Does the order of my training blocks matter?

Blocked vs spread hard sessions: no reliable performance edge at equal load. Pooled over six small studies (107 trained to well-trained athletes, mean 4.9 weeks, range 1-12), block periodization gave small advantages in VO2max (SMD 0.40, 95% CI 0.02 to 0.79) and Wmax (0.28, 0.01 to 0.54); both 95% prediction intervals cross zero (−0.32 to 1.12; −0.18 to 0.73), four of the six studies were first-authored by the review's senior author, and the one significant performance result set a 7-day overload against normal training (REF-946, B+).

The trials behind the edge come from that laboratory: five HIT sessions in week 1 then one a week for three weeks raised VO2max 4.6% against no change on two a week, 8 sessions each (19 trained cyclists; only VO2max differed significantly between groups in the abstract; REF-1214, B); three such cycles over 12 weeks gave VO2max +8.8% vs +3.7% (REF-136, B, n = 15); a 5-1-3-1-1 pattern of the same 11 sessions raised elite skiers' peak power +4% vs −3% when spread (REF-1215, B, n = 19).

Against them: 12 load-matched weeks improved 5-min (+9%) and 40-min power (+8%) in both arms with no evidence of a difference (30 trained cyclists; REF-098, B+); in an independent crossover, nine 5 × 4-min sessions in one week vs three a week for three weeks showed no evidence of a difference in VO2max, economy or a 600-m time trial, point estimates leaning toward the even spread (20 junior skiers, 10 women; REF-1216, B+); seven studies in trained road cyclists gave no basis to choose (REF-103, B+).

Two of the six pooled studies are independent (one is REF-1223): mostly, not all, one laboratory network.

Concentrating easy volume: no evidence of a benefit. One week at 160% of usual volume then lighter weeks, against even weeks of identical total volume, showed no evidence of a different change in VO2max (d = −0.1), maximal speed (d = −0.4) or speed at 3 mmol/L (d = 0.2), with lower mood and perceived recovery in the big week and the week after (20 junior skiers, randomized crossover, abstract; REF-1217, B+).

Pyramidal and polarized: a polarized phase is a bounded VO2peak tool; the order was not shown to matter. In 60 well-trained male runners' 16-week base period (about 6 h a week, about 80% in zone 1, two quality sessions), all four plans improved, including all-pyramidal (5-km −0.6%); pyramidal-then-polarized ended marginally best (5-km −1.5%, d = 0.28) and tied with the reverse on relative VO2peak (+3.0% each), every 5-km change below the paper's 3.2% smallest detectable change, and no race followed (REF-509, A-).

In each arm with polarized weeks the gains came in its first polarized 8 weeks and then held — polarized-then-pyramidal kept them through its pyramidal weeks (VO2peak 0.0%, 5-km 0.0% in weeks 8-16) — and a second polarized 8 weeks added about nothing (all-polarized: VO2peak +2.1% by week 8, 0.0% after) (REF-509, Table 5; the reading is the base's, the authors read a peaking effect).

Over 11 controlled trials polarized training raised VO2peak slightly more than other distributions (SMD 0.24, 0.01 to 0.48), only in interventions under 12 weeks (0.40) and in highly trained athletes (0.46) — largely the same trials — with no evidence of a difference in time trials (−0.01, −0.28 to 0.25) (REF-1213, A-).

Individual data from 13 studies showed no evidence of a difference between polarized and pyramidal for VO2max (SMD −0.06) or time trial (−0.05); recreational athletes may gain more VO2max on pyramidal (subgroup; REF-991, A-). "Pyramidal in preparation, polarized in competition" describes elite runners (REF-811, B+), untested.

Order of interval formats: no evidence of a difference. In 63 cyclists (VO2peak 61), the same 24 maximal-effort sessions over 12 weeks ordered 4 × 16 → 4 × 8 → 4 × 4 min (session lactate 4.7 → 12.7 mmol/L), the reverse, or mixed in every mesocycle all improved 40-min power, peak power and VO2peak by 5-10%, with no evidence of a difference between orders and similar responder rates (REF-096, A-) — one trial, not "order does not matter".

Reverse periodization: no demonstrated advantage, a mixed picture. In 11 small studies of 8-15 weeks (swimmers, runners, triathletes, lifters; six in mostly recreational athletes; no cyclists; eight sharing one author), endurance tests showed no evidence of a difference against traditional or block periodization, two youth-swimming studies favoured reverse on 100 m and both strength studies favoured traditional on 1RM; volumes were similar in 7 of the 11 (REF-948, B+).

In the one independent endurance trial with volume and intensity equated, recreational runners improved 5000 m on both orders against a control, with no evidence of a difference between them (REF-948, full text).

A fixed load stops paying within weeks; raising it restarts the response. In nine probably untrained people training 40 min a day, 6 days a week, VO2max rose for 3 weeks at constant work rates, then held flat, then rose for 3 more weeks once the work rates were raised (half-times 10.3 and 10.8 days; +23% in 9 weeks) (REF-1218, B, uncontrolled — a beginner's timetable).

In 8 highly trained runners, a month of one vVO2max session a week raised vVO2max (20.5 → 21.1 km/h) and the next month at three a week changed nothing (REF-005, B, sequential). Across training studies endurance and HIT training kept improving between 2-6 and 6-10 weeks while sprint intervals did not, and lower starting fitness meant larger gains (REF-465, A, full text).

Qualities not being trained: protect intensity; frequency and session length can fall, within limits. In the three Hickson studies (subjects trained by the study for 10 weeks): cutting work rates by one-third or two-thirds at unchanged frequency and duration lost VO2max and 21% and 30% of long-term endurance (REF-951, B, n = 12); cutting to 4 or 2 days a week at unchanged intensity and length kept VO2max for 15 weeks, VO2max being the only outcome (REF-1219, B, n = 12); cutting sessions from 40 to 26 or 13 min at unchanged intensity and frequency kept VO2max, ~5-min endurance and heart size, while 2-h endurance held at 26 min and fell 10% at 13 min (REF-1220, B, n = 13, abstract).

The review that carries these figures covers general populations and finds athlete data insufficient (REF-587, B+); its frequency figure is, at source, VO2max only.

In well-trained cyclists, one HIT session every 7-10 days through an 8-week transition left power at 4 mmol/L 10.6% (±8.2%) and 40-min power 12.4% (±7.6%) above easy riding only (REF-166, B, n = 13, not randomised), and one weekly sprint set in a 3-week transition went with 20-min power +7.3% vs −1.3% six weeks later (REF-165, B, n = 11 elite men).

One heavy strength session a week kept cyclists' strength gains through 13 racing weeks (REF-800, B) and stopping for 8 weeks lost squat jump, Wmax and Wingate power against controls (REF-744, B); in untrained young women one session every 7 days held strength, quadriceps size and aerobic power for 12 weeks, one every 14 days lost about half the size and aerobic-power gains (REF-1136, B).

After a one-week moderate-intensity block, 10-s sprint power was −1.5% against +1.5% after a HIT block (not significant; fixed order, no control; REF-421, B). Stopping costs more than reducing: −10.1% vs −4.8% VO2max over 5 weeks in top kayakers (REF-950, B); VO2max about 4-7% lower after 12-21 days of complete rest (REF-1133, B+).

Short concentrated blocks can show within two weeks in well-trained cyclists. Six moderate-intensity interval sessions in 7 days plus 6 easy days: power at 4 mmol/L +4.0% vs −1.3% with time-matched regular training (REF-137, B+, randomized crossover, n = 30); five 30/15 sessions in 6 days plus 6 easy days: power at VO2max +3.7% vs +0.7% (REF-048, B, non-randomised, n = 33); six daily 30/15 sessions then a 5-day step taper: Wmax +5.0% vs +0.9% after an 11-day taper (interaction P = .005; VO2peak interaction P = .131) (REF-1198, B, 17 elite men in season, not randomised); seven days of ~2-h sprint-interval sessions then a light week: 20-km time-trial power +4.6-6.8% 14 days after the block while the control's fell 3.3% (REF-1223, B, 28 competitive men in season, matched groups; HIT matched only between its two sprint arms, the control asked to keep total volume similar).

These qualify the premise "no time to absorb a new stimulus" (SYNTH-76) for well-trained cyclists; none started straight after an A-race, and the sessions were race-like or redistributed hard work the riders already did (inference).

Elite practice: more than one order has won. The most decorated winter Olympian won World Cup titles in a season with HIT heaviest early, in seven 7-11-day blocks of 8-13 HIT sessions, and nine years later with HIT rising toward racing, at a TRIMP load chosen to match (15 vs 18 h a week) (REF-1222, C+, n = 1).

Ten world-class kayakers gained VO2peak +8.1% from a block cycle and +11% from a traditional cycle 10 weeks and 120 h longer (two seasons, no control; REF-1221, B, abstract). Twelve Norwegian world-class coaches describe traditional periodization, volume falling and specificity rising toward competition (REF-138, B).

In 46 elite sprint swimmers (105 swimmer-seasons) the fastest pattern had load rising into the second macrocycle and a peak about 5 weeks before the season's best, and a flat pattern whose load fell was slowest (6.0% vs 7.4% off the world top-10 mean; one of five contrasts significant; REF-1224, B, observational).

Block-periodization theory is theory. The case that mixed training gives conflicting responses and too little stimulation, and the residual-training-effect durations coaches quote, come from the model's proponent (REF-1225, C+, narrative review, abstract); its entry read no durations, and this framework cites none.

Do not sell a sequence. At equal load no order has a reliable performance edge (Key Findings 1-5); the named models share untested planning assumptions, a critical review argues (REF-945, C+). Pick the order the calendar, the weather and the athlete can execute.

Order: general → the block's lever → race-specific → taper. Not from a study; the base's own practice: it puts the most race-like work last, as world-class coaches (REF-138, B) and elite runners (REF-811, B+) are described doing — descriptions, not tests. Intensity first is acceptable when winter or the calendar forces short sessions (inference from REF-948, B+: endurance tests showed no evidence of a difference; no cyclists; 1RM favoured traditional).

One primary emphasis per block, progressed inside it, kept while its marker moves and changed at the review once it has stopped. Not from a study; the base's own practice: about 3-6 weeks per block (SYNTH-42, SYNTH-33), one emphasis up to ~8 weeks, the work raised every week or two; direction from REF-1218, REF-509 Table 5 and REF-005. One lever, the thesis and the review date are SYNTH-42.

Every quality not being trained stays on a maintenance dose with its intensity intact. Not from a study; the base's own practice: one session of it every 7-10 days at full intensity and shorter than when building, one heavy strength session a week, a few short sprints inside an easy ride (direction: Key Finding 7). A long-event athlete keeps some long sessions (REF-1220, B).

A loading week is an option, not an upgrade — for an athlete who already tolerates two hard sessions a week and wants variety or has a free week (practice); keep one hard session a week after it (inference across REF-1214, REF-1215, REF-136 against REF-1216); never pile up easy volume expecting more adaptation (REF-1217, B+).

The last block before the taper is what the event demands (Not from a study; the base's own practice); the taper is SYNTH-34.