I'll see what they all have in store after Christmas
I always tell people buy the best they can when they're ready, prices are always in flux, especially this time of year and if you keep waiting for the next/better thing you will be in perpetual wait mode because something else is just around the corner.
Wonderful advice in all areas of life.
Thankfully, here in this country we have a nation-wide auction platform that is very reputable and very advanced — full-blown buyer protection plus a lot of cool functions such as the ability to track searches and get daily notifications when something new pops up in the category. So it doesn't cost me that much time or effort to keep tabs on a bunch of 7–9th-gen CPUs, but yes, I've experienced it very empirically that bargain-hunting is not worth it.
See, I'm a freelancer, like probably a large minority of other posters here. Jobs, or at least offers, inquiries, RFPs and all that jazz come like busses — sometimes there are none for an hour, sometimes six at once. When I get in PC upgrading mode, it interferes with my productivity and prevents me from being able to finish previous projects early and taking on new ones early, thereby making more money in the relevant month — usually obviously by more than the difference that exists between the prices of an i5 and an i7 within the same range or e.g. RTX 2060 vs RTX 2070 or whatever. Hence if it's time for an upgrade, it doesn't really pay to get submerged in the whole seeking-comparing-haggling mindset, it's better to just focus on making the money first, then spending it. I felt this when buying my last computer a couple of years ago, and to some extent also now, that is last week — wasted a bunch of hours scouting keyboards instead of focusing on my work and then buying the best.
It really is the most productive, gainful strategy to focus on the best, provided it doesn't cross the point of diminished returns by a very bad margin.
People tend to focus on not overspending, but that's more psychological distractionst than rationality. People also want the best bang for the buck, bu they fail to realize chasing it is not efficient — it's not the best bang for the buck when it comes to investing their time. For example when your lost wages due to the time invested in saving money via bargains are worth more than the savings (my own case when buying the last computer).
The difficult challenge here is to actually learn, then make the somewhat bold decisive step of nipping the whole analysis in the bud to lock on the top of the line as a target, the task being to get the money for it. Producing more money is often faster, easier, more efficient than making decisions about how to spend the limited resources one already has. This is probably abundance vs scarcity, which would explain why most people don't and won't get it — same reason why most of us aren't millionaires or even successful career people.
From a practical standpoint, overspending on a powerful CPU or GPU can truly be somewhat of a waste, but as long as one isn't lapsing into an episode of legal insanity, one should be fine. It's probably easier to live with 'yeah, well, I overspent a hundred bucks, let me just put some more ultra settings on for the sake of it' (or enjoy some other non-necessary perks) than 'crap, I skimped on a hundred bucks while already spending half a thousand, and now the whole experience is marred by the corners I cut, with my performance tangibly affected by the negligible savings'. It's easier for me, at least, to let go of the wasted bucks than the missing fps.
Plus, if you overspend a little up front you can avoid intermediate upgrades.
However, of course, overspending cuts into your budget for your next rig, so it's still smarter to not buy things that won't be needed, e.g. gaming features if never gaming, or server/workstation/rendering power while mostly just gaming and doing office work.
So I'm trying to apply this to my situation and right now consciously making the decision to just focus on my work and only keep tabs on the prices, stop overthinking. But easier said than done, of course.