Taking Facebook as an example, on the main Route Lab web page I can see two Facebook Connect links:
https://connect.facebook.net/en_US/sdk.js
https://connect.facebook.net/en_US/fbevents.js
I think the first is for things such as a 'Like' button and 'Login using Facebook' type button if you accept Facebook accounts. However, I couldn't see any Facebook elements visible in the page. Maybe MRA were going to use it, then changed their minds, but left the code link in there.
The second one is for gathering data about what people are clicking through on ('measure the effectiveness of your advertising by understanding the actions that people take on your website.'). It can be used for non-advertising purposes too.
The info gathered is normally anonymous - if it isn't the web site / app developer needs to declare that.
As to why they continually hit, I guess the page must be auto-refreshing regularly.
The good news is that comparing the Route Lab web page in a default Chrome browser versus one with tracking disabled seems to give exactly the same result - a useable page with no missing features.
[As well as the DuckDuckGo browser, other good ways to avoid similar tracking, such as:
- use a uBlock Origin plugin
- use a Ghostery plugin
- use the uMatrix plugin if you want full control over individual tracking sites (sometimes you need to allow some of them through to just get a website to draw correctly)]