Two Sessions a Week: The Best-Evidenced Way to Train Durability

Of everything in the durability research canon — fresh-versus-fatigued power testing, carbohydrate dosing, heat, aerodynamics, pacing — training interventions are the weakest link. Most of what claims to “build durability” is observational: athletes who happen to ride more, or train a certain way, also happen to fade less. That's useful context, but it isn't proof that doing X causes Y.

There is exactly one intervention in this evidence base that was tested the way a claim like that needs to be tested — a properly randomised trial, with a performance-matched control group, that used a fatigue-resistance outcome and found a genuinely large effect. It wasn't a fuelling protocol, a pacing drill, or a bike-specific workout. It was strength and plyometric training, delivered twice a week for ten weeks — and it more than doubled a runner's ability to keep going once fatigued, while the control group who trained without it actually got slightly worse at the same task.

This article is about that trial, what it actually showed, and — just as importantly — what it didn't test, because the honest limits matter as much as the headline number.

For a triathlete training 6–10 hours a week around a job and a life, strength training usually gets treated as optional — the session you skip first when the week gets tight, because “durability” sounds like something you build on the bike and run, not in the gym. That instinct is understandable, but it runs directly against the strongest single piece of evidence in this research area.

TriForward has already covered how little strength training a time-crunched triathlete actually needs, and separately compared plyometrics against heavy resistance work for running economy. This article sits next to both of those but asks a different question: not how much strength training you need, or which type improves running form, but specifically whether strength training protects you against the kind of late-race fatigue that durability research is built around — and, if it does, when in the week you should actually schedule it so it doesn't fight with your bike and run training.

What the evidence says

The honest baseline: training improves durability, in a population that doesn't answer your question directly

Before getting to the strength result, it's worth being upfront about what else exists, because the gap matters. The clearest randomised trial that used a full durability outcome — a 3-hour constant-load ride — to compare training approaches was conducted in 35 previously sedentary-to-recreational adults, randomised to ten weeks of either low-intensity or high-intensity training. Both groups improved durability: low-intensity training with a moderate effect (p = 0.03, Hedges' g = 0.49), high-intensity training with a slightly larger one (p = 0.01, g = 0.62), with no significant difference between the two approaches (p = 0.42). Drift was smaller and delayed further after low-intensity training than after high-intensity training.

That's genuinely useful evidence that durability responds to training at all. But the population was untrained, the intervention was general endurance training rather than anything targeted, and neither group added a new component on top of an existing programme — both simply trained differently. It tells you durability is trainable in principle; it does not tell a trained triathlete what to actually add to their week.

Everything else in the training literature is weaker still. Time spent below the first ventilatory threshold and a shift toward polarised training were associated with improved fatigued-state power in 30 U23 professional cyclists (r = 0.43 and r = 0.414), but a separate prospective 8-week study in 10 semi-professional cyclists found no clear relationship between training-load metrics and change in fatigued-state power at all. Associations that appear in one trained sample and disappear in another are exactly the kind of evidence that shouldn't be over-read — which is part of why the strength and plyometric trial below, with an actual control group and a randomised design, carries more weight than any of it.

The strength and plyometric RCT: the strongest, most specific result available

The trial that answers the “what to add” question comes from a different discipline entirely. Twenty-eight performance-matched, well-trained male runners (10 km time 39:02, V̇O₂max 58.6) were randomised to either continue normal training or add maximal-strength and plyometric training twice a week for ten weeks. Both groups then completed a 90-minute heavy-intensity run, and the difference was large.

Running economy at the 90-minute mark improved by 2.1% in the strength group and worsened by 0.6% in the control group (interaction p = 0.04). More strikingly, time to exhaustion at 95% of each runner's pre-test V̇O₂max changed by +35% in the strength group versus −8% in the control group (interaction p = 0.004, η²p = 0.28).

That is a properly randomised trial, with a training-matched comparison group that trained just as much but without the strength component, using a fatigue-resistance outcome as its actual endpoint — a rarer combination in this literature than you'd expect, and the reason this trial anchors the whole article. Where the Matomäki trial compared two active training styles against each other, this one isolated the specific effect of adding strength and plyometric work against a matched group that didn't add it — which is the design that actually answers “should I add this.”

The catch that has to be stated plainly

This was a trial in male runners, not cyclists or triathletes. Running and cycling load the body differently — running is an impact sport with a stretch-shortening cycle at every footstrike, which is exactly the kind of stimulus that strength and plyometric training targets — so the mechanism that plausibly explains the running result does not automatically transfer to a non-impact sport like cycling.

Applying this finding to bike durability is a reasonable, evidence-informed inference — strength training that improves force production and neuromuscular resilience has plausible carryover to any fatigued-muscle problem — but it is inference, not demonstration. Nobody has run the equivalent trial with bike durability as the outcome. For a triathlete, the honest reading is this: strong evidence that strength training protects the run leg specifically, and a reasonable but unproven bet that some of the benefit carries to the bike.

The scheduling rule: when strength work costs you the least

Knowing that strength training helps is only useful if you can fit it into a week without it fighting your other sessions — and there is a specific, evidence-backed rule for that. A meta-analysis of 15 studies and 197 trained male participants found that acute aerobic exercise reduces subsequent muscle strength (a standardised mean decline of 0.79, p = 0.003) but does not meaningfully reduce subsequent power output (SMD 0.04, p = 0.846).

Two details matter for scheduling: the strength decline was larger after more than 30 minutes of prior aerobic work (SMD 1.02), and it was significantly larger after cycling than after running (SMD 0.79 vs 0.28, p < 0.0001).

Put those two findings together and the scheduling logic is straightforward: a strength session stacked directly after a ride of any real length is working against a body that has already lost a meaningful chunk of its strength capacity — more so than the same session after a run of equal length. If you want the strength stimulus to actually land, the gym session belongs before your ride, on a separate day from your longest ride, or at minimum well clear of more than half an hour of prior cycling.

Practical application

• Two sessions a week, for ten weeks, is the dose that produced the measured result — not a daily habit, not an occasional add-on. If you're not currently doing dedicated strength and plyometric work, this is the frequency to build toward, progressively.
• Prioritise technique competence before adding meaningful load or any plyometric/jumping component. The trial used well-trained runners; the exercises themselves — maximal-strength lifts, plyometrics — carry real injury risk if the movement pattern isn't solid first. Build competence with bodyweight and light load before progressing.
• Schedule the strength session away from your longest or hardest ride where possible — ideally on a day without a long bike, or before rather than after one, given that cycling blunts subsequent strength capacity more than running does, and that the effect grows past 30 minutes of prior aerobic work.
• Progress load and plyometric volume gradually. The 10-week trial built up over that period; jumping straight to advanced plyometrics or heavy loading in week one is not what was tested and carries disproportionate injury risk, especially for anyone also running meaningful weekly mileage.
• If you have an existing injury, a history of joint or tendon issues, or you're returning from time off, get sign-off from a physiotherapist or physician before adding plyometric or maximal-strength work — the evidence base here comes from healthy, well-trained runners, and it says nothing about how this protocol behaves in an athlete carrying an existing problem.
• Treat the bike-durability transfer as a reasonable bet, not a guarantee. Keep doing strength work because the running evidence is strong and the injury-prevention and general-strength case for triathletes is well established elsewhere — not because this specific trial proved it will fix your fourth-hour bike watts.
• If choosing between session styles, note the Markov meta-analysis found aerobic exercise reduces subsequent muscle strength but not subsequent power output. That's a reason to be a little more relaxed about placing lighter, power-oriented work near a ride than about placing a genuinely heavy strength session there — though this distinction hasn't itself been tested for durability outcomes, so treat it as a scheduling nuance, not a rule.

Common mistakes

• Stacking plyometrics on top of high running volume without adjusting anything else. The trial added strength and plyometric work to an existing running programme; simply bolting jump training onto an already-heavy run week is a known way to raise injury risk, not the protocol that was tested.
• Assuming the 35% time-to-exhaustion improvement will show up on the bike in the same magnitude. That number is a running result in male runners. There is no equivalent bike-durability number to expect, and treating +35% as a bike prediction overstates evidence that was never collected.
• Doing strength work straight after a long ride “because that's when I have time.” The evidence suggests this is close to the worst placement available — strength capacity is already reduced more by prior cycling than by prior running, and more so once the ride passes 30 minutes.
• Skipping strength training entirely because “it's a running study, not a triathlon one.” The population caveat is real, but it doesn't make the evidence irrelevant — it makes the transfer to cycling an inference rather than a certainty, which is different from having no reason to believe it helps at all.
• Jumping straight to maximal loads or advanced plyometrics without a technique base. The runners in the trial were well-trained and presumably had a reasonable strength-training history; starting there without building competence first changes the risk profile substantially.

How to apply this week

• Pick two fixed slots in your week for strength work and protect them the way you'd protect a key bike or run session — the dose that worked was consistent, twice weekly, not opportunistic.
• If your week includes a long ride, place your strength session on a different day, or earlier in the same day before the ride, rather than immediately after it.
• Start, or restart, with a technique-focused block: bodyweight squats, lunges, single-leg work, and basic hop or skip drills before adding load or box jumps. If you're already competent and simply haven't been consistent, resume at a lighter load than your last peak and build back over 2–3 weeks.
• If you have any current injury, joint pain, or a history of tendon issues, book a session with a physiotherapist before adding the plyometric component specifically — that's the part of this protocol with the highest injury cost if introduced carelessly.
• Log how you feel on your next ride or run the day after a strength session, particularly if you've just added plyometrics. Early soreness or unusual heaviness is a normal adaptation signal, but it should fade within a few sessions — if it doesn't, pull back load or plyometric volume before pushing forward.

References

• Matomäki et al., Frontiers in Physiology, 2023 — 10-week LIT vs HIT durability RCT in untrained adults
• Zanini, Folland, Wu & Blagrove, Medicine & Science in Sports & Exercise, 2025 — strength and plyometric RCT in trained runners
• Markov et al., Sports Medicine, 2022 — meta-analysis of acute aerobic exercise effects on subsequent strength and power
• Spragg, Leo & Swart, European Journal of Sport Science, 2023 — training distribution and fatigued-state power in U23 professionals
• Voet et al., International Journal of Sports Physiology and Performance, 2025 — training-load metrics and fatigued-state power in semi-professionals