The Reserve Behind the Kick
Why two milers with the same personal best can need opposite training.
Anaerobic speed reserve is the gap between an athlete's top sprint speed and the slowest speed at which they're already flat out aerobically. Two milers can run the same 1500m time off completely different engines, one with a big base and a modest sprint, the other with a smaller base and a fast finish. The reserve is how you tell them apart, and it changes what each of them should train.
The idea has moved from a sprinting curiosity to a middle-distance profiling tool over the last few years, and it answers the question the determinants of the mile leave open: when you know an athlete's engine and economy are fine, what's left to work on, and in which direction.
What the reserve is
Maximal aerobic speed (MAS) is close to the speed at VO2max, the engine written as a running speed. Maximal sprint speed (MSS) is flat-out top speed. The reserve is the difference, MSS minus MAS, the span that Sandford, Laursen and Buchheit's 2021 review uses to explain how athletes hold speeds run faster than the one at VO2max. The same idea written as a ratio is the speed reserve ratio, MSS divided by MAS. Sandford used that ratio and cluster analysis to sort elite 800m and 1500m runners into sub-groups, from 400-to-800m types through to 800-to-1500m types.
A higher ratio means a more speed-biased athlete for a given engine. Sandford used it to separate those sub-groups rather than to hand out fixed cut-offs, and that's the safe way to read it, as a position on a spectrum, not a label stuck on a runner for good.
Figure 01 / Interactive
Three milers, similar engine, different reserve
Runner profile / tap to change
Top sprint speed 32 km/h
Aerobic speed 21 km/h
A big reserve and a strong finish, but the aerobic base trails the speed. This runner's next block leans aerobic, threshold work and volume, so the engine catches up to the sprint.
Top sprint speed 30 km/h
Aerobic speed 21.5 km/h
Speed and base roughly in balance. Sharpen whatever the phase of the season calls for, and let race results decide which way to tip the next block.
Top sprint speed 28 km/h
Aerobic speed 22 km/h
A strong base and a blunt finish. This runner needs speed and mechanics, short intervals, sprints and plyometrics, so they aren't outkicked by athletes with the same engine.
Illustrative values, not measured athletes. The aerobic speed is close across all three, so their 1500m potential is similar; what differs is the reserve and the finish. Numbers are placed inside plausible elite ranges to show the shape of the three profiles described by Jiménez-Reyes et al. (2022), not readings from a test.
Read across the three and the point lands: runners can share an aerobic speed, and so a similar 1500m ceiling, while carrying very different reserves and needing very different work.
Same reserve, opposite meaning
Here's the finding that makes the reserve worth measuring. In the 800m a bigger reserve tends to go with faster times. In the 1500m it flips.
Figure 02 / Interactive
A bigger reserve helps the 800m and hurts the 1500m
Event / tap to change
800m: reserve pays off directly. About a third of the race is anaerobic, so a big top-end and the reserve behind it tend to go with faster times.
This is the event where raw speed most obviously wins.
1500m: the relationship reverses. The race is about 88% aerobic, so a large reserve often flags an aerobic base that hasn't caught up to the speed, and goes with slower times.
For a miler, a big reserve is a training clue, not a trophy.
Direction of the relationship, drawn schematically. The 800m relationship is reported by Sandford et al. (2019a) and the reversed 1500m relationship by Sandford et al. (2019b). The lines show the direction, not measured slopes or a real dataset.
Across 1500m runners a larger reserve relates to slower times, the opposite of the 800m, because the 1500m is far more aerobic, about 88% against roughly 66% for the 800m. So for a miler a big reserve can flag an aerobic base that hasn't caught the speed. One guardrail stops people misreading it: the reserve on its own doesn't tell you how good an athlete is, because a runner with a high MAS shows a smaller reserve simply because MAS is high, and that high MAS is itself what makes them fast. Read the reserve next to the actual MAS and MSS, never alone.
The finish is trainable, and the reserve is where it lives
A championship mile is often slow then brutally fast, a 57 or 58 second last lap off a pedestrian first half. Winning that means shifting to near-sprint speed when you're already deep into an aerobic effort, and how much room you have to shift is the reserve. Profile the reserve and you're profiling the kick, which is why it tells you more in tactical racing than in a time trial where everyone runs honest pace from the gun.
Two kinds of miler
1500m runners split into speed-adapted milers, the 800-to-1500 types, and endurance-adapted milers, the 1500-to-3000 types. Comparing the two, the speed-adapted group showed a higher top speed, a bigger reserve, and better sprint and jump ability than the endurance-adapted group. That's a profiling tool. It tells you whether an athlete's next block should chase aerobic development or speed and mechanics, so the hard sessions fit the athlete rather than a template.
Two profiles, two training menus
Once you've placed an athlete on the spectrum, the block writes itself in outline. The aim is a bias, more of the quality the athlete is short of and enough of the other to hold what they already have.
Figure 03 / Interactive
Same event, opposite emphasis
Runner profile / tap to change
Coaching directions, not a prescription. Each menu is the bias a profile points to, following the profiling logic in Jiménez-Reyes et al. (2022) and the individualised high-intensity approach argued by Sandford et al. (2019a). The individual athlete still sets the detail.
Measuring it without a lab
You don't need a treadmill test to start. Top speed comes off a short flying sprint, a 10 to 30 metre effort with a rolling start into a timing gate or a phone camera framed against markers. MAS or vVO2max can be predicted from race performance: Sandford validated predicting vVO2max from 1500m race times, with race speed tracking vVO2max almost perfectly, around r = 0.90. A sprint and a recent 1500m are enough to place an athlete on the spectrum. How well the various field tests for top speed and aerobic speed actually hold up has since been reviewed systematically, so there's guidance on which protocol to trust.
Retest both a couple of times across a season. The reserve moves as the training bias does, and that movement is the feedback: if a block aimed at the base is working, MAS should climb and the reserve should shrink, and if it isn't, the numbers say so before the results do.
What to do with it, and where it's thin
The workflow is simple. Measure top speed and aerobic speed, take the difference, and use it to decide whether the next block leans aerobic or leans speed. Two athletes with the same personal best can be held back by opposite things, which changes what you program for each of them.
The honest edges matter. Most of this work is on men, much of it is built around the 800m, and the profile groups come from small samples, nineteen runners in one of Sandford's studies and fifteen in the milers comparison. Treat the categories as a starting frame for an individual rather than fixed boxes, and note that women are barely represented in any of it. The reserve points you at what to train. The athlete still tells you whether you were right.
References
- Sandford GN, Kilding AE, Ross A, Laursen PB (2019). Maximal sprint speed and the anaerobic speed reserve domain: the untapped tools that differentiate the world's best male 800 m runners. Sports Medicine 49(6):843-852.
- Sandford GN, Rogers SA, Sharma AP, Kilding AE, Ross A, Laursen PB (2019). Implementing anaerobic speed reserve testing in the field: validation of vVO2max prediction from 1500-m race performance in elite middle-distance runners. International Journal of Sports Physiology and Performance 14(8):1147-50.
- Sandford GN, Laursen PB, Buchheit M (2021). Anaerobic speed/power reserve and sport performance: scientific basis, current applications and future directions. Sports Medicine 51(10):2017-2028.
- Jiménez-Reyes P, Cuadrado-Peñafiel V, Párraga-Montilla JA, Romero-Franco N, Casado A (2022). Anaerobic speed reserve, sprint force-velocity profile, kinematic characteristics, and jump ability among elite male speed- and endurance-adapted milers. International Journal of Environmental Research and Public Health 19(3):1447.
- Thron M, Düking P, Ruf L, Härtel S, Woll A, Altmann S (2024). Assessing anaerobic speed reserve: a systematic review on the validity and reliability of methods to determine maximal aerobic speed and maximal sprinting speed in running-based sports. PLOS ONE 19(1).
All links checked August 2026.