Yeah, and let me make my stance clear as maybe it got a bit confusing with the nature of the argument. I think the way VAR is being implemented in terms of the spots they choose to use or not use it hasn't been great. They definitely need to work on the implementation and the clarification but in terms of the offside calls, the level of the technology, and what happens once they've decided to use VAR to look at something...I would always prefer to have the decision made there than just letting the call on the field be the right one because VAR might be a millimeter off, as mentioned above there's a much larger likelihood that the ref or linesmen get it wrong than VAR does in these specific situations. Again, there have been way, way worse missed or wrong offside calls in the past before VAR was introduced. I do think it's a bit difficult to just say because nobody was complaining about an offside that there's no reason to check; it's really hard to see the entire breadth of the game when you're focused on the ball and a lot is going on at once which can happen a lot in those scrambly defensive situations. Overall though I do agree with your point.
I think a lot of it comes down to understanding why offside exists as a rule. It exists to prevent players from cherry picking and to allow defenders to move up the field without always worrying about an attacker behind them. It was never designed to have the type of accuracy demanded in today's game.
That doesn't inherently mean accuracy is a bad thing, but it certainly means that having a category of "too close to call" absolutely makes sense. Have a set of rules that essentially codifies a situation like this one, or the Son armpit or the Pukki armpit, or anything like that, and lets VAR say "it's too close to conclusively call, so it didn't provide a material advantage or impact on the play and thus it stands"
That still lets you use VAR to overturn obviously offside goals (like Sheffield's yesterday) or count ones that were flagged but weren't like the Arsenal leveler against United, but doesn't end up in this stupid place of some guy zooming in and literally trying to identify which pixel is the one that counts and still possibly getting it wrong (since part of the Liverpool defenders leg was behind the attackers arm in the camera shot and thus you can't establish exactly where it was)
That actually seems to be what you're arguing here:
Yeah, if they can conclusively say it touched his arm I'd agree with you that it would be no goal. To me it doesn't look like it actually does (maybe optical illusion, and it's very close) but hard to say one way or the other. He certainly doesn't control it and the ball doesn't change trajectory, but again the rule as it is it would be enough to call the goal back if it did and I wouldn't complain.
There was a similar situation for United against Sheffield. The review seemed to show the ball hitting the attackers arm, but it couldn't be conclusively proven that it was actually his arm and not just his shirt, thus the goal stood.
That cost United 2 points, but I supported it at the time.
Similar situation with offside. Replay *seemed* to indicate it was offside, but it can't be conclusively proven and thus the call on the field should stand.