Substituting one kind of training for another

Name the reason; it decides what is allowed. Injury → the red flags and the rule for the tissue in SYNTH-94 and SYNTH-73 come first, and a clinician leads bone-like pain.

Illness → SYNTH-36: fever or any symptom below the neck means stop, not a session moved to another machine; symptoms above the neck with no fever mean no training on the first day, then easy and short once they are mild and easing, on any machine. Travel, weather, equipment or preference → straight to step 3.

Offer the substitute without waiting to be asked. In a year-long cohort of 49 recreational runners, weeks with an injury had 14.1 fewer minutes a day of moderate-to-vigorous activity (95% CI −22.5 to −6.0), replaced mainly by sitting, while pain alone did not cut activity (REF-1434, B, abstract). It shows the need, not that offering changes what runners do.

Choose by tolerance first, closeness second, availability third. (a) Nothing that provokes the symptoms: the swaps by tissue are SYNTH-94's (REF-1380, A, a guideline whose swap list is expert opinion; REF-1383, C+).

(b) The closer to the sport's muscles and movement, the more carries over (PHYS-40, established — specificity of training adaptation): swimming and arm work transfer least to the legs (REF-1358, C+), and extra swimming gave runners half the gain of extra running (REF-952, B+).

For a runner: deep-water running or cycling (they did alike in REF-1428, B), an elliptical or stepper (REF-1435, B, acute), an anti-gravity treadmill where there is one (REF-1436, B+), then rowing or swimming. For a cyclist kept off the road: the indoor trainer.

Kept off the bike: no trial exists; an elliptical, deep-water running, a rower or the pool by tolerance, and running only for one who already runs (PHYS-22, established; PHYS-37, established). (c) What is there: a hotel gym usually has a treadmill, an elliptical, a rower and a spin bike. The order inside (b) and (c) is practice.

Keep the session's minutes and its purpose; carry the effort, not the pace, watts or heart rate. Minute-for-minute at matched intensity is what held running fitness in the trials (REF-1428, REF-922, B; REF-953, B, intensity matched, minutes not stated, both arms improved).

The same RPE or talk-test level is the bridge between machines: heart rate was "generally similar" on six machines at RPE 11, 13 and 15 while energy cost differed (REF-1430, B); at the maximal lactate steady state RPE was 7.2 in both cycling and running and time to exhaustion 37.7 vs 34.4 min (REF-1432, B+); the talk test located the ventilatory threshold on a treadmill and on a cycle ergometer (REF-929, B).

A longer ride for an easy run is acceptable where time allows — athletes and coaches in one survey estimated about 1.8 times the minutes (REF-1440, C, their opinion; accepting it is the base's practice) — and never a law. Keep the hard sessions hard on the substitute: maintenance runs on intensity (PHYS-41, established — maintenance by kept intensity; REF-587, B+).

Read heart rate per mode; never import zones. At the same effort heart rate runs lower on the bike and in deep water, by amounts that differ between studies; in the pool only the maximum was measured, about 11 bpm under running's; on an elliptical or stepper no group difference was detected at the same share of each device's own peak, with no values reported (conversion table).

Between bike and run the individual error at threshold was 12.1 bpm (REF-1431, B), so heart-rate zones wanted on the substitute are measured there (SYNTH-38); otherwise RPE leads and heart rate is read.

In water, structure, not feel. Intervals, a quick cadence and an RPE anchored to a known land session, with heart-rate targets about 10 bpm under land ones (practice, from REF-1433, B: 8-11 bpm lower at the same oxygen uptake): recreational runners asked to match their land effort ran at 27% lower VO2 with a vest, while competitive runners matched it (REF-1438, B).

Log every substitute session with session RPE and keep a load line per sport. Bike, run and swim scores are not one currency: load-score models built in 15 cyclists missed 59 new athletes' carbohydrate cost by 15-45% and energy cost by 9-31%, worst in runners and kayakers, until recalibrated per sport (REF-1439, B; energy and carbohydrate, not adaptation), and running's mechanical cost is outside every score (inference from REF-235, B+, and PHYS-45, established — internal and external load are two parts).

A summed fitness line is not running fitness after weeks of riding (practice, supported by that measurement).

Expect the engine to hold for a few weeks, and the legs' tolerance of impact not to. In healthy trained runners and young men, VO2max held or dipped as much as with running, and 2-mile to 5-km performance held, over 4-6 weeks of deep-water running or cycling at matched dose (REF-1428, REF-922, B; REF-923, B, no control; REF-1437, B+); where cycling replaced half the hard runs both arms improved alike (REF-953, B, 5 weeks, 6 and 5 runners), and across small trials of 4-10 weeks, mostly in untrained or moderately fit people, neither mode differed significantly from the other (REF-1368, A-); in the injured, in runners beyond 6 weeks and in trained cyclists nothing is known.

Bone builds from impact and heavy loading, not from cycling or swimming (PHYS-37, established); tendon stiffens under high-strain loading over weeks to months (PHYS-38, established), and lost stiffness within 1-2 weeks of limb unloading on crutches or in a cast (REF-1355, B+; a break from running with normal walking was not studied, so how fast it goes after a substitution block is unknown).

Come back as a return, not a resumption. After a substitution block of about two weeks or more, running restarts below its old volume — roughly half to 70% of it is the base's practice (SYNTH-36), and roughly half of the running that hurt after an injury (SYNTH-94) — easy first, intensity last, gated by the injured tissue's rule (SYNTH-94); no single run more than about 10% longer than the longest of the past 30 days, one cohort's risk marker and not a safe line (REF-1102, B+), counted from the running actually done rather than the old peak (inference).

As a race nears, the race-specific sessions go back to the race sport (PHYS-40; SYNTH-76).

At matched dose the usual substitutes did as well as running for a few weeks, in healthy runners. Water running, cycling and running each lost a little VO2max and none changed 2-mile time over 6 weeks in 32 trained young men (REF-1428, B, abstract); deep-water running kept VO2max, ventilatory threshold and economy over 6 weeks in 16 trained men (REF-922, B); a review of 11 deep-water trials found it comparable to land training, effect sizes 0.06-0.10 in two runner trials, nothing pooled, 7 of 11 trials of poor quality (REF-1437, B+); across 7 small trials running-only and cycling-only training did not differ significantly — "not interchangeability" (REF-1368, A-).

Effort carries between machines; pace, watts and calories do not. The same RPE gave generally similar heart rates on six machines with very different energy costs (REF-1430, B); at MLSS cycling and running felt the same (7.2 vs 7.2) and lasted about as long (REF-1432, B+); the talk test located the ventilatory threshold on a treadmill and a cycle ergometer (REF-929, B).

Heart rate at the same effort differs by mode and by person. Lower on the bike: 3 bpm at threshold on average in triathletes with a 12.1-bpm individual error (REF-1431, B), 10 bpm at MLSS in moderately trained men (REF-1432, B+), HRmax often 6-10 bpm lower in triathletes (REF-537, B+, full text); lower in deep water by 8-11 bpm at the same VO2 (REF-1433, B); the swim peak 11 bpm under the treadmill's in fitness swimmers (REF-1441, B+); no detected difference on an elliptical or stepper at the same share of each device's peak (REF-1435, B, no values).

Between different people runners' HRmax sat 2 bpm above cyclists' (REF-3494, B+).

Feel misleads in an unfamiliar mode. Submaximal water running feels harder (REF-1433, B; the abstract does not say the RPE comparison was at equal oxygen uptake), and recreational women matching their land effort ran at 27% lower VO2 with a vest; competitive women did not undershoot (REF-1438, B).

No time factor has been tested. Practitioners' estimates — 1.76 times the minutes of riding for an easy run, 0.54 times of running for an easy ride — changed with session length even in their own answers (REF-1440, C); the trials that held fitness matched minutes and intensity.

Load scores are not one currency across sports. Within cyclists six load metrics tracked a session's carbohydrate and energy cost closely; across sports the cycling-built models missed by up to 45% until recalibrated per sport (REF-1439, B); adaptation and fatigue were not tested.

Injured runners do not substitute on their own. Injured weeks cost 14.1 minutes a day of moderate-to-vigorous activity, mostly replaced by sitting (REF-1434, B).

A treadmill matches the road for oxygen cost, not for feel or race times. At submaximal speeds 0% and 1% both came within about 0.6 mL/kg/min of overground running on group means; heart rate and RPE shifted with speed; endurance performance was poorer on a motorised belt (REF-1429, A-); in one small study 1% was the specific match from about 13.5 km/h (REF-733, B).

The elliptical and the anti-gravity treadmill give a running-like heart and lung stimulus with less impact — acutely. Elliptical and stepper showed no detected difference from the treadmill in VO2max, or in heart rate at the same share of each device's peak (REF-1435, B; no values); unweighting cut vertical force and submaximal heart rate at a given speed and shifted load toward the forefoot (REF-1436, B+). Neither has a training trial in the base.

The substitutes do not train impact tolerance. Impact and heavy loading build bone; cycling and swimming do not (PHYS-37, established; low lumbar-spine density in road cyclists, REF-896, A-); tendon adapts over weeks to months (PHYS-38, established) and lost stiffness within 1-2 weeks of limb unloading (REF-1355, B+; not a running break with normal walking); running damages muscle far more than cycling (PHYS-22, established; REF-235, B+).