Help and FAQ

Output files: which one do you want?

Course files course.fit / course.gpx: how to import them, by brand

Garmin (Edge series)

• Method 1 (wireless): on a computer, open connect.garmin.com → Training & Planning → Courses → import course.fit → save, then hit Send to Device.
• Method 2 (USB): plug the head unit into your computer, drop the file in the Garmin/Courses folder, then restart the head unit.
• Out on the ride, go to Navigation → Courses, pick it and start; the segment cues fire at the right spots.

With a linked Garmin account you can skip all of the above: press Send to head unit on the race plan page and the route goes straight into Garmin Connect, showing up once your watch or head unit syncs.

Wahoo (ELEMNT / ROAM / BOLT)

• In the ELEMNT app on your phone → Routes → the + in the top right → Import from File, pick course.fit → it syncs to the Routes list on the head unit.
• Cue points come up as on-screen notifications.

With a linked Wahoo account you can skip all of the above: press Send to head unit on the race plan page and the route syncs to the ELEMNT through the Wahoo app.

Hammerhead Karoo

• On a computer, open dashboard.hammerhead.io → Routes → Import Route to upload → sync to the Karoo.

With a linked Hammerhead account, press Send to head unit on the race plan page and the route goes straight into the Karoo route library; it appears as soon as the head unit is online.

COROS

• COROS routes only take GPX, so download course.gpx: tap the file on your phone, choose Share / Open with → COROS App → Save → open the route in your Route Library → Sync with Your Device.
• Each section's power cue becomes a waypoint on the route: while you ride there is an extra waypoint screen showing the distance and climbing left to the next one; before the start you can preview every section under Sections on the route.
• Whether a text alert pops up when you reach a waypoint depends on the model and firmware; if it does not, check the waypoint screen or use the race card as well.

Bryton / iGPSPORT (not yet tested)

• Every brand's phone app imports .fit course files through its Routes or Navigation section (Bryton Active, iGPSPORT App).
• Every brand shows prompts differently; some models only show the route and no text, in which case use the race card as well.

Workout file bike.fit: by brand

This is a structured workout file: no map, it just switches power targets section by section by distance, and the head unit warns you if you ride outside the range. Fewer brands support it:

Garmin (Edge series)

• Plug the head unit into your computer over USB, drop bike.fit into the Garmin/Workouts folder, then restart the head unit.
• On the head unit go to Training → Workouts, pick it and hit start; it cues you segment by segment.
• For race day: load the course file (navigation) and the workout file (power targets) both. An Edge runs them together.

With a linked Garmin account, press Send to head unit on the race plan page and the workout is scheduled on race day in your Garmin Connect calendar, showing up once your watch or head unit syncs. Garmin allows at most 50 steps per workout, so plans with more segments have neighbouring segments of similar power merged.

Wahoo (ELEMNT series)

• Plug the head unit into your computer over USB and drop bike.fit into the plans folder → it shows up under Planned Workouts on the head unit.

Hammerhead Karoo

• dashboard.hammerhead.io → Workouts → Import, upload bike.fit → sync.

Bryton / Coros / iGPSPORT

• Most current models will not import a workout .fit directly. Use the course file (which carries the segment cues) or the race card instead.

Zwift .ZWO: the indoor rehearsal

• Zwift on desktop: drop bike.zwo into the Documents/Zwift/Workouts/(your numeric ID) folder.
• Restart Zwift → Workouts menu → Custom Workouts, and your race plan is in there.
• ERG mode locks the power for each segment by default; kill ERG mid-ride if you want the real shifting feel.
• Each block's duration is worked out from an approximate speed. What matters is the order of the intensities and how each block feels, not the distance you cover indoors.

Race card

• On the results page hit Race card (Print) → print from the browser → print it or save the PDF.
• Cut the table out and tape it to your top tube or your stem. The target is a range: stay in it, and stop chasing a single number.

CdA calibration ride: how to ride for the cleanest drag number

When you upload a FIT file to estimate CdA, the quality of the file decides how accurate it is. A race file (the bike leg of a triathlon or a time trial) is the best material there is. With no race file, ride a training session the way described below and the quality comes close. Coaches can also set it as a session for their athletes.

The calibration ride (about 40 min)

• Where: a flat stretch of 2–4 km (under 3% grade) with even pavement, little traffic and few lights. Levee roads, long riverside straights and closed loops are all ideal.
• How to ride it: hold your race position the whole way, cruise steady (over 25 km/h, no surges), and ride two passes each direction. Cover the same stretch both ways and the system pairs the opposing passes automatically, which cancels grade and wind error.
• What not to do: no drafting (drafting makes CdA come out about 30% too low), no braking except at the turnarounds, and do not cut corners (stretches inside bends are dropped automatically, so fewer corners means more usable data).
• Equipment: exactly what you race. Helmet, wheels, kit and bottle placement are all part of your drag.
• Sensors: zero your power meter first; GPS, speed and elevation all have to be recorded (trainer files will not work).
• Weather: a calm day is best. Some wind is fine, since the system corrects with that day's hourly wind direction and the out-and-back pairing, but on a strong-wind day (over 8 m/s) go another day.

How to tell whether the number holds up

• Look at the range printed with the result: a width within 5% of the middle value means good quality. If it is wider, ride it again the same way, or add 2-3 files from other days and estimate them together.
• If the result says it paired N out-and-back stretches, the two directions are cancelling each other out, which is the best quality you can get.
• If a multi-file upload warns that "files differ too much", drop the file it names first.

CdA history: turn your drag into data that builds up over time

Once logged in, every CdA estimate is saved as a record (the CdA analysis page). Records are combined, with the more reliable ones counting for more, to give a value steadier than any single test. Main features:

• Batch upload: drag a stack of FIT files onto the drop zone at the top of the page (12 at a time); each one gets estimated and filed. Coaches switch the athlete view first so the records land under that athlete.
• Confidence (%): how much this record can be trusted. It only scores high when there are plenty of usable stretches and the numbers are steady. Treat anything under 50% as a rough guide; hover it to see the smallest CdA difference this record could tell apart.
• Outlier and low-quality flags: hover "⚠ Outlier" or "Low quality" for the plain-language reason (drafting, wrong bike type, drift from a slow leak, a dead elevation sensor) and what to do about it.
• Name / setup / notes: click any row to open it, then name it (new helmet test, say), tag it with a setup label (matching labels get pooled into an equipment comparison with the time difference over 90 km), and write notes (which helmet, how the ride went).
• Deep analysis on any record: open the row and hit Deep analysis. A plain-language card tells you first whether the record holds up and which stretch is off; below that, the virtual elevation chart (parallel lines mean a clean estimate) and the rolling CdA curve are there when you want to read deeper.
• Converged value in one click: the estimator on the race plan page shows your converged value and drops it into the power plan in one click; on the Coach dashboard you can write an athlete's converged value into their profile.

Head unit auto-sync: link once, outdoor rides come in on their own

Once you link a Wahoo, Hammerhead, COROS or Garmin account in Account settings, every ride lands in AeroPacer as soon as your head unit syncs. No more dragging FIT files in one by one.

• Supported brands: Wahoo (ELEMNT series), Hammerhead Karoo, COROS, and Garmin (since October 2026). For Bryton and other brands, use manual upload for now.
• What happens automatically after linking: every outdoor ride gets a CdA estimate in your history; a ride on race day is compared segment by segment against your plan for that race, the report is emailed to you and shows when you open the plan on the race plan page; records are tagged race or training; if a ride's CdA differs from your usual by more than 10%, you are prompted to give it a setup tag.
• FTP estimate: your best 20-minute power in the last six weeks times 0.95. If it differs from your current FTP by 3% or more, Account settings suggests it with one-click apply; the welcome page fills it in for you too.
• Suggested IF learns from your races: every race you have compared feeds the intensity you actually held back into the suggested IF of your next plan (adjusted by at most ±0.05). A race where you faded in the second half only ever moves it down, never up.
• What gets skipped: trainer sessions, runs, rides without a power meter, and planned workouts that were never ridden. The reason for each skip is listed in the sync list in Account settings.
• Enter your weight first: estimating CdA needs it. If Account settings has no weight, synced rides are skipped until you fill it in, then picked up on the next sync. If your Wahoo profile has a weight, it is copied over automatically.
• COROS is different: it has no instant notifications, so we check every few hours, and again whenever you open the CdA analysis page. Each account allows at most 50 file downloads per day.
• Garmin is different: Garmin doesn't share past rides, so only rides after you link come in. Each ride is sent as soon as your device syncs to Garmin Connect, which is why Account settings has no "Sync now" button. Rides you upload to Garmin Connect by hand come in too.
• Unlinking: press Unlink in Account settings and the authorization and sync ledger are deleted immediately. CdA records already imported stay and can be deleted one by one. How the data is used: see the Privacy Policy.
• Free accounts also get a CdA estimate for every ride and keep them all, with single-ride values for the latest 5. The converged value, trend, outlier reasons and full history are CdA Pro, and open up together when you upgrade.
• It works the other way too: send a plan to your head unit with one click. Once Wahoo, Hammerhead or Garmin is linked, the output area on the race plan page gains a Send to head unit row: the course file goes into the route library and the workout file into the calendar, no cable or app import needed. On Wahoo the workout only shows on the head unit within six days before the race, while the route is visible at any time; for Garmin, tick the workout and course permissions on the consent screen when you link, or that button won't appear; COROS does not allow writing.

Glossary

What the abbreviations in the Power and intensity row of the results mean:

• NP (normalized power): turns your up-and-down watts into the steady wattage that would feel just as hard. On a rolling course NP is higher than average power.
• AP (average power): the plain average of your watts, every second added up and divided by the time.
• IF (intensity factor) = NP divided by FTP: how hard this ride is for you. A full-distance bike leg is usually 0.66-0.72.
• TSS (training stress score): how much the whole ride takes out of you, counting intensity and time together. The bigger the number, the more your body pays.
• VI (variability index) = NP divided by AP: the closer to 1.00, the more evenly you rode. A section-by-section plan tries to buy the most time with the least unevenness.

Terms you will meet in the settings and in CdA analysis (the little question marks on the race plan page, CdA analysis and coach dashboard all link here):

• FTP · Functional Threshold Power: The wattage you could hold for a full hour of all-out riding. Every wattage in this plan is worked out from it. How to test it: ride all out for 20 minutes and multiply your average watts by 0.95. Enter it higher than it really is and the whole plan comes out too hard, and you will blow up in the second half. Fitness changes, so retest every 6-8 weeks.
• CdA · Drag area: How big you and the bike look head on to the wind. The lower and narrower you are, the smaller the number and the less power you need. On the flat, more than 80% of the resistance comes from it. On the aero bars it is about 0.20-0.26; on the hoods of a road bike, 0.30 and up. A look-up value is only a rough guess and can be 15% out; work it back from a race or training ride file instead and you can get within about 2%. Every 0.01 of difference is worth roughly 1.5-2 minutes over 90 km.
• Crr · Rolling resistance coefficient: The effort the road takes back as your tires roll over it. The better the tire and the smoother the road, the smaller the number. A race tire on good tarmac is about 0.003; a puncture-resistant tire or rough tarmac is 0.005 and up. A difference of 0.001 is worth about 2-3 minutes over 180 km. The tire pressure advice below is worked out from it too.
• Climbing power floor: Your easiest gear is what it is, so on a climb you cannot go any slower even if you want to, and you are forced to push more. Below 60 rpm it gets hard to turn the pedals smoothly, so we work out how slowly you could go on that climb in your easiest gear at 60 rpm, and how many watts that takes. That is the floor. If the plan runs into the floor, the climbing check tells you whether to change your gearing or lower your target.
• Official total elevation gain: The climbing recorded in a GPX file is often inflated. Enter the official figure and the system corrects the course to match it. Files record elevation badly in tunnels and on bridges, so self-calculated climbing often comes out several times too high. Enter the official figure and the system scales the gradients of the whole route back to normal. Leave it blank and the raw values are used, with a warning.
• Race format: Sets how hard you should ride, and how many kilometers before T2 you start easing off. Faster finishers ride harder; longer distances ride easier. The suggested intensity does not come from a fixed table. It comes from how long you are likely to be out on this course: the longer you ride, the lower the intensity you can hold. Then some leg is left for the run depending on the format (full distance is usually IF 0.66-0.72, 70.3 about 0.76-0.84, Olympic distance 0.83-0.88, and a pure time trial the highest), and the last kilometres ease off automatically.
• Pacing strategy: Conservative, Standard or Aggressive: the only difference is how much more you push on the climbs than on the flat. All three strategies use the same watts on the flat; they differ on the climbs. Standard is the most efficient way the system found, and it is the recommended one. Aggressive pushes harder on the climbs and saves more time, at the cost of a tougher start to the run. Conservative pushes less on the climbs, rides more evenly, saves less time, and keeps your legs for the run.
• Target NP: The intensity you want to ride for the whole leg, in watts. Leave it blank and the system works it out from your estimated finish time. Filling this in overrides the system's suggestion, and the watts for all six levels move with it (what you enter is the Standard strategy figure; Aggressive comes out a little higher, Conservative a little lower). Do not enter more than you have actually ridden in a race of the same distance.
• Segment granularity: How finely the rolling sections get cut up. A clear climb is always cut out as its own section, whatever you choose here. Fine = more sections that follow the terrain closely, but a long race card; coarse = fewer sections that are easier to remember. It only affects rolling ground; long climbs and descents are not affected.
• Race-day weather: Temperature and humidity decide how thick the air is; the wind direction makes each section feel like a climb or a descent. Within 9 days of the race an hourly forecast is used, earlier than that the average for the same period over several past years. Headwind sections are bumped up a level automatically and tailwind sections down.
• Tire pressure: Front and rear tire pressures worked out from your tire width, tire type, weight and the road surface. Too hard is actually slower on rough tarmac, because the effort is shaken out of you; too soft and you risk a pinch flat. Treat it as a starting point, check it the night before the race, and adjust for the road on the day.
• What segment pacing buys: At the same intensity, how many seconds a section-by-section plan saves over holding one fixed wattage the whole way. Pushing more uphill, easing off downhill and pressing into a headwind is the split that physically saves the most time. On a flat, windless course there were never many seconds to save, and that is not the engine getting it wrong.
• The six power levels: Every section gets a level: Soft, Ease, Steady, Push, Dig or Stomp. Each level is a fixed multiple of your target NP. While riding, just follow the level and watt range your bike computer shows; no numbers to memorise, and the race card uses the same colours.
• Speed score: A 0-100 score for how easy it is to post a good time on this course. Lots of climbing, lots of corners, too much heat and strong wind all cost points. 100 = flat, cool, windless and few corners. It scores how fast the course is, not how fast you are; your time comes from the power plan above.
• CdA source: A look-up gives a typical value, often around 10% off from yours; working it back from a ride file gives your own number, with an error range. The look-up gives a typical value for your bike type and position. Working it back from a file uses the power, speed, gradient and wind you actually rode in, to get your own number; once you are logged in these are saved as your CdA history.
• Post-race comparison: Upload the file from race day and the plan is compared with what you actually rode, section by section, so the next plan can be corrected. It looks at the watts, speed and time difference for every section, and works back what your CdA really was that day. Sections that are far off are usually about position, fuelling or wind, not the engine getting it wrong. With a linked head unit account, the race-day ride is compared automatically, no upload needed.
• Converged CdA: All your estimates combined into one representative value; this is what planning uses. When they are combined, the more reliable records count for more. Different bike types and positions are kept separate. Records of poor quality are kept but left out of the calculation. The plus-minus next to it is the error range.
• Error range (confidence band): The true value is somewhere in this range (plus or minus). The narrower it is, the more certain you can be. Built from the error on each record, with the more accurate ones counting for more. When you compare two setups, the gap between them has to be bigger than the two ranges added together before it really means anything.
• Usable stretches: How many stretches in this file were ridden steadily enough to be used. The more there are, the more reliable it is. A stretch where power, speed and grade all stay steady counts as one segment. Files with too few segments are marked low quality.
• Outliers: This record is more than 10% away from your others, usually because of drafting, a puncture, or your position slipping. The records are combined with a weighted median, so one or two odd numbers cannot pull the result around. If it was a deliberate gear test, keep it; if you know the file is bad, delete it.
• Low quality: Too few usable stretches, or the elevation record looks wrong. The file is kept but left out of the calculation. Common causes: drafting the whole way, a route with too many bends, the head unit's elevation jumping around, or the nearest weather station being so far away that no wind has to be assumed. You can hide them on the chart.
• Setup tag: Give a set of gear a name (New helmet A, Old helmet) and records with the same name are merged for comparison automatically. When they are merged, the more accurate records count for more. To compare gear, use the same day, the same stretch of road and change only one thing at a time, with at least 3 out-and-back runs per setup.
• Virtual elevation: A fake elevation profile worked back from this record's CdA. The closer it sits to the real elevation, the better the estimate. It works by assuming the CdA is right and that energy does not disappear, then turning power and speed back into a height for every second. Where the two lines separate is where CdA changed: your position slipped, you got into someone's draft, or the wind changed.
• Rolling CdA: How CdA changed over the course of the whole ride. It shows whether your position slipped in the second half. The scattered dots are the values for each individual stretch; the line is the smoothed middle of them. A clear rise in the second half usually means your core gave out and you came off the aero bars, which is very common late in a race.
• Detectable difference: The smallest difference this record could tell apart. If the gap between two setups is smaller than this, you cannot measure it. To measure a small difference, such as changing a helmet, you need more usable stretches, steadier roads and less wind.
• Not synced: How far the combined value is from the CdA stored in the athlete's profile. One click updates it. Planning uses the value stored in the profile, not the combined one. This is not flagged when a coach has deliberately picked a single record.
• Needs attention: Athletes whose race is coming up, who have no CdA in their profile, or who have an odd-looking record, are listed first.

FAQ