Preparing for a 5-10 km run
A large easy layer, two quality days (one at vVO2max speed, one at threshold), race pace growing in the last eight weeks, heavy strength for economy and a short taper; the race-day pacing plan is practice, not evidence
The question it answers
A runner has a 5 km or 10 km goal race: what decides it, what the weeks contain, which sessions carry the build, how the last eight weeks and race day change. Not repeated here: distribution by weekly hours SYNTH-26, the shape of a VO2max rep SYNTH-33, threshold format SYNTH-27, gym work SYNTH-44, jumps SYNTH-45, taper detail SYNTH-34, a block's lever SYNTH-42.
What decides the result
Default (what to prescribe first)
For a healthy Tier 1-3 runner 8-16 weeks from the race, already running at least three days a week:
Most running easy, and genuinely easy. Anchor the top of easy at the first threshold, which sat at 82.3% of critical speed across 26 studies and 527 runners (80.6% in the slowest, 84.2% in the fastest CS group) (REF-178, A).
Run below it, not pinned to it: in 20 trained amateurs, easy running at the first threshold gave the same 6-week gains as 10-15% below it, at a higher psychophysiological load (REF-705, B+).
With hard work held equal (8.3% vs 8.5%), among 12 sub-elite runners in two groups, the group at 80.5% easy improved a 10.4 km race by 157 s vs 121.5 s for the group at 66.8% easy over five months (REF-094, B).
In 8 well-trained sub-elite runners followed for 6 months, time below the first ventilatory threshold (4581 ± 979 min, vs 1354 and 487 min in the higher zones — pyramidal) tracked the 10.1 km cross-country time (r = −0.97; 4.2 km race r = −0.79, P = 0.06) (REF-817, B, correlational).
Two quality sessions a week: one at about vVO2max speed (95% and 100% were the speeds tested, [REF-813]), one at threshold / critical speed. Elite distance runners did at least one threshold session and one zone-3 session each week (REF-811, B+, systematic review of 10 descriptive studies).
Not from a study; the base's own practice: the count for amateurs follows SYNTH-26's hour bands. In 17 well-trained runners, 4 weeks of 2 interval sessions a week at 95% or 100% vVO2max both improved 5000 m time and speed at 3.5 mmol/L; only 100% also improved vVO2max, economy and 1500 m time; VO2max moved in neither (REF-813, B, no non-interval control).
In 8 highly trained runners, one vVO2max session a week raised vVO2max 20.5 → 21.1 km/h; three a week added nothing (REF-005, B, uncontrolled). Rep shape: SYNTH-33 (mostly cycling and skiing trials); threshold: SYNTH-27.
Above critical speed, work and recovery intensity and duration together decide how much W′ returns, and W′ recovery varies widely between athletes (REF-809, B+, conceptual) — set recoveries deliberately.
Not threshold-dominant: in 48 well-trained endurance athletes over 9 weeks it added nothing while polarized training improved most (REF-303, A-, mixed sports); a review of 16 running studies (6 interventions) favoured pyramidal and polarized over threshold training, though some top marathoners train threshold-heavy (REF-773, B+).
Sequence the distribution: pyramidal first, more polarized in the last ~8 weeks. In 60 well-trained male runners over 16 weeks, all four distributions improved 5 km time; pyramidal-then-polarized improved it most, by about 1.5% (REF-509, A-; men only, modest gap between arms). Elite runners are described doing the same, pyramidal in preparation and more polarized in the competitive period (REF-811, B+; description, not a tested effect).
Race pace grows as the race nears. In world-class distance runners the share of race-pace running rises toward the main competition (REF-089, B+, narrative, Tier 5); international runners do fewer threshold and more race-pace sessions before competition (REF-097, C+).
Not from a study; the base's own practice: from ~8 weeks out, one quality day becomes race-pace reps (5 km: 400 m-1 km; 10 km: 1-2 km reps or a continuous 3-5 km at goal pace), progressed by total distance at pace.
One long easy run a week. In 26 well-trained men matched on 10 km time (~39 min), those doing regular ≥90-min runs lost less economy over a 90-min run at lactate threshold (+3.1% vs +6.0%), and economy durability tracked the length of the weekly longest run (r = −0.67) (REF-160, B, cross-sectional). Inference: whether that matters over a 20-60 min race is untested.
Not from a study; the base's own practice: 60-90 min easy, not beyond about a third of the week's running time.
Strength: heavy lower-body, twice a week while building (SYNTH-44): in middle- and long-distance runners, high-load training improved time-trial performance moderately and combined methods largely, plyometrics alone not significantly (REF-865, A-); heavy loads helped economy more at higher speeds, plyometrics only at ≤12 km/h (REF-801, A-); explosive work replacing 32% of training for 9 weeks improved 5 km time in 10 well-trained endurance athletes (REF-802, B). Jumps: a small dose on top of lifting (SYNTH-45; REF-518, A, in track-and-field athletes: running economy −1.96% at p = 0.05).
Taper per SYNTH-34 for content, shorter than its two-week default: volume down 41-60%, intensity and session frequency kept (REF-107, A, 27 studies, 2-week optimum; REF-3492, A, 14 studies, 174 endurance athletes per the full text: time trials improved, VO2max and economy did not).
Elite British 3000 m steeplechase-10,000 m runners reported a median 6-day taper, continuous volume to 71% and interval volume to 67% of normal, peak interval speed 112% of race speed (REF-818, B, survey, no outcome); world-class runners start 7-10 days out (REF-089, B+).
Not from a study; the base's own practice: 7-10 days for a 5 km, 10-14 days for a 10 km or after a heavy block.
Decision
| Situation | Do this | Source and how sure |
|---|---|---|
| New to running or <3 runs a week (Tier 0-1) | The lever is regularity and gradual volume; one quality session at most. Keep week-to-week increases under 20%; there is no validated "10% rule" | REF-623 B+ (261 recreational runners: 20-60% jumps → more injuries at 21 days, gone by 8 weeks); REF-565 B+ (10% vs 24% same risk); REF-619 B (874 novices, >30% over 2 weeks: non-significant trend). Session cap = house practice; gate = SYNTH-42 |
| Recreational, 3-5 runs a week | Default above, prescribed as sessions, not percentages | REF-086 B+ (30 recreational 10 km runners, 10 weeks: polarized 5.0% vs between-thresholds 3.6%, not significant; clearer only in the 12 who kept to plan); session count = SYNTH-26, house |
| Trained, 16-week build | Pyramidal first half, more polarized second half | REF-509 A- (well-trained men; transfer to women and novices is inference) |
| 5 km vs 10 km goal | 5 km: vVO2max-speed and faster reps weigh more; 10 km: threshold/CS and longer race-pace reps weigh more | Not from a study; the base's own practice — no entry compares the two builds. Nearest: reps at 95% vVO2max matched 100% for 5000 m, only 100% improved 1500 m (REF-813, B, 17 runners, 4 weeks) |
| Short of time for a block (tested in Tier 2-3) | A 6-9-week block of less volume with short near-maximal work, then back to the base | REF-869 B: 25% less training plus 6-12 × 30-s sprints 3-4×/wk for 6-9 weeks: 3 km 10.4 → 10.1 min, 10 km 37.3 → 36.3 min, VO2max unchanged (12 vs 5 controls); REF-870 B: 10 moderately trained runners (3 women; 8 controls) on 10-20-30 only for 7 weeks, 54% less distance: 5 km −48 s, 1500 m −21 s. Short blocks, not a year-round plan |
| Last ~8 weeks | One quality day turns race-specific; short above-VO2max speed-endurance reps are an option | REF-089 B+, REF-097 C+ (elite shape); REF-697 B (narrative review, trained athletes: speed-endurance improved 10-60 min events without changing VO2max); REF-695 B (with an aerobic base it helped 10 km running); format and dose → SYNTH-46 |
| Masters | Keep the intensity and the load; widen gaps between hard days on what recovery shows, not on age | SYNTH-35 B; REF-595 B (former elite men over 22 years: ~6% VO2max loss per decade with continued hard training vs ~15% after stopping; it graded the amount of training, not intensity) |
| Women | Same default; do not restructure around cycle phase; missed periods plus bone pain → energy availability, not only load | SYNTH-37 B+; REF-343 B (59 elite distance athletes: amenorrhoea and low testosterone went with ~4.5× bone-injury prevalence); REF-509 and REF-160 tested men only |
Race day: warm-up and pacing
How to progress / periodise
Not from a study; the base's own practice, using SYNTH-42's horizon bands: >16 weeks out, build volume and the engine; 16-8 weeks, pyramidal with both quality days; the last 8 weeks, more polarized with race pace growing (REF-509 for the direction); the last 1-2 weeks, taper. Progress running volume by less than 20% a week (REF-623).
Check the build every 4-6 weeks with a 3-5 km time trial, the same race-pace session repeated, or a critical-speed test: 3600, 2400 and 1200 m on a track in one visit, 30 min apart, matched a multi-day treadmill CS (r = .89; repeat variation 0.4%) in 10 male distance runners, but its D′ was neither valid nor repeatable (REF-876, B). Reading the response: SYNTH-42.
When NOT to use it / red flags
What the evidence does and does not show
Evidence gaps
How this entry was checked and changed
(the skeleton rests on an A- running RCT with a 5 km outcome REF-509, an A meta-analysis anchoring the easy zone REF-178, an A- strength review REF-072 and A taper meta-analyses; the session doses rest on B entries, 5-10 km pacing on descriptive elite entries only, and warm-up on no 5-10 km running entry)