He talks about a few things to do. The most notable is how he lapped the lugs. It’s seems pretty easy to do. That’s definitely something I’d like to try with one of my factory Rem 700.

So he changed multiple things without testing in between each change and has no idea what improved precision. No doubt he made some good changes but most seem to be personal preference. Like why would the keymod to m-lol handguard help accuracy? Or changing the stock? Good results, poor testing platform to diagnose improvement.

I wish he’d done one thing at a time and re-tested.

Yep. The thing we need to do is take each of his changes, test it, log the results, and then go on to the next suggested change. "Shotgunning" can fix a lot of things but when you're asked to say which change did what and by how much, it's tough to answer! He did good work; just needs to be laid out in a chronology of some sort.

bad canvas

I think the most significant change that he did was lapping the bolt lugs. It seems to be the biggest bang for buck.

How would you know that based on his testing? That’s the problem, only speculation based on crappy testing protocol.

Lapping the lugs can change the headspace. The RPR in video has a barrel nut and can be adjusted.
With a Remington 700 it’s not so easy.

I did this on my Savage action based on some other youtube video. It certainly made the whole bolt closing stroke smoother and I could definitely see how uneven it was initially and how much better the contact surfaces looked after. It was done on only the action only because I too agree it could definitely change the headspace. I'm a bit skeptical that alone made that difference.

Similarly, I did a bunch of other things to my rifle at the same time, mostly upgrade recoil lug and partial bedding, so hard to say. My shooting skills are still limited so i can't ever really shoot better than about a 0.5moa being my best group ever. But after the rifle modifications, I feel like I had more consistent days. Before, a single 0.5moa was about it that day with others varying widely up to like 1.5moa on the same load. Now if I can get one group close to 0.5moa, then the others will mostly be between 0.5 to 1.0moa.