In case you don't understand how the site works:
If you filter by all, you get stats from ALL players who have at least 1 top200 on any mode. This includes 4v4 random or 2v2 AT. So it's more likely you get polluted stats if someone plays more than a single mode.
Top200 is better when possible because on top of having a stable sample of players, all members have to be top200 which filters as much as possible matchmaking deficiencies.
More power to my argument when we considered that there are less UKF players on the ladder, as only considering top200 makes it a way more homogeneous sample of players (i'll argue the cut should be made at top100 but we have what we have).
...
I agree but we have to work with the limited tools we have at our disposal. The site offers a filter to look at how each faction performs against each other, but as of the moment, we don't have a bigger pool of information as that feature was added recently.
You have to look at it weekly.
Week 17:
Brits vs OH/OKW / Number of games respectively
All
47%/47% - 260/265
Top200
51%/44% - 96/88
Week 18
All
53%/50% 310/228
Top
45%/48% - 101/42
So the LIMITED data points we have atm are not telling us that UKF is doing fine against OKW or leeching wins against it.