Swingy in Dominion context means that it's quite luck dependant -> the less skilled player is more likely to win, right?
If so, I think Chapel is one of the least swingy cards in the game. My vote would go to either Swindler or Jack
I'm not sure that's actually a good measure. At least, it doesn't really fit my definition of "swinginess" at all. That definition is more, "is this card both strong and simple to use?" If so, then yeah obviously it's going to favor the less skilled player. Embassy does not seem luck-dependent to me. Quite the opposite, in fact; it sifts as it draws. But it is a card that has a very simple strategy, so sometimes a novice player employing that strategy will beat the more nuanced deck. Ditto Jack of all Trades.
I think there are two metrics to consider here. There's "how much does the presence of this card in the game downplay skill" and there's "how much can one draw of this card vary."
The rrenaud thing I love linking to is of course the first metric. Embassy makes the game more about luck by making skill less relevant; I don't need to do anything clever, I can just buy Embassy. The tricky things you can do may be competitive, but they don't just beat up the Embassy guy. In multiplayer it's magnified; I always remember that game from a few years back where three people had turn one Embassy. My buys went Embassy, nothing, Embassy, Province, Province, Province, Province. I did not learn anything about whatever we were playtesting.
Tournament is an example of the second. Maybe you draw it with Province, maybe you don't; it's a big difference. This makes it feel like it's all about luck, even though the card actually favors the better player.
Well, cards that work in strong simple strategies favor weak players for sure, but in a match-up between two highly skilled players, Embassy doesn't really favor the slightly less skilled player as much as something like Chapel does. It depends on the board of course; if it's just Embassy/BM and there's nothing better either player could do, it's just a matter of luck basically, but more often than not, there's an engine that can beat the BM. For example, maybe on a certain board, I would have a 30% chance of beating Stef with Embassy/BM, which is actually pretty good if you're not very skilled, but being me, I might want to aim for something like 40% so I would mirror his strategy anyway. And if there's Chapel, that's a great opportunity for him to draw badly.
I think you're focusing too much on matches between two skilled players.
The whole time, your argument has been, "given two highly skilled players that are reasonably close in skill level, card X makes the less skilled player win more often than they should because of A, B, and C."
rrenaud's chart measures how uncertain a game's outcome is, based on
all matches, including games where people aren't close in skill level.
I have played several games where I drew Chapel with 4 Copper, and my opponent drew it with 3 Estates, and I still won because they made mistakes in the mid-game. I've won games in other situations that looked similarly bad, often because of things like my opponent not respecting potential 3-piles, or getting overconfident in their engine and not trashing down as much as they should have. Against a player of similar skill level, those mistakes shouldn't happen that often and I shouldn't be able to make comeback victories.
I've also played games where my Chapel went to turn 5, my opponent autopiloted the engine, and it felt like there was nothing I could do. Those games exist. But by symmetry there are also games where I get dream draws and there's nothing my opponent can do. At a single game level, a bad Chapel draw can really ruin your chances, but on average I find it easy to believe Chapel helps the better player win more often, and rrenaud's numbers back it up.
If you restricted the matches used to only ones that were between players of similar level, and then further restricted it to only matches among people above a certain level threshold, I imagine you'd get different results.
Edit: more to the point, I'm not sure you and Donald actually disagree on anything, I think you're just arguing from different definitions.