"pings alone cannot prove a problem. all routers are designed to prioritise routing traffic over pings. Pings need more CPU time so are designed to be dropped and given low priority."
Indeed. But (a) the evidence is not pings alone and (b) VM do not make available to us other tools.
And something changed for me on 21 December at c.04:00. Prior to that packet loss from pinhs was rare and small. After it wasn't. I have posted the BQM charts showing the dramatic change. Of course that might be the result of a change in the way the BQM charts are compiled. But it's supported by the results of simply pinging (with a Windows command).
Then there's the other evidence I've posted and told VM about: erratic speeds downloading from news.virginmedia.com; and pages not loading.
By all means suggest how these effects can arise locally - despite a VM engineer calling yesterday and pronouncing that he'd fixed the problem ("noise") by changing 3 F plugs.