I used to do things with my reel to reels starting with rerecording then cutting and splicing

Can't be too much longer before I start doing the same on files. Perhaps this year....

I no longer mess with files, although at one time I was processing them with a little upsampling to get the best from them.

It's quite difficult doing anything with clipped files, although some reconstruction filters can rebuild the 'overshoot' from the data samples.

How on earth does that work ?!

Would that make it 99% right, with the possibility of the estimate being wrong ?

I dabble with Audacity when importing from records, but I am still learning how to get better results. I'd like to assume that everyone here is familiar with it, but in the event anyone is not, Audacity is an open source freeware audio editor that is surprisingly robust out of the box and has a large and active community surrounding it with tons of plug-ins and components available to make it do virtually anything you could want, which means it is more than enough for my current needs.

My process for vinyl at the moment looks something like this:

Playback (after cleaning): Sony PS-LX310BT Belt Drive Turntable
That player has hardwired RCA cables, which go to TISINO Female RCA to XLR Male Adapter
Then to Focusrite Scarlett 2i2 USB Audio Interface for capture into Audacity.

Before I start editing anything I adjust the levels on the import. My output from that player is extremely unbalanced, with the left channel being very quiet. The Scarlett allows for independent gain adjustment on the channels, so I balance it there and boost both channels to be roughly as loud as I can get them without clipping.

Once imported I run a variety of effects on the raw .wav in Audacity:

Normalize (DC offset only, no gain)
Change Speed (if necessary, I can only play at 33 1/3 & 45, but this effect can correct other speeds played wrong such as 78)
Equalization (various curves based on media; Invert RIAA, roll-off for spoken word, old media standards, etc.)
Click Removal (high threshold on the entire track, to get any extremely loud clicks/pops)
Noise Reduction (sample the surface noise from the lead-in then apply to the entire track)
Then other various effects based on the source, such as High Pass or Low Pass filtering, Compressor in rare circumstances if the audio is extremely unbalanced, etc...
Then spot editing/fixing with the Repair effect or other minor tweaks. "Repair" works surprisingly well on clips, though it has a max of 128 samples, so you can only fix a fraction of a second at a time.
Once everything is good I finish with Amplify on the entire track to match standard CD audio volume.

Final step is to add track breaks and labels, then export as individual tracks in .flac for playback.

Not sure if this is the kind of editing you were referring to or not, and I suppose it's worth noting that I have a pretty strong PC with a good GPU as I do graphics and video editing as well with Adobe Photoshop Elements 13 & Premiere Elements 13. I'd love to upgrade to the full Creative Suite, but can't justify the price at the moment (they don't sell it anymore, it is only available as a monthly fee now). I might spring for the latest version of Elements the next time there is a good sale, since they seem to do annual updates and I am 5 versions behind, but I digress from the topic at hand.

I always welcome suggestions on what I can do to improve, so if you have any thoughts on my Audacity process please share!

Thanks.