The headline is the question that's been popping into my In Box more and more frequently since the big Canon and Nikon mirrorless announcements. As people study the current options and consider what the future options for lenses might be, there's a decidedly chilling effect going on. People are suddenly concerned about putting money into a new lens with an old mount.
The problem, of course, is that you need to both have some idea of where you're going with the camera transition that's currently happening, plus you need to read the minds of the inscrutable Japanese companies who have proprietary locks on their mounts. That last bit means that you also have to guess at how fast reverse engineering happens at the third party makers, and whether those makers will hedge their bets by just doing a quick and dirty "adapter type" change—witness the Sigma Art lenses for the FE mount—or actually design something that takes advantage of the short flange distance with a big throat that the new RF and Z mounts have.
Nikon did the right thing and Canon the wrong thing for a change (role reversal): Nikon gave you a specific plan for new lenses. So we know what Z lenses are coming in the next 18 months or so (20mm f/1.8, 24mm f/1.8, 50mm f/1.2, 58mm f/0.95, 85mm f/1.8 for primes, and 14-24mm f2.8, 14-30mm f/4, 24-70mm f/2.8, 70-200mm f/2.8 for zooms).
So let's start with Nikon and those zooms, because there the answer is a little more clear: if you think you're moving to the Z system in the next couple of years and you were thinking about buying the 14-24mm f/2.8G, 16-35mm f/4G, 24-70mm f/2.8E, or 70-200mm f/2.8E in the F-mount, well, you probably ought to just delay your plans on purchasing those lenses. The expectation is that Nikon will deliver better Z versions than F versions, even though the F versions are already D850-worthy and quite good.
I'd tend to say the same thing about the primes that we know are coming, particularly the 50mm (the current F-mount 20mm, 24mm, and 85mm f/1.8 lenses are all quite good, the current 50mm lenses, not so much).
From there, things get very fuzzy very fast. It's clear that recent E and AF-P lenses seem to work just fine on the Z cameras, and there's no telephoto beyond the 70-200mm on Nikon's road map. Thus, things like the 70-300mm AF-P, the 300mm/500mm PF, and even the exotics are lenses I'd not be worried about picking up right now. Moreover, the folks that shoot these kinds of lenses probably shouldn't be abandoning DSLR for mirrorless in a rush: the DSLRs still have better continuous focus options and far deeper buffers than any Z, so I'd say there's a high likelihood that on safari I might be doing something like putting a Z-mount 70-200mm on my Z7 and an F-mount 400/500mm on my D850. Ditto for sports.
So the true D500, D850, and D5 shooter shouldn't be having much in the way of qualms about picking up new F-mount lenses that they need, particularly for recent lenses and telephoto ones.
Which brings me to another point: opportunity cost. If you delay your decision and then don't take certain types of pictures (or use less optimal gear) until such time as Nikon finally delivers the lens you need for Z cameras, well, you don't get the photo. Certainly not the photo you envisioned.
One of those asking me the headline question was specifically an astro photography shooter. Okay, so what astrological events might occur between now and the time a Z lens you might prefer comes along? We've got a couple of upcoming eclipses, plenty of meteor showers, and all kinds of moon/landscape alignments you might want to shoot. What are you going to shoot them with? The Z lens you're still waiting for? ;~)
Canon shooters have a bigger problem: lack of future information to plan with. Basically Canon has said "we'll do more RF lenses in the future," but the signals about the future of EF-M and EF-S mounts are mixed, for sure, and there's little guidance on basic EF, which just makes the lack of information even more problematic for some shooters. Heck, do you even buy an EOS M camera now as part of your transition from DSLR to mirrorless?
The bigger the information gap, the larger the chilling effect.
The good news on the Canon side is that they've been designing EF lenses fully electronically for some time, and almost certainly have been designing recent EF lenses for use with dual pixel focus on the sensor, which the EOS R also has. The initial news on the Nikon FTZ adapter has been that existing lenses work fine with it for most uses, and I'd expect that we'll have even more confirmation from the Canon side soon that EF lenses also work fine with the adapters (there's three adapter choices).
I've always advocated that you purchase using these criteria in order: (1) need; (2) capability/quality; and (3) price.
So, if you have a need, I wouldn't be reluctant to purchase an F-mount lens. Indeed, I ended up ordering a 500mm f/5.6E PF (F-mount) at the same time as my Z7 order.
Capability/quality is still a slight unknown. You probably need more confirmation that the adapters work fine. A perfect adapter would make it less likely that I'd pick up the Z series 70-200mm f/2.8, for instance, as the current 70-200mm f/2.8E is superb. I really only need to confirm and compare autofocus performance to be certain, but I'm already 70% there.
Price is going to be the interesting factor.
I noted that a lot of people were writing silly things like "US$600 for a 50mm f/1.8 lens, what is Nikon thinking?" They're thinking you might want a really good 50mm after suffering from mostly poor ones during the DSLR era, that's what. Still, the US$220 F-mount version isn't completely terrible, and Nikon is asking 3x the price for the new Z version, so you can see that a lot of folk are going to wonder if there's an economy way to go mirrorless with Nikon (you need to spread the US$150 cost of the FTZ adapter over the number of lenses you'll do this with to get the price leveled correctly, though).
This is where Nikon Z and Canon RF gets interesting. Both companies are making broad, sweeping, and strong claims about a new era of optical performance coming because the new mounts lift previous design restrictions. Both companies have emphasized quality in their initial lens offerings for the new mount. Quality above and beyond what their DSLR lenses attain, even though that is already quite high in many cases.
Companies don't offer high quality at low price (e.g. the 50mm price differential). But then there's that adapter which does allow the consumer to trade off some quality and use older less expensive lenses on the new quality-targeted cameras.
My suspicion is that we'll see far more Canon R purchasers opt to mostly use lenses via adapter than Nikon Z purchasers. But this is just a thesis based upon the generalized profiles of the typical Canon and Nikon DSLR purchaser.
Getting back to the headline question, the answer, I think, is deceptively simple: concentrate on the known knowns. If you know you need/want a certain lens and you know that a mirrorless version is coming in a time period you can tolerate, then sure, wait. But if you're trying to manage the known unknowns in your decision, I'd say you're in analysis paralysis. Trying to guess what lens will come when at what price with what level of quality is a parlor game, not something you should be making decisions about.
