Yeah I was going to post this yesterday, here are my results compared to the TBB Results:
[Ookla Single Download Result]
https://www.speedtest.net/result/7899470691
[Ookla Multi Download Result]
https://www.speedtest.net/result/7899474063
[TBB IPv4 Result]
https://www.thinkbroadband.com/speedtest/15454348346...
[TBB IPv6 Result]
https://www.thinkbroadband.com/speedtest/15454349465...
What I did with the Ookla Speed Tests, was to pick a London Server (BT screwed up with allocating my connection with an IP which is Geolocated IP up north 250miles away, yeah I could just keep resetting my connection until I get a good IP, but I cannot be bothered ATM) and then did single tests on all of them and then used the highest one to compare them.
And as you can see all 4 match up.
Here is the closest server they say are close to me due to geolocation LOL:
[Middlesbrough: Single Download Result]
https://www.speedtest.net/result/7899506475
[Middlesbrough: Multi Download Result]
https://www.speedtest.net/result/7899502657
As you see in the above single download result its very low, I am guessing that routing to that server has issues, where as with the multi downloads results are a lot better but still 2 thirds my actual speeds, so something might be wrong with their servers or due to where its 250 miles away, maybe its under load.
[Leeds: Single Download Result]
https://www.speedtest.net/result/7899511117
[Leeds: Multi Download Result]
https://www.speedtest.net/result/7899503884
The results from the Leeds Servers are a lot better and close to my actual speed considering they are about 200 to 210 miles from me.
So like I said you need to pick the one that gives you the best single test result and then use them for that test.
Paul