You don't like thinking of it that way because, hey, you could have bought a more expensive card for that amount. But I say you just take that into account when you give it a rank. You say that's hard to do, but I say you're really already doing pretty similar tradeoff type estimations when you ranked all the other cards.
The reason why I don't like this argument (that you can just take into account MP at higher costs) is that this is tantamount to saying that we could very well rank IGG on the $2 list. Why can't you just take into account that it costs more? As Robz earlier demonstrated when he began trying the LGG exercise, it's just hard. The trade-off estimation for cost is something that we've attempted to abstract away by the very method of dividing the lists according to cost.
But IGG does not have a variable cost. The "effective cost" of MP is defined on an infinite (okay, effectively infinite) domain. There is no reason to put IGG on the $2 list, while there is a reason to put Stonemason on the $2 list (namely, it doesn't belong on any other list. The presence of Stonemason on any particular list of given cost does not tell you its strength as a $2 card that allows you to overpay any amount, because you've fixed the amount of overpay. So then you say, well, you take that into account when you put it on whatever list you decided to put it on, and then I say that's my point, you take the variable overpay amount into account when you decide how good of a card it is.
If it's going to go on a list other than $3, it has to go on all of them. Otherwise you're picking arbitrarily and inaccurately.
I disagree with this. The choice is not neccesaarily arbitrary nor inaccurate. In the example with Masterpiece, I have provided reasoning again and again for why I feel $6+ is the best place for it. I could do the same for Stonemason. I CANNOT do it for Doctor, which is why I concede there. And Herald I would put on $4 anyway, so there's no disagreement there. As far as accuracy goes, I am willing to bet that Masterpiece would be bought a negligible number of times at $3, and also very rarely at $4. So, $5 or $6+? I don't know, you got me there. But either one is more accurate that $3.
In the end, your argument seems to be "rank Masterpiece on the $3 list because that is what it says on the card". But my argument is, yes, that's what it says on the card, but that is not the cost at which it is bought.
I can't see how Masterpiece being on the $5 or $6+ list is more accurate than $3. You seem to be considering all cases of gaining a card other than when you by them during your buy phase (or Black Market) an edge case, but I think there are way too many of those cases to be considering them all edgey. Masterpiece at $3 plays differently than Masterpice at $5, even if "you'd never by it at $3," as you say. But it matters. For Swindler, for Forager, for Forge, for every trash-for-benefit card, for possible three-piling scenarios, etc. It's cost is simply $3, and that matters during the game.
Peddler has a cost of $8, but no one has ever bought a Peddler for $8 (okay, maybe it's happened about as often as Halley's comet comes). It's most often bought for $2 or $0. But it really matters that it's an $8 card.. it's very important in a non-trivial amount of cases. Heck, the article on dominionstrategy about Peddler specifically keyed in that it's $8 price point was a big deal for trashing-with-cost-consideration scenarios. Now even moreso with Butcher and Stonemason (turning a Peddler into two Grand Markets is cool).
I don't agree with your premise that "that number on the card" is just a technicality. It's an important part of the mechanics of the game.