Methodology
How formulas get built, what evidence they rest on, and what open-source means in the context of sports nutrition.
Why this exists
Sports nutrition is a category built on margin, not science. A 30 g gel retails for $3–5. The ingredient cost — maltodextrin, fructose, a pinch of sodium citrate — is under $0.40. The markup covers branding, distribution, and the sustained fiction that the formulation itself is proprietary.
It mostly isn't. The underlying science has been available for decades in peer-reviewed journals: which carbohydrate combinations oxidise fastest, what sodium concentrations support hydration, how osmolality affects gastric emptying. None of this is secret.
What doesn't exist — at least not in one place — is a clean translation of that science into reproducible, auditable formulas that anyone can make and improve. OpenFuel is that translation. Not a brand. Not a product. A specification.
How formulas are built
Every ingredient and dose in an OpenFuel formula is justified by a specific evidence source. We apply a strict hierarchy:
Tier 1 — RCTs and isotope tracer trials
Randomised controlled trials with direct measurement of carbohydrate oxidation rates or performance outcomes. These carry the highest weight. When RCT data exists for a dose or ratio, we use it.
Tier 2 — Meta-analyses and systematic reviews
Quantitative synthesis of multiple RCTs. Used to establish effect sizes, identify dose-response relationships, and confirm consistency across populations.
Tier 3 — Position statements
Expert consensus from established bodies (NATA, ACSM, IOC). Used for practical dosing guidance where RCT data is present but nuanced.
Tier 4 — Narrative reviews and industry guidelines
Used for context, mechanistic explanations, and to identify research gaps. Not used as primary justification for doses.
Each formula in the Research Library includes the specific studies cited. The research alignment score in the Formula Explorer is a simplified composite — it is directional, not prescriptive.
The open-source model
"Open-source" here means something specific. It does not mean "free to use" (though it is). It means:
- 1Formulas are versioned. Every change is documented. You can compare v1.0 and v1.1 of a formula and see exactly what changed and why.
- 2Formulas are reproducible. The specification is complete enough that two people following it independently should produce equivalent products.
- 3Formulas are auditable. Every ingredient, dose, and ratio has a cited reason. Not vibes. Not "our proprietary blend".
- 4Formulas can be forked. If you think the sodium is wrong, or you want a higher-fructose variant, build it. Share it. The format supports that.
The analogy to software open-source is imperfect — food formulas don't have compilers or test suites — but the intellectual property model is the same: if the knowledge is valuable, it should be shared, not hoarded.
Formulation principles
Five core principles govern every OpenFuel formula. They are not arbitrary — each has a specific evidence base.
Multi-transportable carbohydrate strategy
The SGLT-1 intestinal transporter saturates at approximately 60 g/h of glucose. Above that rate, additional glucose-only carbohydrates are not absorbed and contribute to GI distress. By combining maltodextrin (glucose polymer, SGLT-1) with fructose (GLUT-5 transporter), total oxidation rates can reach 90–120 g/h. All OpenFuel carbohydrate formulas use this dual-transporter approach. The default ratio is 1:0.8 glucose:fructose — supported by multiple RCTs as the best balance of oxidation rate and GI tolerance.
Concentration and osmolality control
Fluid absorption peaks in the isotonic range (~270–330 mOsm/kg). Hypertonic solutions — including concentrated gels — draw fluid into the gut rather than out of it, which is why all gel formulas specify water co-ingestion. Drink mixes are formulated to hit near-isotonic concentrations in their intended mix volume (typically 500–750 mL). Osmolality is not just a safety consideration — it directly affects gastric emptying rate and therefore fuel delivery timing.
Sodium individualisation
Sweat sodium concentration varies from approximately 600 mg/L to 1,800 mg/L between individuals — a threefold range. A fixed-dose formula cannot account for this variation. OpenFuel formulas are calibrated to the mid-range (NATA recommendation: 500–700 mg/L) as a starting point, with explicit guidance for high-sweat athletes to scale up. Hyponatraemia risk in ultraendurance contexts is real and is addressed in the disclaimer.
GI tolerance as a primary design constraint
Performance is irrelevant if the formula causes gastric distress. GI symptoms — nausea, bloating, cramping — are the primary reason athletes abandon fuelling strategies mid-race. Every formula is designed with GI tolerance as a first-order constraint, not a secondary consideration. This means: respecting osmolality limits, not exceeding fructose thresholds for untrained guts, and flagging that all high-carbohydrate strategies require gut training to tolerate.
Manufacturing and food safety as part of the spec
A formula that cannot be produced safely is not a formula — it is a hypothesis. Each OpenFuel formula includes manufacturing notes, pH targets (which determine microbial safety and shelf stability), preservative rationale, and explicit limitations on claimed shelf-life without validated testing. This is not a regulatory requirement for personal production, but it reflects how food scientists think about specifications.
What we don't claim
Intellectual honesty requires being explicit about what OpenFuel formulas cannot do.
These formulas are not validated for individual athletes.
The research they are based on used group means. Individual variation in carbohydrate oxidation capacity, GI tolerance, sweat rate, and sodium concentration is large. The formulas are starting points, not prescriptions.
Gut training is not optional at high intake rates.
Protocols targeting 90–120 g/h require progressive gut training to tolerate. Taking a high-carbohydrate formula into a race without training at that intake rate is a risk, regardless of formulation quality.
We do not validate shelf life.
Commercial shelf-life validation requires accelerated stability testing under controlled conditions. OpenFuel formulas include preservative rationale but do not claim validated shelf life. pH and water activity data can indicate safety, but this is not a substitute for testing.
Regulatory status varies by jurisdiction.
Food vs. supplement classification, labelling requirements, and manufacturing standards differ significantly between the US, EU, Australia, and other markets. OpenFuel formulas are educational resources — not commercially approved products.
Performance claims are not made.
We describe what the research says about ingredient mechanisms and population-level outcomes. We do not claim that any formula will improve your performance. That claim would require an RCT with the specific formula in the specific population — which doesn't exist.
Contribution model
OpenFuel is designed to evolve. Formulas are versioned precisely because we expect them to improve as new research emerges, as production feedback comes in, and as the community identifies errors or gaps.
GitHub integration — coming soon
Formula specifications, research citations, and ingredient source data will be maintained in a public GitHub repository. Contributions — corrections, new sources, formula variants, additional research citations — will be accepted via pull request. Each change will require a cited justification.
Until then, if you find an error, have a better source, or want to propose a formula variant, the contact details are in the footer.
The only contribution we won't accept: undocumented opinion. If you want to change a dose, bring a paper.