Touching on a few things mentioned so far (although I skipped a few pages). Obviously the following contains not just a little opinion, so please take it in that light. Feel free to counter my assumptions if you've got a good argument but please let's not devolve into the "640k ought to be enough for anybody" type comments about what is, and isn't, "good enough". That metric changes, if for no other reason.
Vinyl is different partly because the motor generates it's own voltage (like a microphone or loudspeaker) while a DAC is at least partly dependent on a power supply. It would be more remarkable if they *didn't* sound different.
I completely disagree that vinyl degrades with the first and every subsequent play. If the cartridge/arm is poorly setup, that will be the case. However with properly aligned arm/motor combinations, I don't detect any significant degradation with albums that have many plays on the odometer. I find it similar to magnetic tape ... possible in theory, but not common in practice. Changing to a new stylus profile can also result in access to a different, unplayed area of the groove, if wear is an issue on a given pressing.
The built-in phono preamps on classic gear were adequate but not commonly excellent. I too had both a PSAudio, and a Marcoff, plus a Peter Moncrief phono pre back in the day* (at different times) and the difference was audible vs otherwise competent integrated amps and receivers. Not so much on preamps, where good phono performance was more common, but not unheard of either.
The layout of a PCB is fairly critical on any component, but when you are dealing with the gain and signals typical of phono cartridges, excellence in layout pays dividends. That takes engineering skill, which should be rewarded. Similarly, component selection can play a role.
*If* a decent phono preamp retails for $US 300, then we know from the usual industry cost models that the BoM should probably be about $US 75 or thereabouts. Casework and connectors can eat perhaps $25 pretty quickly, leaving maybe $50 for the circuit and PCB themselves. Looked upon in that light, it does not seem unreasonable to me, price-wise. Perhaps more significantly, should you cut the case/connector BoM down to $US 10 or even $US 5, and use adequate quality but off-brand parts elsewhere, it would not be difficult to meet a $50~75 price point. Not easy, but not difficult (which sounds contradictory, I know, but really it's not meant to be).
A decent performing basic phono pre in a reasonably practical casework can be built via following pretty much exactly the Linear Technology application note, for about $100 and a "Good Engineer" could shave that as well. (I am not a Good Engineer; I rely on the 4 or 5:1 BoM to Retail ratio to cover my ass and hopefully if I DIY something I'm happy if at least it wasn't more expensive than a boxed, warrantied unit of similar performance). Speaking from experience, the DIY game gets pretty poor pretty quick as far as saving money goes; it's addictive (always a bad sign) and you end up with stuff that cost you money but isn't making any music, but still need for the next project.
The somewhat famous Muffsy is essentially that exact Linear Technology circuit, and in my estimation is roughly the equal of those early inexpensive 1970's attempts (PS Audio, Moncreif, Marcoff, dB Systems, etc.). I owned a Luxman Lab Reference Series integrated amp during the time I had those various phono pre's in my system, no slouch sonically, and I found value in those examples over the built-in pre. The LT/Muffsy is pretty much right at the entry level DAC range.
Phono cartridges, being motor devices, are varied in their impedance requirements and output levels. Capacitance will affect the HF response and corner frequency with MM variants and apparently isn't irrelevant with every MC cartridge either. There is not one single equalization curve to follow, and even if limited to just the RIAA curve only, cutters do not necessarily follow it exactly. A phono preamp that is flexible in the interface must cost more.
Finally, let's not forget we can hear at least some information below the noise floor (especially anything non-random). In theory, again, something like -87dB/1m is apparently perceived as noise-free via human hearing, (or at least that's what Rod Elliot tells us) yet for some reason components that exceed that figure do sound better.
Is there justification for mega-buck phono preamps when (taking the assumption as gospel) a $300 preamp is perfectly acceptable? I suppose it depends on what you want. I don't expect someone to hand-assemble a phono preamp and earn $10 from the activity. If that example is competing with a machine-made alternative stuffed at the PCB fab in China, maybe $10 is all that is needed to turn a profit.
There will always be these kinds of price variances in Audio components even if we could agree that there is a ceiling on component BoM, if only to amortize the design cost over different sales volumes. Cosmetics are actually pretty expensive, and that matters to some. I read somewhere (Stereophile, perhaps) that one well regarded designer specified a custom Japanese-manufactured switch for one mid-priced High End phono pre (it used two) that in standard form at Qty:1 is something like $28 a copy. There's maybe $US 150 retail right there, and we haven't even built anything that actually performs a function yet.
Finally I do not resent someone who can afford things I cannot, I am perfectly fine with there being Ferraris and McLarens in this world, and if they want to buy megabuck components, and if someone wants to supply such products, I say more power to them both.
* I use a Acoustic Signature Tango now (German or Austrian COO, I forget which); MSRP when current was $US700 although I bought it on closeout at something like $450. A friend uses one of those $35 jobs with his recent vintage Pioneer receiver. It certainly doesn't sound "bad" ... but there is a clear difference compared to the Tango. Thankfully ;-)