What this site keeps, and what it never sees

Two different things are promised here, and it helps to keep them apart. Your rides are never uploaded, never kept and never shown under your name; that promise does not move. The site pays for itself with advertising, and the ad network may set a cookie with your consent; that is the only tracking on it, and it is described below.

Your rides

• Your ride files are read in your browser and never uploaded. The calculators run entirely on your machine.
• The analysis, the comparison and the prediction run on the server. Your browser decodes the file and sends the decoded samples: time, distance, speed, power, altitude, heart rate, cadence, temperature and GPS position, one row per second or so. The server computes the answer, sends it back, and keeps nothing: no file, no samples, no result, no account. While it works it fetches the open terrain tiles under the ride (Mapzen on AWS), the roads under it (OpenStreetMap, through Overpass) and the wind over it (Open-Meteo), which tells those services the area of the ride, the same disclosure the map makes, and for the wind the date as well.
• The roads are kept; your ride is not. What comes back from OpenStreetMap is the shape of public roads, so a climb is described the same way for every rider. It is stored under no account and carries nothing about you or your ride: a square of map, and when it was last read.
• Map tiles. The map under a route is fetched by your browser from the tile service you pick; the note under the map names it. A tile request tells that service which squares of the world you asked for.
• Weather. When you analyse or compare a ride, the server asks Open-Meteo what the wind did that day, and sends it the ride's date and its midpoint rounded to about a kilometre, and nothing else. It happens with the answer rather than on a click, because the wind is part of how your drag area is measured. The note under the wind chart names the service.
• Nothing is kept unless you ask for it. The one exception happens only when you press a button: saving an answer behind a link, below.

The site's own code sets no cookies and writes nothing to your browser's storage; every visit starts fresh. The one thing on a page that may is the advertising, below, and only with your consent where the law asks for it.

Saving an answer behind a link

Any of the three ride tools will keep its answer and give you a link you can send to someone. No account is involved. Nothing is saved unless you press the button, and the page lists what will be published before you do.

What is kept. The answer as it was on your screen: the distance, the climbing, the times, the power, your weight and where it came from, the temperature, your head unit by model, and, if you chose to include it, the line of the route. The name you gave it. Nothing else.

What is taken out, automatically, before it is stored.

• Your file name, wherever it appears. When two rides cannot be told apart by their dates the comparison names them after their files, so the name is also the label under every chart; both go, and a ride named after its file becomes "Ride A".
• The time of day. The date is kept; the clock is rounded down to midnight, because leaving at 06:14 on Tuesdays is a pattern and a date is not.

The route is your choice, offered when you save: the ends trimmed (the default: the first and last 500 m are dropped, so the line does not start at your door), the whole route, or no map at all.

Who can see it. Anyone holding the link. It is a random 128-bit id, so it cannot be found by guessing. Nothing on the site lists saved answers, nothing searches them, no page links to one, and each tells search engines to stay out. But treat the link as the whole of the protection: anyone you send it to can send it on.

Removing it. Saving shows you a second link, once, that takes the page down. Once, because the site keeps no memory of you between visits: there is no account to attach it to and no address to send it to. Keep it, or the page stays up. Nobody else can remove your page, and we cannot look your key up for you.

Retention. A saved answer is kept until someone uses its removal link.

How the site pays for itself

Ads. The site shows a small number of ads, served by Google and marked as such. With your consent, the ad network may set cookies and use them to choose which ads you see. Where the law requires consent, you are asked before any of that happens, and declining leaves every tool working exactly as it does with it; without consent the ads are shown without personalisation. What the ad network collects is governed by Google's own privacy policy, not this one. None of it is your ride: no figure from an analysis, no file name and no route is ever passed to an advertiser, and nothing you see on a results page is influenced by one.

Subscriptions, later. When there is a paid tier, an account will hold an email address and what is needed to bill it, and this page will say so before it exists. An account will never hold your ride files; whether it holds decoded rides, and what for, is a separate decision that this page will describe when it is made, with a way to delete all of it in one step.

What the site does not do

• No social layer. Nobody's rides, figures or name are shown to anyone else, except what you yourself choose to save behind a link, anonymised, and can take down.
• No tracking beyond the ad network's, and none of that without consent where consent is required.
• No selling, lending or aggregating of ride data.
• No use of your data to train anything.
• No third-party sync. When Wahoo or Garmin sync arrives it will be opt-in, and this page will say what those services receive.

If any of this is found to be untrue of the code, the code is wrong.

See also the terms of use.