741 | Not yet checked |
Heart, lungs and oxygen
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E3 | Oxygen used each minute equals the blood the heart pumps each minute (cardiac output) times the oxygen each litre of blood gives up on its way round: the Fick principle. The lab uses typical values for a fit adult of 70 kg, not measurements. | Fick principle (physiology), with typical textbook values | Model |
| E4 | When trained people exercise with large muscle groups, their VO2max is limited mainly by how much oxygen the heart and blood can deliver, rather than by how much the muscles can use. | Bassett and Howley 2000, Med Sci Sports Exerc 32:70-84 | Not yet checked |
| E5 | Endurance athletes reach a much higher maximal cardiac output than untrained people, mainly because each beat pumps more blood; maximal heart rate hardly changes with training. | Levine 2008, J Physiol 586:25-34 | Not yet checked |
VO2max
4 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E6 | The idea of a maximal oxygen uptake, a ceiling that oxygen use reaches and cannot pass however much harder you work, comes from A. V. Hill and colleagues in the 1920s. | Hill and Lupton 1923, Q J Med 16:135-171 | Not yet checked |
| E7 | In about 480 sedentary adults from 99 families who did the same 20-week training programme, VO2max rose by about 400 mL a minute on average, but gains varied widely between people and ran in families. Up to 47% of the variation in gains was estimated to be inherited. | Bouchard et al. 1999, J Appl Physiol 87:1003-1008 (HERITAGE Family Study) | Not yet checked |
| E8 | In 810 healthy adults aged 21 to 87 followed for a median of about 8 years, peak oxygen uptake fell faster with each decade: about 3 to 6% per 10 years in the 20s and 30s, and more than 20% per 10 years from the 70s. More active people were fitter at every age. | Fleg et al. 2005, Circulation 112:674-682 (Baltimore Longitudinal Study of Aging) | Not yet checked |
| E9 | Pooling 28 studies of healthy adults aged 18 to 45, both high-intensity interval training and continuous endurance training raised VO2max a lot compared with no training; intervals gave a possibly small extra gain over continuous training. | Milanović, Sporiš and Weston 2015, Sports Med 45:1469-1481 | Not yet checked |
Lactate and thresholds
4 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E10 | Lactate is not a waste product. Muscles, the heart and the brain take it up and use it as fuel, and the liver can turn it back into glucose. | Brooks 2018, Cell Metab 27:757-785 | Not yet checked |
| E11 | There are many different definitions of a lactate threshold. A review found them useful for setting training and predicting endurance performance, though they do not all measure the same thing. | Faude, Kindermann and Meyer 2009, Sports Med 39:469-490 | Not yet checked |
| E12 | In the critical speed model, the distance covered in an all-out effort rises in a straight line with its duration: distance = critical speed × time + D′, a fixed reserve of distance you can run above critical speed. | Monod and Scherrer 1965, Ergonomics 8:329-338; Jones et al. 2010, Med Sci Sports Exerc 42:1876-1890 | Model |
| E13 | Critical speed (or critical power) marks the boundary between efforts where oxygen use and blood lactate settle to a steady level and efforts where they keep rising until you have to stop. | Jones et al. 2019, Physiol Rep 7:e14098 | Not yet checked |
Heart-rate zones
6 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E14 | The common "220 minus age" formula for maximal heart rate did not come from a study designed to produce it, and was never an accurate guide for individuals. | Robergs and Landwehr 2002, J Exerc Physiol Online 5(2):1-10 | Not yet checked |
| E15 | A meta-analysis of 351 studies (18,712 people), checked in a lab study of 514 people, gave maximal heart rate ≈ 208 − 0.7 × age, with a standard error of about 10 beats a minute. "220 minus age" underestimated it in older adults. | Tanaka, Monahan and Seals 2001, J Am Coll Cardiol 37:153-156 | Not yet checked |
| E16 | Training intensity can be set as a share of heart rate reserve: resting heart rate plus a percentage of the gap between maximal and resting heart rate. | Karvonen, Kentala and Mustala 1957, Ann Med Exp Biol Fenn 35:307-315 | Not yet checked |
| E17 | The American College of Sports Medicine classes 30 to 39% of heart rate reserve as light intensity, 40 to 59% as moderate, 60 to 89% as vigorous and 90% or more as near maximal. | Garber et al. 2011, Med Sci Sports Exerc 43:1334-1359 (ACSM position stand) | Not yet checked |
| E18 | During long, steady exercise, heart rate slowly drifts up while the blood pumped per beat falls: cardiovascular drift. Heat and dehydration can make it larger. | Coyle and González-Alonso 2001, Exerc Sport Sci Rev 29:88-92 | Not yet checked |
| E19 | In 50 adults exercising on a treadmill, bike and elliptical, a chest-strap monitor agreed most closely with an ECG. Wrist-worn optical monitors were less accurate, most of all on the elliptical. | Gillinov et al. 2017, Med Sci Sports Exerc 49:1697-1703 | Not yet checked |
Muscle fibre types
5 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E20 | Motor neurons are recruited in order of size, smallest first: the size principle. So small, slow, fatigue-resistant motor units switch on first, and bigger, faster ones join as the force needed rises. | Henneman, Somjen and Carpenter 1965, J Neurophysiol 28:560-580 | Not yet checked |
| E21 | Calf muscle samples from 14 elite distance runners averaged 79% slow-twitch fibres. | Costill, Fink and Pollock 1976, Med Sci Sports 8:96-100 | Not yet checked |
| E22 | Among college track athletes, sprinters had a larger share of fast-twitch fibres in the calf than distance runners. | Costill et al. 1976, J Appl Physiol 40:149-154 | Not yet checked |
| E23 | In rats trained to run, the muscles' capacity to use oxygen (their mitochondrial enzymes) roughly doubled. | Holloszy 1967, J Biol Chem 242:2278-2282 | Not yet checked |
| E24 | The Muscle fibres lab is an illustration. Its fibre mixes and the efforts at which each kind switches on are chosen to show the idea, not measured. | Illustration | Model |
Carbs and fat as fuel
6 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E25 | As exercise gets harder, the body shifts from burning mostly fat towards burning mostly carbohydrate: the "crossover" concept. | Brooks and Mercier 1994, J Appl Physiol 76:2253-2261 | Not yet checked |
| E26 | In five trained people tested at 25, 65 and 85% of VO2max, fat supplied the largest share at moderate intensity (65%); at 85%, muscle glycogen became the main fuel. | Romijn et al. 1993, Am J Physiol 265:E380-E391 | Not yet checked |
| E27 | A diet high in carbohydrate raised muscle glycogen and let people exercise for longer at a fixed workload than a diet low in carbohydrate. | Bergström et al. 1967, Acta Physiol Scand 71:140-150 | Not yet checked |
| E28 | Running on the flat costs about 1 kcal for each kilogram of body mass for each kilometre, and this changes little with speed. | Margaria et al. 1963, J Appl Physiol 18:367-370 | Not yet checked |
| E29 | The fuel tank widget is arithmetic with typical values: fat gives about 9 kcal per gram and carbohydrate about 4 kcal per gram. | Arithmetic (standard food energy values) | Model |
| E43 | Low energy availability, when the energy eaten does not cover the energy used in exercise and daily life, can impair many body systems, including bone health. The IOC calls this relative energy deficiency in sport (RED-S). | Mountjoy et al. 2018, Br J Sports Med 52:687-697 (IOC consensus statement) | Not yet checked |
Why humans can run far
4 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| E39 | Humans have many features that help endurance running but matter little for walking, such as springy leg tendons, a ligament that steadies the head and a large gluteus maximus. The authors argue that endurance running helped shape human evolution. | Bramble and Lieberman 2004, Nature 432:345-352 | Not yet checked |
| E40 | The gluteus maximus works far harder in running than in walking, where it helps keep the trunk from pitching forward. | Lieberman et al. 2006, J Exp Biol 209:2143-2155 | Not yet checked |
| E41 | Persistence hunting, chasing an animal in the heat until it is too hot to go on, has been recorded among hunter-gatherers in the Kalahari. | Liebenberg 2006, Curr Anthropol 47:1017-1026 | Not yet checked |
| E42 | Humans can keep running in the heat because they lose heat by sweating, while many four-legged mammals cool mainly by panting, which is hard to combine with galloping. | Carrier 1984, Curr Anthropol 25:483-495 | Not yet checked |
| ID | Claim | Source | Grade |
|---|---|---|---|
| T1 | In 12 sub-elite runners trained for about 5 months, a programme with more low-intensity running (below the first ventilatory threshold) improved a simulated 10.4 km cross-country race by more, about 157 s against 121 s, than one with more moderately hard running. Both had a similar, small share of high-intensity work. | Esteve-Lanao et al. 2007, J Strength Cond Res 21:943-949 | Not yet checked |
| T2 | In 48 well-trained runners, cyclists, triathletes and skiers over 9 weeks, a "polarised" programme (mostly low intensity, some high intensity, little in between) raised peak oxygen uptake by about 12%, more than threshold, high-intensity or high-volume programmes. Groups were small. | Stöggl and Sperlich 2014, Front Physiol 5:33 | Not yet checked |
| T3 | The intensity mix calculator is arithmetic: each kind of running as a share of your total running time. | Arithmetic | Model |
Workouts explained
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| T9 | The workout examples and the talk-test efforts on this page are common coaching practice, not tested prescriptions. | Coaching practice | Model |
Plans from 5K to marathon
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| T4 | Pooling studies of tapering, the largest performance gains came from a taper of about 2 weeks in which training volume was cut by roughly 41 to 60% while intensity and frequency were kept. | Bosquet et al. 2007, Med Sci Sports Exerc 39:1358-1365 | Not yet checked |
| T8 | The example weeks on this page are illustrations of a common pattern (mostly easy running, one or two harder sessions and a long run), not tested plans. | Illustration | Model |
Cross-training
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| T5 | In highly trained athletes, less than 4 weeks of stopped or much reduced training brings a rapid fall in VO2max and blood volume. | Mujika and Padilla 2000, Sports Med 30:79-87 | Not yet checked |
| T6 | A review found that trained endurance athletes kept their aerobic performance with deep-water running for up to 6 weeks. In the water, peak heart rate and oxygen uptake are lower than on land. | Reilly, Dowzer and Cable 2003, J Sports Sci 21:959-972 | Not yet checked |
What your watch metrics mean
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| T7 | Pooling 14 validation studies, wearables that estimate VO2max from resting data overestimated it by about 2 ml/kg/min on average, with wide errors for individuals. Estimates made from exercise data were unbiased on average, but an individual reading could still be off by roughly 10 ml/kg/min either way. | Molina-Garcia et al. 2022, Sports Med 52:1577-1597 | Not yet checked |
Warm-up studio
5 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| W1 | In more than 1,500 army recruits, adding short static stretches to the warm-up made no meaningful difference to injury risk. | Pope et al. 2000, Med Sci Sports Exerc | Abstract only |
| W2 | In more than 2,700 runners, a 3 to 5 minute stretch before running neither prevented nor caused injury; runners who switched routines had more injuries (a link, not proof). | Pereles et al. 2011, AAOS annual meeting (conference report, not a journal paper) | Abstract only |
| W3 | Holds of 60 s or more per muscle just before testing cut performance by about 4.6%, shorter holds by about 1.1%; dynamic stretching gave about +1.3%. | Behm et al. 2016, Appl Physiol Nutr Metab 41:1-11 | Abstract only |
| W4 | A review of 5 trials leaned towards fewer injuries with a warm-up, but the evidence was not conclusive. | Fradkin et al. 2006, J Sci Med Sport | Abstract only |
| W5 | Pooling 25 trials (26,610 people), stretching programmes showed no clear protective effect; strength training programmes were linked with the biggest drop in injuries. | Lauersen et al. 2014, Br J Sports Med 48:871-877 | Abstract only |
Strength training for runners
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| M1 | A review of 24 studies in middle- and long-distance runners found that adding strength training generally improved running economy, by about 2 to 8% against running alone, and tended to improve time-trial performance over 1.5 to 10 km. VO2max and body composition were usually unchanged. | Blagrove, Howatson and Hayes 2018, Sports Med 48:1117-1149 | Not yet checked |
| M2 | The American College of Sports Medicine advises adults to do resistance exercise for each major muscle group on 2 to 3 days a week. | Garber et al. 2011, Med Sci Sports Exerc 43:1334-1359 (ACSM position stand) | Not yet checked |
| M5 | The Strength library shows each exercise on a generic 3D runner, with the main muscles it works lit. The movements and the doses (sets and repetitions) are common starting points from coaching practice, not tested prescriptions. | Illustration | Model |
Running drills
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| M7 | Pooling 16 studies of endurance athletes, adding explosive training or heavy weight training to endurance training improved running economy by similar small amounts, about 4.8% and 3.7%. Longer programmes went with bigger gains. | Denadai et al. 2017, Sports Med 47:545-554 | Not yet checked |
Stretching after runs
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| M3 | Pooling 11 trials (452 people), strength training and stretching improved range of motion by similar amounts. The trials differed a lot from one another. | Afonso et al. 2021, Healthcare 9:427 | Not yet checked |
| M6 | A Cochrane review of randomised trials found that stretching before, after, or before and after exercise did not produce clinically important reductions in delayed muscle soreness in healthy adults. Stretching after exercise lowered soreness a day later by about 1 point on a 100-point scale. | Herbert, de Noronha and Kamper 2011, Cochrane Database Syst Rev (7):CD004577 | Not yet checked |
| M8 | The ACSM advises flexibility exercises for the major muscle groups on at least 2 to 3 days a week, holding static stretches for 10 to 30 s to the point of tightness or slight discomfort, for about 60 s in total per exercise. Daily stretching gives bigger gains in range of motion. | Garber et al. 2011, Med Sci Sports Exerc 43:1334-1359 (ACSM position stand) | Not yet checked |
Foam rolling
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| M4 | Pooling 21 studies, foam rolling before exercise slightly improved sprint performance (about 0.7%) and flexibility (about 4%). After exercise it slightly reduced muscle pain (about 6%) and the drop in sprint performance. The authors judged the effects small. | Wiewelhove et al. 2019, Front Physiol 10:376 | Not yet checked |
Everyday eating
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| F1 | A joint position statement suggests daily carbohydrate of about 3 to 5 g per kg of body mass for light training, 5 to 7 g/kg for moderate exercise of about an hour a day, 6 to 10 g/kg for 1 to 3 hours a day, and 8 to 12 g/kg for very heavy training. | Thomas, Erdman and Burke 2016, Med Sci Sports Exerc 48:543-568 (ACSM, Academy of Nutrition and Dietetics, Dietitians of Canada) | Not yet checked |
| F2 | The same statement suggests daily protein of about 1.2 to 2.0 g per kg of body mass for athletes, spread over the day and after key sessions. | Thomas, Erdman and Burke 2016, Med Sci Sports Exerc 48:543-568 | Not yet checked |
| F16 | The IOC describes relative energy deficiency in sport as impaired body function caused by eating too little for the energy used. It includes, but is not limited to, effects on metabolic rate, menstrual function, bone health, immunity, protein synthesis and heart health, and it affects men as well as women. | Mountjoy et al. 2018, Br J Sports Med 52:687-697 (IOC consensus statement) | Not yet checked |
Before and during runs
6 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| F3 | Before exercise lasting more than an hour, the statement suggests 1 to 4 g of carbohydrate per kg of body mass, eaten 1 to 4 hours before. | Thomas, Erdman and Burke 2016, Med Sci Sports Exerc 48:543-568 | Not yet checked |
| F4 | Suggested carbohydrate during exercise rises with its length: little or none is needed under an hour, up to about 30 g an hour for 1 to 2 hours, up to about 60 g an hour for 2 to 3 hours, and up to about 90 g an hour beyond about 2.5 hours. Above about 60 g an hour, a mix of glucose and fructose is needed. | Jeukendrup 2014, Sports Med 44 Suppl 1:S25-S33 | Not yet checked |
| F5 | The gut adapts to what it gets: a diet high in carbohydrate raises its capacity to absorb carbohydrate. Practising eating and drinking during training may reduce stomach problems, though few studies have tested this directly. | Jeukendrup 2017, Sports Med 47 Suppl 1:101-110 | Not yet checked |
| F17 | For speedy refuelling when there are less than 8 hours between two demanding sessions, the statement suggests about 1 to 1.2 g of carbohydrate per kg of body mass per hour for the first 4 hours, then a return to daily needs. | Thomas, Erdman and Burke 2016, Med Sci Sports Exerc 48:543-568 | Not yet checked |
| F20 | Eating, or drinking large amounts (especially of sugary drinks), shortly before exercise seems to make a stitch more likely. | Morton and Callister 2015, Sports Med 45:23-35 | Not yet checked |
| F6 | The carbohydrate planner is arithmetic on the ranges in F4: a target per hour for your expected time, the total, and how many of your gels or drinks that is. | Arithmetic | Model |
Sleep
4 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| F9 | Sleep experts recommend that adults sleep 7 hours or more a night on a regular basis. | Watson et al. 2015, Sleep 38:843-844 (AASM and Sleep Research Society consensus) | Not yet checked |
| F10 | In 112 school athletes, those who slept less than 8 hours a night on average had about 1.7 times the odds of injury. A small study looking back: a link, not proof. | Milewski et al. 2014, J Pediatr Orthop 34:129-133 | Not yet checked |
| F11 | In 11 college basketball players who extended their sleep for several weeks, sprint times and shooting accuracy improved. | Mah et al. 2011, Sleep 34:943-950 | Not yet checked |
| F19 | Single doses of 100 mg of caffeine may affect sleep in some adults if taken close to bedtime. | EFSA Panel on Dietetic Products, Nutrition and Allergies 2015, EFSA Journal 13(5):4102 | Not yet checked |
Rest days
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| F12 | Training improves fitness when hard work is matched with enough recovery. A short spell of overload with recovery can be followed by improvement; when recovery falls short for long, performance can stay down for weeks or months (non-functional overreaching), or longer in the overtraining syndrome. | Meeusen et al. 2013, Med Sci Sports Exerc 45:186-205 (ECSS and ACSM consensus) | Not yet checked |
Choosing a race
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| C2 | In results from several big-city marathons over many years, finishing times slowed progressively as heat stress (the wet-bulb globe temperature) rose from about 5 to 25 °C, and slower runners were slowed more than the fastest. | Ely et al. 2007, Med Sci Sports Exerc 39:487-493 | Not yet checked |
Pacing
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| C1 | In about 92,000 performances in 14 US marathons, men slowed more than women in the second half, and were about three times as likely to slow markedly (a second half more than 30% slower than the first), even allowing for age, experience and finishing time. | Deaner et al. 2015, Med Sci Sports Exerc 47:607-616 | Not yet checked |
| C5 | The Course fly-over is an illustration on made-up courses, with the hills drawn taller than they are so you can see them. Its pacing plan keeps the energy cost per minute steady using the hill curve in H17, limits how much faster it lets you run downhill, and ignores wind, heat, turns and tiredness. | Illustration and arithmetic | Model |
Race-day fuel
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| C3 | For events longer than about 90 minutes, the statement suggests carbohydrate loading: about 10 to 12 g of carbohydrate per kg of body mass a day for the 36 to 48 hours before. | Thomas, Erdman and Burke 2016, Med Sci Sports Exerc 48:543-568 | Not yet checked |
| C4 | The consensus statement advises using thirst to guide how much to drink. Drinking more than is lost, over hours, is the main cause of exercise-associated hyponatraemia (low blood sodium). | Hew-Butler et al. 2015, Clin J Sport Med 25:303-320 | Not yet checked |
Race-time estimator
3 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| P3 | The Riegel formula predicts a time at a new distance from a known one: new time = known time × (new distance ÷ known distance) to the power 1.06. It came from a power law fitted to world-record performances lasting about 3.5 to 230 minutes. | Riegel 1981, Am Sci 69:285-290 | Not yet checked |
| P4 | In a survey of 2,303 recreational runners, the Riegel formula predicted times well up to the half marathon, but predicted marathon times that were too fast: by at least 10 minutes for half of the runners. | Vickers and Vertosick 2016, BMC Sports Sci Med Rehabil 8:1 | Not yet checked |
| P5 | The race-time estimator is arithmetic: the Riegel formula (P3) applied to the race you enter, with the pace for each distance. | Arithmetic | Model |
Heart-rate zones
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| P8 | The heart-rate tool is arithmetic: your measured maximum, or 208 − 0.7 × age (E15) if you have none, and your resting heart rate give your heart rate reserve (E16), which is split into the ACSM ranges (E17). | Arithmetic | Model |
Cadence counter
2 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| P1 | The often-quoted 180 steps per minute comes from coach Jack Daniels' count of elite runners' cadence at the 1984 Olympics, not from a study of what suits each runner. | Daniels, Daniels' Running Formula (book) | Not yet checked |
| P2 | Step length is speed divided by steps per second: at 6:00 per km (2.78 m/s) and 165 steps per minute, each step is about 1.01 m. | Arithmetic | Model |
Plan builder
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| P6 | The plan builder is a rule of thumb, not a tested plan. It starts from your current week, raises weekly distance by at most about 10% a week, makes every fourth week easier (about three quarters of the biggest week so far), grows the long run by about a kilometre or 10% a week up to a cap by race distance, and keeps it to about half the week or less. With a two-week taper, the last two weeks are about 30% and 50% below the biggest week; with a one-week taper, race week is about 40% below. Typical lengths and long runs come from T8, the taper from T4. | Arithmetic on common coaching patterns | Model |
Fuel planner
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| P7 | The fuel planner is arithmetic on the carbohydrate ranges in F4: a target per hour for your expected finishing time, how many of your gels or drinks that is, and when to take each one. | Arithmetic | Model |
Myths and evidence
1 claims
| ID | Claim | Source | Grade |
|---|---|---|---|
| L1 | Pooling 17 studies (114,829 people), hip and knee osteoarthritis was found in about 3.5% of recreational runners, against 10.2% of people who did not run and 13.3% of competitive (elite or international) runners. Links, not proof of cause. | Alentorn-Geli et al. 2017, J Orthop Sports Phys Ther 47:373-390 | Not yet checked |
General information, not medical advice.