If we assume that games have, for instance, no replay value,
That's a really bad assumption to have as a part of any argument. Many games have replay value, that's a given. So it's hard to draw any conclusions relying on an assumption that is plainly wrong and not useful. So I can't even understand why that assumption leads to giving strength to BladeOfWraith's comparison. And yes there are unknown variables in his comparison, but since when do we throw our hands up in the air and go 'oh well, there are unknown variables, so anything we do or say is pointless'
Under that assumption (which, we both agree, is inaccurate) it is perfectly accurate to equate buying a game, playing it once, and then giving it away to buying a game, playing it once, and then giving away a copy. Without that assumption (or some other assumption), though, the two situations are no longer identical. That's all I meant by that.
I could pirate a $50 game and play it once. The publisher would (in some vague sense) be "losing" $50 dollars.
The publisher would be "losing" $50 in a very real sense, and you would be committing a crime.
No, and yes. No, they wouldn't actually be losing anything except relative to the case where I would otherwise have bought the game for $50. Piracy is a big deal, and no one is disputing that it's a crime. But not every act of piracy translates to lost revenue: it's perfectly possible that someone pirates a game to try it before buying it (and otherwise wouldn't have bought it at all) or simply pirates a game that they wouldn't have played at all otherwise. That doesn't make it any less illegal, but it does render useless any direct equation of piracy to lost revenue.
The license to play allows you to replay the game, I don't understand what you are going on about. I suppose you are making a hypothetical argument. It's true that I have to pay twice if I go and watch a film at the cinema twice, not that I'm personally someone who does that, though I know some people will watch a movie two or three times if they love the movie. They must pay each time.
Well, you do know what I'm going on about as evidenced in the rest of your post, and it's simply this: the license also allows you to resell the game, or at least it could/used to. In the same sense that a game that can only be played once is a different product (with different value) from a game that can be played over-and-over, a game that can be resold is a different product from one that cannot be resold, and consequently is (at least for some people) of different value.
Well to say the comparison doesn't actually accomplish anything is defeatist and I wonder how any of us get anything done, because nothing can be done perfectly. So many things that we do, we do inadequately.
The question is not and never has been whether the comparison can or should be made approximately. BladeOfWraith made a direct comparison that had all of these implicit assumptions, and then used very specific numbers. I did the same thing with my (very stupid) comparison regarding replayable games: the point is that if you assume equivalency where there is none, it's possible to derive anything (it's the principal of explosion, really).
It's only defeatist to give up at this point. The appropriate thing to do is to instead acknowledge that your comparison is approximate, come up with some reasonable (evidence-based) bounds for how the comparison actually works in practice, and carry this uncertainty through to the end of the derivation. Only then is it possible to (begin to) make sense of how much revenue would be gained or lost in different situations (sharable games, unsharable games, games with different kinds of piracy protection, replayable games, etc.)
Or, we can just assume that any arbitrary equation is always perfect, and we end up with nonsense like this:
Don't like that used games are the exact same thing as piracy?
Hint: they're not. To start with, one is generally illegal, and the other is not.