Why the Ironman Marathon Falls Apart After 25 km: Running Durability and Heart-Rate Drift

The professional races at the 2026 IRONMAN World Championship in Kona were decided late in the marathon. Magnus Ditlev started the run 5 minutes 34 seconds ahead of Vincent Luis, but slowed dramatically and was reported walking before Luis passed him. Pre-race favourite Kristian Blummenfelt was dropped and finished eighth. In the women's race, Solveig Løvseth led with less than 8 km to go, then went down twice in the final kilometre while Taylor Knibb ran past to win.

Elite collapses make headlines, but the pattern is familiar to age-groupers. You run the first half of an Ironman marathon at a pace that feels sustainable. Somewhere after halfway, the same pace starts to cost more and more. By 30 km you're working harder to run slower.

Sports scientists increasingly describe this as a problem of durability. This article explains what that means for running, what the largest datasets show, and what you can do about it.

What durability means

In cycling, TriForward's bike durability articles looked at durability through power: why your hour-four watts aren't your fresh watts. The same concept applies to running.

Maunder and colleagues, writing in Sports Medicine in 2021, defined durability as the time of onset and the magnitude of deterioration in physiological characteristics during prolonged exercise. In plain terms: how long can your fresh fitness last before things start to slide, and how far do they slide?

Traditional testing ignores this. A lab test or a 10 km time trial measures your fitness when you're fresh. Two runners with identical fresh results can behave very differently four hours into a race. That difference is durability.

What 82,303 marathoners showed

The largest study of running durability so far comes from Barry Smyth and colleagues, published in Sports Medicine in 2022. They analysed race data from 82,303 recreational marathon runners, including 13,125 women, who wore heart-rate monitors.

They compared internal workload (heart rate as a percentage of maximum) with external workload (running speed as a percentage of each runner's estimated critical speed). In a perfectly durable runner, the ratio between the two would stay the same all race. In reality, heart rate drifts up relative to speed. The researchers called this drift decoupling, and compared the 35–40 km segment with the 5–10 km segment.

Their findings:

• Decoupling was the norm. On average, the ratio rose by a factor of 1.16, and decoupling was first detected at around 25 km (25.2 ± 9.9 km).
• Runners split into three similar-sized groups. About a third showed low decoupling, a third moderate, and a third high.
• Low decouplers held on longest and finished fastest. The low group's decoupling began at around 33 km, and they finished in an average of 217 minutes. The high group's began at around 19 km, and they finished in an average of 239 minutes.
• The least durable runners weren't running harder. The high-decoupling group raced at a lower average percentage of their critical speed (82%) than the low group (88%), yet still faded more.
• Men decoupled slightly more than women (1.17 vs 1.12) and slightly earlier (25.0 vs 26.3 km).
• Durability improved predictions. Adding a measure of durability reduced the error in predicting marathon time from 6.45% to 5.16%.

The key message is that durability varies a lot between people, and it matters for performance independently of fresh fitness.

How often the wall arrives

A second analysis by Smyth, of more than four million marathon race records, looked at the outcome rather than the physiology. He defined "hitting the wall" as slowing by at least 25% for at least 5 km after the 20 km point, compared with each runner's pace over 5–20 km.

By that definition, 28% of male runners and 17% of female runners hit the wall. It was also more common in the three years leading up to a runner's recent personal best (36%) than in earlier years (23%), suggesting that runners in a phase of chasing faster times took more risks with pacing.

Why an Ironman run is a harsher test

Everything in those marathon datasets started from fresh legs at the start line. An Ironman marathon doesn't. By the time you leave T2, you have typically been racing for somewhere between five and nine hours as an age-grouper. Your muscles have done thousands of pedal strokes, your glycogen stores have been drawn down for hours, and in many races the run starts in the hottest part of the day.

Inference: if decoupling in fresh marathoners appears at around 25 km on average, it's reasonable to expect it to start earlier in an Ironman run, possibly within the first half. That would explain why so many age-groupers describe the Ironman marathon as two different races: a first half that feels controlled and a second half that's about survival.

There's a second difference. In a stand-alone marathon, most runners can choose their pace from the gun with some confidence, using recent training and race results. In an Ironman, your fresh running fitness tells you much less about what you can sustain, because so much depends on how the swim and bike went, what you ate and the conditions on the day. That uncertainty is another reason to start conservatively and adjust later, rather than commit early to a fixed pace.

What deteriorates

Several things change during a long run, and they interact.

Heart rate drifts

Cardiovascular drift is the gradual rise in heart rate at a steady workload during prolonged exercise. Coyle and González-Alonso's 2001 review described it as a progressive decline in the volume of blood pumped per beat, starting after about 10–20 minutes, with heart rate rising to compensate. They argued that the rise in heart rate itself, rather than extra blood flow to the skin as body temperature climbs, is the main reason stroke volume falls. Heat makes drift worse, as the next sections show.

In practice, this is why your heart rate at 6:00/km in the last hour of a long run is higher than in the first hour, even if the effort felt identical to begin with.

Running economy worsens

Running economy is the oxygen or energy cost of running at a given speed. Hausswirth and Lehénaff's 2001 review described how it tends to worsen during marathons and triathlons, and linked this to rising core temperature, fluid losses and a shift towards fat metabolism.

A 2024 study by Zanini, Folland and Blagrove measured this directly in 44 well-trained male runners during a 90-minute run at an easy, steady intensity. Running economy worsened in everyone, but by different amounts. In the faster runners (10 km times around 31 minutes), the energy cost of running rose by 2.3%. In the slower runners (10 km times around 42 minutes), it rose by 4.3%. Better runners had more durable economy, at least over 90 minutes.

An Ironman marathon lasts far longer than 90 minutes and follows many hours of swimming and cycling. Inference: the deterioration in economy by the end of an Ironman run is likely to be larger than in this study, but nobody has measured it in that setting.

Fuel runs short

Glycogen stores are limited, and an Ironman run starts after hours of using them. In a study of 221 endurance athletes by Pfeiffer and colleagues, higher carbohydrate intake was associated with faster finishing times in two Ironman races. That's not proof of cause, but it fits with laboratory evidence that carbohydrate supports performance in prolonged exercise. TriForward's articles "Feeding the Fade" and "90–120g of Carbs an Hour" cover the fuelling evidence in more detail.

Heat makes everything worse

Heat increases cardiovascular drift and the energy cost of running. Takayama and Aoyagi's 2025 opinion paper on elite marathoners noted that heat can worsen drift and that data from world-class runners in competition are scarce. Kona's 2026 race was described by commentators as "really, really humid," and the heat and humidity hit several professionals hard. TriForward's "Humid, Not Just Hot" looks at why humidity matters on its own.

Pacing the start of the run

Smyth and Muniz-Pumares's 2020 study of more than 25,000 recreational marathoners estimated each runner's critical speed from their training data. On average, runners completed the marathon at about 85% of critical speed. Runners who started the race faster than 94% of their critical speed were more likely to slow by more than 25% in the second half.

In an Ironman, you start the run after hours of exercise. Your legs often feel surprisingly good for the first few kilometres, partly from relief at getting off the bike. Inference: that sensation is a poor guide to what you can sustain for 42 km. A conservative start costs seconds. A start that triggers early decoupling can cost tens of minutes.

Checking your own durability in training

You can get a rough sense of your durability without a lab. This is a practical method, not a validated test:

1. Choose a steady long run of at least 90 minutes on a flat or repeatable route.
2. Run at an even, easy effort, aiming for a consistent pace.
3. Compare your average heart rate for a given pace in the first third with the last third.
4. Repeat the same session every few weeks under similar conditions.

A small rise suggests good durability at that duration. A large rise suggests you're running out of something, whether that's fitness, fuel, fluid or heat tolerance. Some training platforms report this as "aerobic decoupling." Heat, hills, sleep, stress and illness all affect the result, so look for trends rather than single numbers.

You can repeat the check on a run straight off a long ride. That's closer to what an Ironman demands.

Here's a worked example, with illustrative numbers rather than data from any study. Over the first 30 minutes of a flat 90-minute run at 6:00/km, your heart rate averages 140 beats per minute. Over the last 30 minutes, at the same pace, it averages 150. That's a rise of about 7%. Four weeks later, after more long runs and better fuelling, the same session shows a rise from 140 to 145, about 3.5%. The pace hasn't changed, but the cost of holding it late in the run has fallen. That's the kind of trend worth watching.

If instead the rise grows over a few weeks, ask what changed. Hotter weather, poor sleep, a cold coming on, or a heavier training week can all push heart rate up independently of durability.

Training durability

There's little direct trial evidence on how to train running durability. These approaches follow from the physiology and from the cycling durability research, and are labelled as reasoning:

• Long runs and long bricks. Durability is specific to duration. You can't train what happens at 3 hours by only running for 60 minutes. TriForward's "How Long Does Your Ironman Long Run Really Need to Be?" covers how to do this without excessive injury risk.
• Practise fuelling. Taking carbohydrate in long runs trains the gut and supports durability. See "Gut Training."
• Build heat tolerance before a hot race. See "Heat Acclimation for Race Day."
• Strength training. TriForward's "Two Sessions a Week" reviews evidence that strength training supports durability in cycling. The running evidence is thinner.
• Ride with the run in mind. A steady, controlled bike leg leaves more for the marathon. See "Smooth Beats Surgy."

What we don't know

• No Ironman-specific durability data of this scale exist. The 82,303-runner analysis and the four-million-record analysis are of marathons.
• The running-economy study included only men, and only lasted 90 minutes.
• Heart-rate data are imperfect. Wrist-based monitors, heat and caffeine all affect readings, and the large datasets rely on consumer devices.
• Whether durability can be trained independently of overall fitness is still debated. Better runners showed better durability, but that doesn't tell us whether training durability directly helps.

Common mistakes

• Starting the run at the pace you'd like to average. In most Ironman marathons, the second half is slower than the first. Start below your target.
• Judging fitness only by fresh tests. A great 10 km time doesn't guarantee a great Ironman marathon.
• Ignoring heart-rate drift in long runs. It's useful information about fuelling, heat and durability.
• Under-fuelling the run. Durability falls apart faster without carbohydrate.
• Training long runs that are too short or too hard. Durability needs time at steady effort, not just intensity.

References

• 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
• Smyth B. How recreational marathon runners hit the wall: a large-scale data analysis of late-race pacing collapse in the marathon. PLoS ONE, 2021;16(5):e0251513. doi:10.1371/journal.pone.0251513
• Maunder E, Seiler S, Mildenhall MJ, Kilding AE, Plews DJ. The importance of 'durability' in the physiological profiling of endurance athletes. Sports Medicine, 2021;51(8):1619–1628. doi:10.1007/s40279-021-01459-0
• Coyle EF, González-Alonso J. Cardiovascular drift during prolonged exercise: new perspectives. Exercise and Sport Sciences Reviews, 2001;29(2):88–92. doi:10.1097/00003677-200104000-00009
• Hausswirth C, Lehénaff D. Physiological demands of running during long distance runs and triathlons. Sports Medicine, 2001;31(9):679–689. doi:10.2165/00007256-200131090-00004
• 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
• Smyth B, Muniz-Pumares D. Calculation of critical speed from raw training data in recreational marathon runners. Medicine & Science in Sports & Exercise, 2020;52(12):2637–2645. doi:10.1249/MSS.0000000000002412
• Pfeiffer B, Stellingwerff T, Hodgson AB, Randell R, Pöttgen K, Res P, Jeukendrup AE. Nutritional intake and gastrointestinal problems during competitive endurance events. Medicine & Science in Sports & Exercise, 2012;44(2):344–351. doi:10.1249/MSS.0b013e31822dc809
• Takayama F, Aoyagi A. Do heart rates of elite marathon runners exhibit room for drift? Implications for durability. Frontiers in Sports and Active Living, 2025;7:1571498. doi:10.3389/fspor.2025.1571498
• NBC Sports. Ironman Kona: Taylor Knibb wins world title with pass in final mile, 10 October 2026. nbcsports.com
• TRI247. IRONMAN Kona 2026 as it happened as Knibb and Luis crowned champions (live race report), 10 October 2026. tri247.com