Conditioning — evidence & provenance

Where the numbers come from. The conditioning tools — fitness testing, the speed profile, the prescription, and the verify step — aren't invented in-house: every formula, anchor, and zone boundary is taken from published sport-science or established strength-and-conditioning practice. This page is the single place that traces each idea back to its source, so you can check the working.

Three tiers of provenance, tagged throughout:

[paper] — peer-reviewed research the tool implements directly.
[practice] — an established S&C coaching method or coach resource (widely used, not always a single citable trial).
[ours] — a pragmatic estimate or product decision we made, informed by the above but not lifted from a paper. These are the ones to scrutinise.

1. The anchors — what we measure

Maximal Aerobic Speed (MAS)

The slowest running speed that elicits VO₂max; the aerobic anchor for interval prescription. It comes from a fixed-distance field test (the 2 km time-trial, straight-line; or the 1.2 km shuttle), as MAS = distance ÷ time.

[practice] MAS-based conditioning was popularised for field sport by Dan Baker and is now standard (e.g. World Rugby's coaching syllabus). The field-test distance ÷ time estimate and the %MAS interval grid below come from this tradition.
[paper] Academic roots: Léger & Boucher (1980), An indirect continuous running multistage field test: the Université de Montréal Track Test (Can J Appl Sport Sci 5(2):77–84) — an incremental field test for maximal aerobic velocity; and the vVO₂max interval-training tradition (Billat).

Velocity at the 30-15 IFT (VIFT)

The final running speed of the 30-15 Intermittent Fitness Test — a maximal, intermittent (30 s run / 15 s walk), shuttle-format beep test. VIFT integrates aerobic power, change-of-direction cost, and recovery ability, which is why Buchheit recommends prescribing HIIT directly from it rather than converting it to MAS. For that reason a VIFT result never moves the MAS anchor.

[paper] Buchheit M. (2008). The 30-15 intermittent fitness test: accuracy for individualizing interval training of young intermittent sport players. J Strength Cond Res 22(2):365–374. (Test developed ~2000.)
[paper] Buchheit M. (2021). The 30-15 IFT — two decades of learnings (SPSR 148), and the earlier 10-year review.
[practice] Starting velocity — we open at 12 km/h, not the textbook 8. Buchheit sanctions starting well-trained squads at 10 or even 12 km/h to save time. Because VIFT is the absolute speed of the last stage reached (not a stage count), skipping the trivially easy early stages does not change an athlete's result — it only removes ~6 min of dead wind-up per group. Set in IFT_START_KMH (src/lib/testing/ift.ts); the increment (0.5 km/h/stage) and everything downstream is unchanged.
[paper] Smith T.B. et al. (2025). Estimating Maximal Aerobic Speed in Academy Soccer Players: Time-Trial Methods vs the 30-15 Intermittent Fitness Test. Eur J Sport Sci. — Compares TT and 30-15 estimates of MAS in a comparable population. Supports using the 2 km TT as the straight-line MAS protocol and treating VIFT as a separate, non-interchangeable measure.
[paper] Bellenger C.R. et al. (2015). Predicting Maximal Aerobic Speed Through Set-Distance Time-Trials. Eur J Appl Physiol 115(12):2593–2598. — Validates the distance ÷ time formula across multiple fixed-distance TTs. Confirms the 2 km TT produces a MAS estimate consistent with laboratory measures.
[paper] Thron M. et al. (2022). Overestimation of Maximal Aerobic Speed by the Université de Montréal Track Test and a 1500-m Time Trial in Soccer. Front Physiol 13:1023257. — Shorter TTs overestimate MAS due to higher anaerobic contribution; supports our protocol decision to use 2 km over 1500 m or the UMTT, and explains why shuttle MAS (1.2 km) reads lower than straight-line MAS.

Maximal Sprinting Speed (MSS)

The sprint ceiling. The structural/mechanical anchor. It has three sources, in precedence order manual override > tested > observed:

observed — from GPS, a rolling 12-week top-speed percentile. A floor: the athlete is at least this fast, but only if they happened to sprint flat-out in the window.
tested — a controlled maximal-velocity measurement that takes precedence over the observed floor (it's a deliberate max, not an incidental one). Two protocols, both deriving MSS: a Flying 10 m timed sprint (MSS = 10 ÷ time, the classic 20 m fly-in fly-out) and a GPS max sprint (the unit's peak read off a controlled effort). A tested result survives the ingest auto-refresh exactly as a manual override does.
[paper] Used as the speed anchor in the anaerobic-speed-reserve literature below (Sandford; Weyand). The flying-sprint protocol for max velocity is standard S&C practice; [practice] e.g. the 10–30 m fly-in is the common field method for MSS = distance ÷ time.

Yo-Yo Intermittent Recovery Test (IR1 / IR2) — a benchmark, NOT an anchor

A beep-paced 2×20 m shuttle test with 10 s active recovery, run to exhaustion; the score is the total distance covered. It yields an estimated VO₂max, not a runnable velocity, so — unlike MAS/VIFT/MSS — it never feeds the prescription engine. It is a fitness screen (trend + squad benchmarking + a familiar VO₂max number), kept on the "describe" side alongside observed MSS and jumps. This is the same non-interchangeability discipline drawn for VIFT, one step further: VIFT at least anchors a velocity; an aerobic-capacity score does not, so no Yo-Yo→MAS/velocity conversion is fabricated.

[paper] Bangsbo J., Iaia F.M., Krustrup P. (2008). The Yo-Yo intermittent recovery test: a useful tool for evaluation of physical performance in intermittent sports. Sports Med 38(1):37–51 — the test, its two levels, and the VO₂max regressions (IR1: VO₂max = distance·0.0084 + 36.4; IR2: distance·0.0136 + 45.3).
[paper] Krustrup P. et al. (2003). The Yo-Yo intermittent recovery test: physiological response, reliability, and validity. Med Sci Sports Exerc 35(4):697–705 — the original validation (Yo-Yo IR1).
⚠ implementation note the on-device beep schedule encodes the canonical speed/bout progression, but the generated audio must be validated against an official Yo-Yo recording before competitive use (a wrong table invalidates the result); the IR2 early bout counts in particular should be checked.

2. The speed profile — turning anchors into zones

Anaerobic Speed Reserve (ASR) & locomotor profiles

ASR = MSS − MAS. The reserve an athlete has above their aerobic ceiling; tolerance to supramaximal work tracks ASR better than aerobic power alone. Athletes classify as Speed / Hybrid / Endurance by where MSS sits relative to MAS (the speed ratio MSS ÷ MAS, with boundaries at 1.7 and 1.9).

[paper] Sandford G.N., Laursen P.B., Buchheit M. (2021). Anaerobic Speed/Power Reserve and Sport Performance: Scientific Basis, Current Applications and Future Directions. Sports Med 51(10):2017–2028 — the source for ASR-based individualisation and the speed/hybrid/endurance typology.
[paper] Sandford G.N. et al. (2019). "Question Your Categories": the Misunderstood Complexity of Middle-Distance Running Profiles… Front Sports Act Living 1:28 — the ASR profile spectrum.
[paper] Bundle, Weyand et al. (2003); Weyand et al. (2000). Energetics of high-speed running — the physiological basis of the speed reserve.

Why profiles should change training selection

The Speed/Hybrid/Endurance classification isn't just descriptive — athletes with different fibre typologies respond differently to the same training stimulus, and this is the physiological rationale for profile-specific session recommendations.

[paper] Bellinger P. et al. (2020). Muscle Fiber Typology Is Associated with the Incidence of Overreaching in Response to Overload Training. J Appl Physiol 129(4):823–836. — Fast-twitch dominant athletes (Speed profile) show higher overreaching incidence and are more sensitive to volume accumulation. Supports showing short, intense RST as the priority format for Speed athletes rather than long aerobic intervals.
[paper] Lievens E. et al. (2020). Muscle Fiber Typology Substantially Influences Time to Recover from High-Intensity Exercise. J Appl Physiol 128(5):1370–1381. — Speed-profile athletes need more between- session recovery, while Endurance athletes tolerate higher training densities. The evidence base for why the same template is labelled "primary" for one profile and "caution" for another.
[paper] Thron M. et al. (2025). Anaerobic Speed Reserve and Acute Responses to a Short-Format High-Intensity Interval Session in Runners. J Sci Med Sport 28(3):XX. — ASR predicts the acute physiological response (HR, RPE, blood lactate) to a session, not just classification. Strengthens the case for individualising work intensity by profile rather than prescribing group averages.

VIFT-preferred high zones; estimates when an anchor is missing

When a VIFT result exists it splits the boundary between the high zones (the intermittent ceiling); without it, the high zones fall back to the ASR scheme. When an anchor is missing we estimate MAS ≈ VIFT ÷ 1.25 and MSS ≈ VIFT × 1.4.

[practice] The VIFT-as-ceiling idea and the rough VIFT-to-MAS relationship (≈ VIFT ÷ 1.25) come from Buchheit's 30-15 guidance.
[ours] Estimating a sprint ceiling as ≈ VIFT × 1.4 when no GPS top speed exists is a pragmatic placeholder — flagged in-product with a * ("estimated"). It is not a universal constant, and it is replaced the moment real GPS sprint data lands.

3. The prescription — turning zones into sessions

HIIT format families (the "weapons")

Long intervals at MAS, short 15-15 / 30-30 intervals at the intermittent ceiling, speed-endurance and repeated-sprint formats — and the principle that work:rest, intensity, and exercise mode are the levers you manipulate per goal.

[paper] Buchheit M., Laursen P.B. (2013). High-Intensity Interval Training, Solutions to the Programming Puzzle. Sports Med — Part I: Cardiopulmonary Emphasis 43(5):313–338; Part II: Anaerobic Energy, Neuromuscular Load & Practical Applications 43(10):927–954.
[practice] Laursen & Buchheit, Science and Application of High-Intensity Interval Training (Human Kinetics, 2019) and hiitscience.com — the "weapons" framing for the curated session catalogue.

Why the dual-pathway (VIFT vs speed-reserve) design

The Session Builder and Drill Library offer two prescription pathways: VIFT-based (when a 30-15 IFT result is present) and speed-reserve/ASR-based (when only MAS + MSS is available). This isn't cosmetic — these approaches predict meaningfully different training speeds for the same athlete.

[paper] Collison J. et al. (2022). Supramaximal Interval Running Prescription in Australian Rules Football: %MAS vs %ASR vs 30-15 IFT. J Strength Cond Res 36(11):3121–3128. — Direct field comparison of the three prescription methods in a team-sport cohort. %ASR and %VIFT produce individualised supramaximal speeds that %MAS alone cannot; the paper is the primary evidence for why underdog prioritises VIFT when available and falls back to ASR rather than %MAS for high-intensity work.

Repeated-sprint training (RST builder)

The RST builder's work:rest defaults (5 s/25 s = 1:5; 8 s/52 s ≈ 1:6.5), set structure (3 sets × 6 reps), and the time-based individualisation logic (distance scales with each athlete's Z6 speed) are grounded in the repeated-sprint ability literature.

[paper] Girard O., Mendez-Villanueva A., Bishop D. (2011). Repeated- Sprint Ability — Part I: Factors Contributing to Fatigue. Sports Med 41(8):673–694. — Identifies the key determinants of RSA performance: PCr re-synthesis, oxidative capacity, and neuromuscular factors. Establishes that short sprint durations (3–10 s) with full phosphocreatine recovery (rest ≥4× work) maximise quality.
[paper] Bishop D., Girard O., Mendez-Villanueva A. (2011). Repeated- Sprint Ability — Part II: Recommendations for Training. Sports Med 41(9):741–756. — Prescriptive companion to Part I. Supports the 3–10 s effort window, 1:5+ work:rest ratios, and multi-set structure used in the RST builder.

Tempo running (Tempo builder)

Extensive and intensive tempo runs — the GAA / team-sport high-speed-running staple. A fixed rep distance (chosen in 20 m increments, up to 300 m per rep and longer efforts of 600–800 m total work) is run at a target pace expressed as a percentage of maximal sprinting speed (%MSS), so each athlete's own top speed individualises the pace while the distance stays fixed for the group (the squad falls into pace lanes). Extensive = more volume at a lower intensity (~~60–72 % MSS); intensive = less volume at a higher intensity (~~72–85 % MSS). Straight-line or a there-and-back over cones. Work:rest defaults to 1:2, with 1:3 and 1:4 options.

[practice] Tempo running is long-established sprint/team-sport conditioning practice (Charlie Francis' extensive/intensive tempo; widely adopted in GAA S&C) — submaximal high-speed running dosed as a fraction of top speed to build high-speed-running capacity and repeatability without the neural cost of maximal sprinting.
[ours] The extensive/intensive %MSS bands (60–72 % / 72–85 %), the tempoZone label mapping, and the reference-pace used only to seed a default rest from the work:rest ratio (REF_MSS_MPS = 8.0, never used in a prescription) are our conventions for readability, not hard physiological boundaries. The real per-athlete run time varies with each player's MSS; the coach tunes the absolute rest.

Absolute-pace prescription (Absolute pace builder)

The no-GPS, no-testing pathway. Every other velocity pathway in this engine resolves off a testing-spine anchor (%MSS, %ASR, %MAS, %VIFT) — and so does what the app calls its HSR threshold, which is hsr_pct_of_mss, a percentage of each athlete's own top speed. A club with no GPS unit and no test day therefore cannot express an intensity at all.

Coaches solve this with arithmetic that needs no anchors: fix the distance, give a time window, and the speed is specified. d / t = v. A window rather than a single time because a whole squad cannot hit one number:

Prescription	Implied band	The 5.5 m/s line
100 m in 18–20 s	5.00–5.56 m/s	18.2 s
200 m in 37–42 s	4.76–5.41 m/s	36.4 s

The prescription is deliberately uniform — same distance, same clock, every athlete. That is what makes it programmable and sendable to players who have never been tested, which is the off-season / early-pre-season club case.

[practice] The 5.5 m/s high-speed-running threshold is the STATSports default and is in wide use across GAA S&C; 7.0 m/s is the common companion sprint threshold. These are squad-independent conventions, deliberately not derived from anyone's anchors — that is the entire point of this mode. A club that has tested should be using the individualised %MSS thresholds instead (Settings → Speed thresholds).
[practice] Prescribing a fixed distance against a target time is long-standing club practice (e.g. "100 m in 18–20", "200 m in 37–42") and long predates GPS.
[ours] The intermediate moderate zone boundary (4.0 m/s), the decision to target the slow end of the window rather than the midpoint, and suggestWindow's default 2-second slack are our conventions. Targeting the slow end makes the prescription a floor every athlete must clear, which matches how whole-session Verify already treats a prescription (a minimum dose, only under is flagged). The coach's window is described against the thresholds but never silently adjusted to fit a zone.

Lane banding (all builder modes)

Every prescription is individual, but a coach lays out a handful of cone lanes. src/lib/prescription/lane-bands.ts turns N individual target speeds into the lane count the coach asks for (2–6), and resolves what distance each lane runs.

Two rules do the work:

Grouping is by clustering, not by an equal-width pace grid. The coach names the number of cone groups; the squad is split into exactly that many groups by exact 1-D k-means (dynamic programming — Ckmeans.1d.dp), which minimises the pace spread within each lane. A pinned band width is available as an override for coaches who want fixed km/h steps between lanes.
On a fixed-distance rep, lanes run one shared time to staggered cones. The rep time is set by the coach's chosen distance at the squad's low-anchor pace; every lane then runs that same time, so a quicker lane's cone sits a few metres further out and the whole squad still goes on one whistle.

[practice] Running a fixed-time interval with the group split into ability lanes at different distances is standard team-sport conditioning practice — it is what makes a squad session coachable from one clock, and it is already how the MAS-interval and COD-shuttle families in this engine work (distance individualised over a fixed work window). Staggering tempo cones applies the same convention to the one family that previously did not: "same 60 m, each at your own pace" is not policeable on a pitch. The Drill Library shares the same clustering and lane budget (bandDrillLanes), but not the staggered-cone mode: a drill's per-athlete distance is its own cell-validated shuttle geometry, and the lane reports the group's representative distance exactly as the source sheet's "Group" column does.
[ours] The lane budget range (2–6, default 4), the choice of exact 1-D k-means over equal-width bands, and LOW_ANCHOR_PCT = 0.1 (which athlete the coach's chosen rep distance describes) are our conventions. The percentile is the one to scrutinise: anchoring on the outright slowest athlete is the more literal reading, but it is not robust — on the live Roscommon roster two broken MSS anchors (6.3 and 6.49 m/s against a 9.26 median) would have turned a 60 m prescription into 86–95 m reps for the rest of the panel. The 10th percentile keeps the coach's number meaning "near enough the slowest player" while refusing to let one duff anchor set the session; anyone below it runs a shorter rep.

%MAS interval grid

Per-rep distance = MAS × %MAS × work-seconds, across 10-10 / 15-15 / 20-20 / 30-30 work:rest ratios and 50–140 % MAS. A faithful, generative version of the classic coach's "MAS Calculations" worksheet, validated number-for-number against it.

[practice] Classic MAS-interval prescription (Baker; World Rugby). The canonical short-interval recipe — ~120 % MAS for 15–30 s with equal passive rest, accumulating 15–20 min of work — is the well-known default.
[ours] The way each %MAS is labelled with a training zone (e.g. 100 % = aerobic base, 110 % = aerobic power, 130 % = speed endurance) is our convention for readability, not a hard physiological boundary.

Change-of-direction (COD) shuttle correction

For run-to-the-line shuttle HIIT, a per-turn time penalty shortens the distance an athlete covers versus a straight run at the same target speed (the cost of decelerating into the turn and accelerating out of it).

[practice] A faithful version of Martin Buchheit's "Squad Shuttle Run Prescription" worksheet, validated number-for-number against its worked rows.
[paper] The acceleration/deceleration energetics behind the penalty: di Prampero et al. (2005) (below).

4. GPS metrics that feed the anchors

Metabolic power / High Metabolic Load Distance (HMLD)

The di Prampero "equivalent slope" model for the energy cost of accelerated running, used to credit hard accelerations that a plain speed threshold misses.

[paper] di Prampero P.E. et al. (2005). Sprint running: a new energetic approach. J Exp Biol 208(14):2809–2816 — accelerated flat running ≈ uphill running at constant speed; energy cost ~3.6–4.0 J/kg/m.
[paper] Osgnach C. et al. (2010). Energy cost and metabolic power in elite soccer: a new match analysis approach. Med Sci Sports Exerc 42(1):170–178 — the team-sport application.

Individualised high-speed running & sprint thresholds

High-speed running and sprinting are defined relative to each athlete's own MSS, not absolute cut-offs. (See the Metrics reference for how these are used day-to-day.)

[paper] Mendez-Villanueva A., Buchheit M., Simpson B., Peltola E., Bourdon P. (2011). Does on-field sprinting performance in young soccer players depend on how fast they can run or how fast they do run? J Strength Cond Res 25(9):2634–2638 — the case for individualising speed thresholds to MSS.
[paper] Spencer M., Lawrence S., Rechichi C., Bishop D., Dawson B., Goodman C. (2004). Time–motion analysis of elite field hockey, with special reference to repeated-sprint activity. J Sports Sci 22(9):843–850 — defines a repeated-sprint sequence as ≥3 sprints with mean recovery <21 s.

Reference list

Every peer-reviewed entry below was checked against its PubMed record or DOI in June 2026 (title, authors, journal, year, volume, pages). Links go to PubMed or the DOI resolver — stable, free to open — plus the authors' own copies where they exist.

#	Reference	Used for	Open
1	Buchheit M. (2008). The 30-15 Intermittent Fitness Test: accuracy for individualizing interval training of young intermittent sport players. J Strength Cond Res 22(2):365–374.	VIFT capture	PubMed 18550949 · doi
2	Buchheit M. (2021). The 30-15 IFT — two decades of learnings. SPSR 148.	VIFT guidance	PDF · 30-15ift.com
3	Buchheit M., Laursen P.B. (2013). HIIT, Solutions to the Programming Puzzle. Part I: Cardiopulmonary Emphasis. Sports Med 43(5):313–338.	Session catalogue	doi
4	Buchheit M., Laursen P.B. (2013). … Part II: Anaerobic Energy, Neuromuscular Load & Practical Applications. Sports Med 43(10):927–954.	Session catalogue	PubMed 23832851
5	Laursen P., Buchheit M. (2019). Science and Application of HIIT. Human Kinetics (ISBN 9781492552123).	"Weapons" framing	hiitscience.com
6	Sandford G.N., Laursen P.B., Buchheit M. (2021). Anaerobic Speed/Power Reserve and Sport Performance… Sports Med 51(10):2017–2028.	ASR, profiles	doi · PDF
7	Sandford G.N. et al. (2019). "Question Your Categories"… Middle-Distance Running Profiles. Front Sports Act Living 1:28.	Profile spectrum	doi
8	di Prampero P.E. et al. (2005). Sprint running: a new energetic approach. J Exp Biol 208(14):2809–2816.	Metabolic power, COD penalty	PubMed 16000549 · doi
9	Osgnach C. et al. (2010). Energy cost and metabolic power in elite soccer… Med Sci Sports Exerc 42(1):170–178.	HMLD (team-sport)	PubMed 20010116 · doi
10	Léger L., Boucher R. (1980). An indirect continuous running multistage field test: the UMTT. Can J Appl Sport Sci 5(2):77–84.	MAS test heritage	PubMed 7389053
11	Mendez-Villanueva A., Buchheit M. et al. (2011). Does on-field sprinting performance depend on how fast they can run or how fast they do run? J Strength Cond Res 25(9):2634–2638.	Individualised HSR	PubMed 21768891
12	Spencer M. et al. (2004). Time–motion analysis of elite field hockey… repeated-sprint activity. J Sports Sci 22(9):843–850.	Repeated-sprint definition	PubMed 15513278
13	Buchheit M. "Squad Shuttle Run Prescription" / "MAS Calculations" coach worksheets.	COD shuttle + %MAS prescription	martin-buchheit.net
14	Dan Baker / World Rugby — MAS conditioning method.	MAS test + %MAS intervals	World Rugby
15	Collison J. et al. (2022). Supramaximal Interval Running Prescription in Australian Rules Football: %MAS vs %ASR vs 30-15 IFT. J Strength Cond Res 36(11):3121–3128.	Dual-pathway rationale	doi
16	Girard O., Mendez-Villanueva A., Bishop D. (2011). Repeated-Sprint Ability — Part I: Factors Contributing to Fatigue. Sports Med 41(8):673–694.	RST builder (work:rest, PCr recovery)	doi
17	Bishop D., Girard O., Mendez-Villanueva A. (2011). Repeated-Sprint Ability — Part II: Recommendations for Training. Sports Med 41(9):741–756.	RST builder (structure, effort windows)	doi
18	Bellinger P. et al. (2020). Muscle Fiber Typology Is Associated with the Incidence of Overreaching in Response to Overload Training. J Appl Physiol 129(4):823–836.	Profile-dependent session priority	doi
19	Lievens E. et al. (2020). Muscle Fiber Typology Substantially Influences Time to Recover from High-Intensity Exercise. J Appl Physiol 128(5):1370–1381.	Profile-dependent recovery / session priority	doi
20	Thron M. et al. (2025). Anaerobic Speed Reserve and Acute Responses to a Short-Format HIIT Session in Runners. J Sci Med Sport 28(3).	Profile-individualised prescription	doi
21	Smith T.B. et al. (2025). Estimating MAS in Academy Soccer Players: Time-Trial Methods vs the 30-15 IFT. Eur J Sport Sci.	TT vs 30-15 protocol comparison	doi
22	Bellenger C.R. et al. (2015). Predicting Maximal Aerobic Speed Through Set-Distance Time-Trials. Eur J Appl Physiol 115(12):2593–2598.	2 km TT validity	doi
23	Thron M. et al. (2022). Overestimation of MAS by the UMTT and 1500-m TT in Soccer. Front Physiol 13:1023257.	Protocol length + shuttle vs straight-line MAS	doi
24	Bangsbo J., Iaia F.M., Krustrup P. (2008). The Yo-Yo intermittent recovery test: a useful tool for evaluation of physical performance in intermittent sports. Sports Med 38(1):37–51.	Yo-Yo IR1/IR2 + VO₂max regressions	PubMed 18081366 · doi
25	Krustrup P. et al. (2003). The Yo-Yo intermittent recovery test: physiological response, reliability, and validity. Med Sci Sports Exerc 35(4):697–705.	Yo-Yo IR1 validation	PubMed 12673156 · doi

Entries 13–14 are [practice] coach resources, not journal articles, so they link to a live page rather than a DOI. The [ours] items — our VIFT-to-sprint- ceiling estimate and our %MAS zone labels — have no external citation by definition; that's exactly what the tag is there to flag.

← PreviousRunning fitness tests Next →Load management