Your Watch Tracks Everything Except What to Eat: How to Turn Wearable Data Into a Race-Day Fueling Plan

Your Watch Knows More Than You Think

If you run with a Garmin or Apple Watch, your wrist is collecting an impressive amount of data every single session. Heart rate zones, training load, estimated calorie burn, workout duration, pace, cadence, elevation gain — some watches even estimate sweat loss.

And yet, when race morning arrives, most runners are still Googling "how many gels do I need for a marathon."

That's the gap. Your watch is a world-class data collector, but it's a terrible nutritionist. It can tell you your average heart rate was 158 bpm during last Sunday's long run. It cannot tell you to drink 600 ml of fluid and take a gel at the 45-minute mark during your half marathon next month.

This isn't a knock on Garmin or Apple. These devices do what they're designed to do: track. The problem is that tracking and planning are two completely different things — and most athletes don't realise how close they are to having a genuine fueling plan, if only they knew which numbers to look at and what to do with them.

The Three Metrics That Actually Matter for Fueling

You don't need to understand every screen on your Garmin to build a better nutrition plan. Three data points carry most of the weight.

1. Training Load (and What It Tells You About Energy Demand)

Training load measures the cumulative stress your body has absorbed over recent sessions. Garmin calls it "Training Load," Apple Watch surfaces it through workout strain and trends, and Strava's Relative Effort does something similar.

Why it matters for fueling: a runner averaging 6 hours of training per week has fundamentally different glycogen demands than one averaging 12. Your weekly training load shapes your baseline carbohydrate needs — not just on race day, but in the days leading up to it. If your training load has been climbing in the final weeks before a taper, your glycogen stores need more topping up, and your carb-loading window matters more.

Most nutrition advice treats all marathon runners the same. A 55 kg runner doing 40 km per week and a 85 kg runner doing 80 km per week get the same "eat pasta the night before" guidance. Training load is the metric that starts to separate the two.

2. Heart Rate Zones (and What They Reveal About Fuel Source)

Your body burns a mix of fat and carbohydrate during exercise, and the ratio depends heavily on intensity. At lower intensities (Zone 1–2), fat provides the majority of fuel. As you push into Zone 3 and above, carbohydrate becomes the dominant energy source.

Your watch records exactly how much time you spend in each zone during every workout. Over weeks and months, this builds a picture of your typical race-day intensity — and by extension, how fast you'll burn through glycogen on the day.

Here's where this gets practical. A runner who spends 80% of their marathon in Zone 2 burns carbohydrate at a very different rate than one who drifts into Zone 3–4 for the second half (which is what happens when you go out too fast). The first runner might need 40–50g of carbs per hour. The second might need 70–80g to avoid bonking at mile 20.

Your watch already knows which runner you are. You just need to translate that data into a fueling number.

3. Estimated Sweat Loss (and Why It's a Starting Point, Not a Finish Line)

Many Garmin models now estimate fluid loss per workout. Apple Watch doesn't provide this natively, but apps can estimate it from heart rate, duration, and ambient conditions.

The number your watch gives you is a rough guide. Research shows these estimates can be off by 15–30%, which is why we've written before about calculating your sweat rate manually. But even a rough estimate is better than the generic "drink 400–800 ml per hour" you'll find in most marathon guides.

What matters is the trend. If your watch consistently estimates 1.2 litres of sweat loss on a 90-minute run in mild conditions, you know your baseline hydration need is roughly 800 ml per hour. If the same run in 25°C heat shows 1.8 litres, you know warm conditions push your needs up by 50%. That difference should change your race-day hydration plan — and for most runners, it doesn't, because the plan was never connected to the data in the first place.

Where Garmin Connect+ Falls Short

Garmin launched nutrition tracking in Connect+ at the start of 2026, and it's worth addressing directly. The feature lets you log food, scan barcodes, and use AI image recognition to estimate calories and macros. It's a food diary.

What it does not do is take your training data — your load, your zones, your sweat estimates — and convert it into a fueling plan for your next race or long run. You still get generic messages like "you'll need a lot of carbs for this ride" without a specific gram target, fluid volume, or timing schedule.

This isn't unique to Garmin. TrainingPeaks offers nutrition planning through a separate Fuelin integration — a second app, a second subscription, and a workflow that still doesn't pull your training data into a unified plan. Apple Health collects data passively but offers no intelligence layer at all.

The gap across all three: data in, nothing actionable out.

Building a Plan From Your Data (Step by Step)

If you want to do this manually, here's the framework. It takes about 30 minutes and four weeks of training data.

Step 1: Establish your average training intensity. Look at your last four long runs (90+ minutes). What percentage of time did you spend in Zone 3 or above? If it's under 20%, your race-day carb needs are likely in the 40–60g per hour range. If it's 30–50%, you're looking at 60–80g per hour.

Step 2: Estimate your sweat rate. Weigh yourself before and after a 60-minute run (no fluid intake during). The difference in grams roughly equals your sweat loss in millilitres. Do this in conditions similar to your race — temperature matters enormously. Repeat two or three times to get a reliable average.

Step 3: Calculate your carbohydrate target. Multiply your estimated carb-per-hour need by your expected finish time. For a 4-hour marathoner needing 50g per hour, that's 200g of carbohydrate total during the race — roughly 8 standard gels, or a mix of gels, chews, and sports drink.

Step 4: Map it to a timeline. Starting at 30–40 minutes in, space your fuel intake evenly. For 50g per hour using 25g gels, that's one gel every 30 minutes. Pair every gel with 150–200 ml of water (not sports drink — the sugar concentration will be too high and your stomach will protest).

Step 5: Test in training. Run your next two long runs using this exact plan. If you experience GI distress, your gut needs training — start at a lower carb intake (30g per hour) and build up by 5–10g per week.

Why Most Runners Skip This

The honest answer: it's tedious. Pulling data from your watch, cross-referencing zone splits, weighing yourself before and after runs, doing the maths, converting grams to gels, building a timeline — most runners would rather just grab a few gels and hope for the best.

And that's exactly why 40% of marathon runners bonk. Research shows the average marathoner consumes just 21.7 grams of carbohydrate per hour — roughly one gel every 70 minutes — despite knowing they should be taking in more. The plan isn't wrong. The plan just doesn't exist in a form that's easy to follow.

This is the problem Sweatr was built to solve. Connect your Garmin, Apple Watch, or Strava account, and Sweatr pulls your training load, heart rate zone data, and activity history to calculate your personal carbohydrate and fluid targets. It builds a race-day fueling timeline — specific grams, specific millilitres, specific minutes — and maps it to real products you can buy and carry.

No spreadsheets. No bathroom-scale sweat tests. No guessing how many gels "a few" actually is.

The Metrics Your Watch Should Track (But Doesn't Connect)

Here's a quick reference for which watch metrics feed into a fueling plan and what's missing from each platform.

What Garmin gives you: Training load, heart rate zones, estimated sweat loss, Body Battery, VO2 max estimate, recovery time. What's missing: any connection between these metrics and a specific fueling recommendation.

What Apple Watch gives you: Heart rate zones, workout duration, active calories, cardio fitness (VO2 max estimate). What's missing: sweat loss estimation, training load aggregation, and again — any fueling output.

What Strava gives you: Relative Effort, heart rate analysis, pace and elevation data, training log. What's missing: all physiological metrics are estimates, and there's no nutrition layer whatsoever.

Each platform gives you pieces. None of them assemble the puzzle.

What Changes When the Data Is Connected

When your fueling plan is actually built from your data — not from a generic table in a magazine — three things change.

You fuel enough. The single biggest race-day nutrition mistake is under-fueling. When your plan says "take 55g of carbs per hour starting at minute 35," you have a specific action to follow, not a vague intention.

You hydrate for your body, not an average. A runner who loses 1.5 litres per hour in April heat needs a fundamentally different plan than one who loses 0.8 litres. Your sweat data makes that distinction possible.

You stop guessing and start adapting. Conditions change. A cooler-than-expected morning means less fluid loss. A hillier-than-expected course means more time in Zone 3. When your plan is data-driven, adjustments have a logic to follow — not just "I feel okay so I'll skip this gel."

Start With What You Have

You don't need to buy a new watch or a sweat-analysis patch. The data you need is almost certainly already sitting in your Garmin, Apple Watch, or Strava account. The gap has never been data collection — it's data translation.

If you want to do it manually, the five-step framework above will get you a solid race-day plan in about 30 minutes. If you'd rather skip the spreadsheet, Sweatr does it in about 30 seconds — pulling your wearable data and building a personalised fueling plan you can follow gel by gel, sip by sip.

Your watch already did the hard part. Now let the data actually work for you.