How Long Does Your Ironman Long Run Really Need to Be?

Ask ten Ironman coaches how long your longest training run should be, and you'll get answers from 90 minutes to a full marathon. Some argue you need to know you can cover 42 km. Others say the run is the most injury-prone part of triathlon training and the long ride does most of the endurance work.

TriForward's "How Long Does the Long Ride Actually Need to Be?" looked at this question for cycling. This article does the same for running. The honest answer is that no trial has tested long-run length in Ironman athletes, so most of the evidence comes from marathon runners. That evidence is still useful, as long as you keep the gap in mind.

Why the long run is different in triathlon

For a marathon runner, the long run is the key session of the week. For a triathlete, it competes with long rides, swim sessions and bricks for the same recovery and the same hours.

Running also has a higher injury cost than swimming or cycling. Each stride loads bones, tendons and muscles with forces of a few times body weight, and that load adds up over hours. A long ride, by contrast, is mostly limited by fatigue and fuel, not by tissue damage.

So the question for an Ironman athlete isn't just "What long-run length gives the best performance?" It's "What long-run length gives the most benefit for the least cost, given everything else I'm doing?"

What 441 marathoners showed about the longest run

The most directly relevant study comes from Fokkema and colleagues in the Netherlands, published in the Scandinavian Journal of Medicine & Science in Sports in 2020. They followed 441 recreational marathon runners and 556 half-marathon runners through their preparation, asking about weekly training volume, their longest run, injuries and race results. About 70% of the marathoners were men, and their average age was 41, which is close to the typical age-group triathlete.

For the marathon, compared with a longest run of 30–35 km:

• A longest run under 25 km was associated with finishing about 13 minutes slower (95% confidence interval 5 to 22 minutes).
• A longest run over 35 km wasn't associated with a faster finish. The estimate was about 5 minutes quicker, but the confidence interval ran from 14 minutes faster to 4 minutes slower, so the study couldn't rule out no effect.
• Longest-run length wasn't associated with injury. Runners who went beyond 35 km had about the same odds of injury as the 30–35 km group.

Weekly volume mattered too. Running less than 40 km a week was associated with a finish about 6 minutes slower, and more than 65 km a week with a finish about 14 minutes faster, than the middle range.

Injuries were common. More than half the marathon runners (55%) reported a new running-related injury during the study period. But no training characteristic, including volume or longest run, was associated with injury risk.

This is an observational study. Runners who choose longer runs may also be fitter, more experienced or more motivated, so the associations can't prove that the long run itself caused the faster times. But it gives a useful shape: a clear penalty for keeping the longest run short, and no clear gain from going very long.

The bigger picture from 85 studies

Doherty and colleagues pooled the marathon training literature in a 2020 meta-analysis with meta-regression in the Journal of Science and Medicine in Sport. They included 85 studies and 137 groups of runners, from elite to recreational, of whom 25% were women.

Every training variable they looked at was associated with faster finishing times. That included average weekly distance, the number of runs per week, the biggest week, the number of runs of 32 km or more in the pre-marathon block, training pace, the longest run and weekly running hours. Each explained between 38% and 81% of the differences between groups.

The message is that more running, in total and in long sessions, goes with faster marathons. What the meta-analysis can't tell you is where the benefit levels off, or what the injury cost was. It's also group-level data: faster groups trained more, which doesn't mean any individual will get faster by copying their volume.

What we know about Ironman training

Ironman-specific evidence is thinner. Knechtle and colleagues' 2015 review of 35 years of Ironman research found that high training volume and high training speed were both linked to faster race times, with volume mattering more than speed. Running speed in training was linked to marathon split times in women and to overall race time in men.

The review didn't identify the longest run as a predictor of Ironman performance, because the studies it covered didn't report one in a way that could be tested. Inference: the absence of evidence here is a gap in the research, not proof that long runs don't matter.

Why the injury risk is about spikes, not distance

If longer runs don't obviously raise injury risk, why do so many triathletes get hurt building towards an Ironman? A large Danish study suggests the problem is how quickly the longest run grows.

Frandsen and colleagues, publishing in the British Journal of Sports Medicine in 2025, followed 5,205 runners (average age 46, 22% women) for 18 months using their GPS watch data. That included 588,071 runs and 1,820 running-related injuries.

They compared each run with the longest run in the previous 30 days. Runs that were up to 10% longer, or shorter, were the reference. Compared with those:

• A run 10–30% longer was associated with a 64% higher injury rate (hazard rate ratio 1.64).
• A run 30–100% longer was associated with a 52% higher rate (1.52).
• A run more than twice as long was associated with more than double the rate (2.28).

Changes in weekly distance from one week to the next weren't associated with injury in the same way. TriForward's "Forget the 10% Weekly Rule" covers this study in more depth.

Combined with the marathon findings, the picture is fairly coherent: the eventual length of your long run matters less for injury than how big a jump each long run is from the last.

Durability: what the long run is really for

In an Ironman, you won't run the marathon fresh. The long run's main job is to build durability: how well your physiology holds up over hours. TriForward's "Why the Ironman Marathon Falls Apart After 25 km" covers the 82,303-runner study by Smyth and colleagues, which found that heart rate and pace came apart at about 25 km on average, and earlier in less durable runners.

Zanini and colleagues found that over a 90-minute steady run, running economy worsened by 2.3% in faster male runners and by 4.3% in slower ones. Inference: if the costs of running climb with duration, and those costs are larger in less-trained runners, then at least some training at long durations is needed to prepare for them. That argues for long runs that are long in time, not necessarily in distance.

Turning marathon data into Ironman guidance

This section is reasoning from the evidence above, not a tested prescription.

A longest run of around 2.5 to 3 hours is probably enough for most age-groupers. Inference: at a long-run pace of 5:00 to 6:00 per km, the 30–35 km longest run linked to faster marathons in the Fokkema study takes roughly 2.5 to 3.5 hours. Ironman athletes also do long rides that build overall endurance, and the run is only one of three disciplines. Inference: going beyond about 3 hours of running adds recovery cost for little likely gain, and a longer run probably increases the chance of reaching race day carrying a niggle.

Don't reach that length in a single jump. The Frandsen data suggest keeping each new longest run within about 10% of your longest in the previous month. Inference: going from a 90-minute longest run to 3 hours safely is likely to take several months, not several weeks. This is one reason Ironman preparation needs a long run-up.

Count time on your feet in bricks. A long ride followed by a run of 30 to 60 minutes trains the specific durability you'll need in the race, often at lower injury cost than a very long stand-alone run, because the run is shorter. TriForward's "The Bike-to-Run 'Brick' Problem" covers what changes when you run off the bike.

Keep long runs mostly easy. The marathon studies don't separate long-run pace from length very well. But hard long runs increase recovery time, which costs the rest of your training week. Short sections at goal Ironman effort late in a long run are a reasonable compromise.

Weekly consistency matters at least as much as the longest run. In both marathon datasets, weekly volume was linked to performance as strongly as the longest run. A triathlete who runs four or five times a week, consistently, is likely to be better prepared than one who runs twice a week but includes a heroic Sunday long run.

Walking breaks are allowed. If you plan to walk aid stations on race day, practise it in long runs. TriForward's "Walking the Aid Stations" looks at the run-walk evidence.

A worked example

Here's how this might look in practice for an age-grouper with a 3:00 longest run as the goal, using illustrative numbers. In January, their longest run in the past month is 75 minutes. Increasing the longest run by no more than about 10% each time, with a few weeks at the same length in between, they reach around 90 minutes by late February, 2 hours by April and 2:30 by late May. Their longest run of about 2:45–3:00 comes around six weeks before an August race. In the same period, they build a long brick from a 4-hour ride plus 20-minute run to a 5-hour ride plus 45-minute run.

Inference: this is a slow progression on purpose. It costs nothing in fitness to reach the long run gradually, and the evidence suggests the injury cost of rushing it is real.

Adjusting for your situation

The population averages above are a starting point, not a rule. A few situations call for adjusting them.

• Newer runners and returners should build frequency first. If you're running twice a week, adding a third and fourth easy run is likely to do more for your Ironman marathon than stretching the long run.
• Athletes with a history of running injuries may do better with a shorter longest run and more long bricks, or with run-walk long runs that keep total time on feet without as much continuous impact.
• Faster runners will cover more distance in the same time. Inference: thinking in hours rather than kilometres keeps the load more comparable between a 4-hour and a 6-hour Ironman marathoner.
• Masters athletes often report needing more days between long runs. The studies here don't test that directly, but it's a sensible reason to space long runs out rather than doing one every weekend.

Fuelling the long run

A long run is also a chance to practise race fuelling. Taking carbohydrate on long runs helps you train your gut and teaches you what you can tolerate when running. TriForward's "Gut Training" covers the evidence. Repeatedly doing long runs without fuel, on the other hand, increases the risk of low energy availability, which is linked to bone stress injuries. TriForward's "Under-Fuelled, Not Undertrained" covers this in detail.

Common mistakes

• Running a marathon in training "to know you can do it". The marathon data show no clear benefit beyond 35 km, and the recovery cost in an Ironman build is high.
• Jumping straight to a long run you haven't done in months. A single run much longer than your recent longest is the strongest injury signal in the data.
• Neglecting weekly frequency. Two runs a week plus one very long run isn't the same as four or five consistent runs.
• Making every long run a test. Long runs are for building durability, not proving fitness.
• Skipping fuel to "teach the body to burn fat". See TriForward's "Train Low, Race High — And Probably Don't".

What we don't know

• No Ironman-specific trial exists. The distances and durations in this article are reasoned from marathon studies.
• The marathon studies are observational. Fitter and more experienced runners tend to do longer runs, so cause and effect can't be separated.
• Women are under-represented. About 30% of the Dutch marathon group, 25% of the meta-analysis cohorts and 22% of the Danish injury study were women.
• The interaction with cycling is unknown. Nobody has tested whether a long ride reduces the long run you need, or whether long bricks can replace long runs.
• Age effects are unclear. Masters athletes may need more recovery between long runs, but the studies don't test this directly.

References

• Fokkema T, van Damme AADN, Fornerod MWJ, de Vos RJ, Bierma-Zeinstra SMA, van Middelkoop M. Training for a (half-)marathon: training volume and longest endurance run related to performance and running injuries. Scandinavian Journal of Medicine & Science in Sports, 2020;30(9):1692–1704. doi:10.1111/sms.13725
• Doherty C, Keogh A, Davenport J, Lawlor A, Smyth B, Caulfield B. An evaluation of the training determinants of marathon performance: a meta-analysis with meta-regression. Journal of Science and Medicine in Sport, 2020;23(2):182–188. doi:10.1016/j.jsams.2019.09.013
• Frandsen JSB, Hulme A, Parner ET, Møller M, Lindman I, Abrahamson J, Simonsen NS, Jacobsen JS, Ramskov D, Skejø S, Malisoux L, Bertelsen ML, Nielsen RO. How much running is too much? Identifying high-risk running sessions in a 5200-person cohort study. British Journal of Sports Medicine, 2025;59(17):1203–1210. doi:10.1136/bjsports-2024-109380
• Knechtle B, Knechtle R, Stiefel M, Zingg MA, Rosemann T, Rüst CA. Variables that influence Ironman triathlon performance – what changed in the last 35 years? Open Access Journal of Sports Medicine, 2015;6:277–290. doi:10.2147/OAJSM.S85310
• Smyth B, Maunder E, Meyler S, Hunter B, Muniz-Pumares D. Decoupling of internal and external workload during a marathon: an analysis of durability in 82,303 recreational runners. Sports Medicine, 2022;52(9):2283–2295. doi:10.1007/s40279-022-01680-5
• Zanini M, Folland JP, Blagrove RC. Durability of running economy: differences between quantification methods and performance status in male runners. Medicine & Science in Sports & Exercise, 2024;56(11):2230–2240. doi:10.1249/MSS.0000000000003499