Most athletes think more carbs during a race means more fuel for the muscles. The science is more interesting. The real reason 120 g/hr outperforms 60 g/hr has almost nothing to do with glycogen sparing.
Gut training is a 2–4 week process of systematically overloading your intestinal carbohydrate transporters during exercise, teaching them to absorb more per hour. Without it, 120 g/hr causes GI distress. With it, 120 g/hr keeps muscle damage markers 67% lower than standard fueling protocols.
Why Your Gut — Not Your Legs — Is Limiting Your Race
Up to 93% of Ironman athletes report GI symptoms during the race. Seven percent drop out because of them. Those aren’t bad-luck numbers. They’re what happens when you ask an untrained gut to absorb carbohydrates faster than it was prepared for.
Your intestine uses two independent transport systems. SGLT1 moves glucose and maltodextrin. It saturates at around 60 g/hr — your floor, not your ceiling. GLUT5 moves fructose on a separate track. Think of it like a second fuel pump alongside the first.
When you mix glucose and fructose, you’re running both pumps. Peak oxidation with glucose-fructose blends reaches 1.75 g/min (~105 g/hr) in trained athletes, compared to 1.0–1.1 g/min with glucose alone. Gut training is what raises the pump’s flow rate over time.
The Carbs Per Hour Ceiling: What Transporter Math Actually Means
The 60–90 g/hr guideline most coaches cite came from Smith et al. 2013 (n=51, 12 doses). At the time, 60–80 g/hr was the optimal range. For athletes without gut training, it still is.
The landscape shifted when Podlogar et al. (2022) tested 90 vs. 120 g/hr in trained cyclists. At 120 g/hr, exogenous carbohydrate oxidation reached 1.51 g/min — 17% higher than the 1.29 g/min at 90 g/hr (p=0.026, n=11). That’s a meaningful gap.
Here’s the part most articles skip. Podlogar also measured endogenous carbohydrate oxidation — how much muscle glycogen was being spared. The answer: none. Endogenous oxidation was 2.15 g/min at 120 g/hr versus 2.20 g/min at 90 g/hr (p=0.786). A true null result. Statistically indistinguishable.
So the extra exogenous carbohydrates being oxidized at 120 g/hr aren’t replacing glycogen at all. Something else is going on.
The Real Mechanism: Muscle Damage Suppression
Viribay et al. (2020) ran 20 elite mountain marathon runners at 60, 90, or 120 g/hr over a 42.2 km race with 3,981 m of elevation gain. At 24 hours post-race, they measured creatine kinase (CK), the primary blood marker of muscle damage.
The results are stark:
The 60 g/hr and 90 g/hr groups were nearly identical: CK of 1529 U/L and 1553 U/L respectively. The 120 g/hr group came in at 499 U/L. The LDH (lactate dehydrogenase) increase, another damage marker, was 8.5% in the 120 g/hr group versus 36.7–46.7% in the others (p<0.001).
That’s not a marginal difference. That’s a different race.
Internal exercise load was also lower in the 120 g/hr group (3805 vs. ~4690 AU, p=0.019). Same external work, less physiological cost.
The Podlogar and Viribay data tell a consistent story. 120 g/hr doesn’t save your glycogen. It protects your muscle tissue. The likely mechanism is reduced protein catabolism plus lower inflammatory response when carbohydrate availability stays high.
The 4-Week Gut Training Protocol
Gut training works by upregulating intestinal SGLT1 transporter density. Cox et al. showed a 16.5% increase in exogenous carbohydrate oxidation after 28 days of high-carbohydrate availability training (n=16 cyclists, p<0.01). SGLT1 protein density increases within days in animal models; human adaptation takes 2–4 weeks.
Costa & Miall (2017) showed that 2 weeks of repetitive gut-challenge cut GI symptoms by 60% and halved hydrogen breath test readings (13±6 to 6±3 ppm, p=0.004). Runners also gained 0.6 km in a 1-hour time trial.
Below is the 4-week escalation protocol. All sessions must be at moderate-to-high intensity (above Zone 2 aerobic base) — gut adaptation is stimulated by exercise stress, not rest.
| Week | Target g/hr | Glucose:Fructose Ratio | Session Duration | Product Mix | Monitoring Cue |
|---|---|---|---|---|---|
| 1 — Baseline | 60 g/hr | 2:1 | 90–120 min | 2 gels (22g each) + 500 ml sports drink (16g) | Note any bloating or nausea |
| 2 — Build I | 80 g/hr | 1:0.8 | 90–120 min | 2 gels + 750 ml mixed drink | GI should calm vs. Week 1 |
| 3 — Build II | 100 g/hr | 1:0.8 | 120–150 min | 3 gels + 750 ml mixed drink | Any cramping = back off 10 g/hr |
| 4 — Race Sim | 120 g/hr | 1:0.8 | 150–180 min | 3 gels + 1 L mixed drink + banana or rice cake | Full race conditions |
One detail matters: switch from 2:1 to 1:0.8 glucose:fructose at Week 2. The 2:1 ratio underdoses fructose at higher intakes and leaves GLUT5 underutilized. Jeukendrup’s 2022 review confirms 1:0.8 is optimal above 90 g/hr. Most newer gels (Maurten, SiS Beta, Spring) already use this ratio.
Building the 120 g/hr Stack
A typical hour at 120 g/hr looks like this in practice:
| Source | Amount | Carbs | Notes |
|---|---|---|---|
| Energy gel (1:0.8 ratio) | 3 x 40g gel | 75–90g | Check label for fructose content |
| Sports drink (6–8% CHO) | 500 ml | 30–40g | Don’t mix above 8% — osmolality risk |
| Banana (optional, >3 hr efforts) | 1/2 medium | 12g | Real food improves palatability on long bike legs |
| Total (gels + drink) | 105–130g | Adjust drink volume to hit 120g |
A few practical rules that cut GI risk significantly:
- Drink mix concentration matters as much as volume. Above 8% solution, osmolality pulls water into the gut and causes cramping. At 6–8%, absorption is near-optimal.
- Take carbohydrates every 15–20 minutes rather than every 30. Smaller, more frequent doses avoid osmotic spikes.
- Don’t change your product mix in the final 2 weeks before a race. The gut training is specific to what you’ve been practicing with.
Pro teams already do this. Team Visma runs 3 gels per hour on flat Tour stages (90 g/hr, 2:1) and switches to 3 x 40g products on mountain stages (120 g/hr, 1:0.8). Cam Wurf consumed 200 g/hr during his record Ironman bike split, only possible with years of specific gut adaptation.
Your aerobic decoupling score in long rides is a useful proxy for whether your fueling is working. If heart rate drifts substantially in the final third of a well-fueled long ride, the gut hasn’t adapted yet — the fuel isn’t getting through fast enough to support the effort.
When to Stay Below 90 g/hr
120 g/hr isn’t always the right target. It’s only appropriate for events lasting 2.5 hours or more at moderate-to-high intensity. Pushing 120 g/hr during a sprint triathlon, a 1-hour tempo run, or an easy recovery day creates unnecessary GI load with no performance payoff.
Three situations where staying at 60–90 g/hr is the right call:
- Short events (under 2.5 hr). Liver and muscle glycogen hold enough for ~90 minutes at race pace. The 120 g/hr benefit on a 1-hour event is near zero.
- Low-intensity training. Zone 2 rides and easy runs are fat-burning sessions. Flooding them with carbohydrates blunts the signal you’re after.
- Fructose malabsorption. ~30% of people have impaired GLUT5 function. If mixed-carb products cause distress but glucose alone doesn’t, get tested first.
Training load context also matters: if your fitness score and fatigue score show you’re in a recovery week, gut training sessions can be paused — the adaptation happens when you’re absorbing stress, not when you’re resting.
A Case That Illustrates the Protocol
Sarah, a 42-year-old age-grouper preparing for Ironman Texas, had raced three IMs with the same pattern: controlled through Mile 80 on the bike, then fell apart on the run. Her fueling was 60–70 g/hr with a 2:1 mix.
Eight weeks out, she started gut training. Week 3 at 100 g/hr brought one rough long ride with mild nausea, so she dropped to 85 g/hr for two sessions then resumed the build. Race week, she simulated 120 g/hr on a 3.5-hour brick. No distress.
On race day, she hit 115 g/hr on the bike and 70 g/hr on the run. Her marathon split was 9 minutes faster than her previous IM best. Her legs felt less damaged at Mile 15 than they had at Mile 5 before. That’s the Viribay finding in practice: less muscle breakdown, not more glycogen sparing. Her run fade wasn’t a pacing error. It was a fueling error compounded by hidden muscle damage.
The Ratio, the Timing, and the Science Summary
The standard 90 g/hr advice isn’t wrong. It’s incomplete. For races above 2.5 hours, 120 g/hr with gut training produces a different physiological outcome. The primary mechanism isn’t what you’ve been told.
A quick reference for how the three common intake levels compare:
| Metric | 60 g/hr | 90 g/hr | 120 g/hr |
|---|---|---|---|
| Exogenous oxidation rate | ~1.0–1.1 g/min | ~1.29 g/min | ~1.51 g/min |
| Recommended glucose:fructose ratio | 2:1 | 2:1 | 1:0.8 |
| SGLT1 saturation risk | Low | Moderate | High without training |
| CK at T+24h post-marathon (Viribay) | 1529 U/L | 1553 U/L | 499 U/L |
| LDH increase post-race | +46.7% | +36.7% | +8.5% |
| Gut training weeks required | 0 | 1–2 | 4 |
| Minimum event duration warranted | Any | 90+ min | 2.5+ hr |
| Who it suits | All athletes | Most trained athletes | Gut-trained, long-course |
Three numbers to walk away with: 17% more exogenous carbohydrate oxidized at 120 vs. 90 g/hr (Podlogar 2022); zero additional glycogen spared (p=0.786); 67% less muscle damage in the group that got there (Viribay 2020).
The gut isn’t a passive pipe. It’s a trainable organ.
AthleteOS’s Carb Fueling Planner takes your race distance, expected finish time, and event type to generate a personalized hourly carbohydrate target, the right glucose:fructose ratio, and a 4-week gut training schedule counting back from race day. Try it here.