Also, yes, I did heavily debate using 'fewer' vs. 'less' before ultimately deciding on the more lol-cat-worthy 'less'. 
As DG already mentioned, though, I feel like buying power is really what limited you in the early game. i.e. no Silver. Both Sea Hag and Lookout are notorious for drastically limiting your deck's buying power, and imo really require a heavy investment in your deck economy to compensate for. Edit: with the curse-heavy game, and only the slow trashing of Lookout to clear them out, this isn't going to be a good set for an engine, thus the need for a big economy to out-slog the opponent. 
of course, if you're playing IRL you can just have symmetric opening turns...
A lot of players don't seem to like this option (maybe because it's more of a house 'rule'), though. Not to mention I don't know of any tournament/competition that uses symmetric opening turns. Artificial kingdom selection, though, is often used for competitive play, so having the option to have a card picker automatically check for really lopsided 5/2 vs. 4/3 split openers makes sense to me.. provided I can come up with some method of adequately (and on average) evaluating the relative 'strength' of each 5/2 and 4/3 opener pair available in the Kingdom.
I don't think you're goign to convince anyone that money density after 4 turns (or any other $ based metric) is the end-all and be-all way to measure the discrepancy you're after, though I might be a bit more sympathetic to it than SCSN. The effects of trashers, attacks, and gainers are all ignored by that kind of metric (or at you don't see their full benefits), and that means you're missing a lot. Money is always useful, but it's not always the priority.
That's not to say that as an approximation, the metric you're suggesting isn't worth investigating. If you got it working, you could look at what kinds of boards it rejects, and ask whether it's filtering out the right things.
Even that, though, had definite problems at times, based on the exact Kingdom composition (in particular with combo cards that are bought right after the openers). An effective-treasure-value assessment after 4 turns (or 8, or 10, or any number of turns) seems like a pretty crude method of checking for overly-strong 5/2 or 4/3 openers, but I'm at a loss for coming up with a better one currently. Also important is the question of just how 'close' does the 'best' 5/2 vs. 4/3 opening need to be, for players not to feel like the match came down to just opening split luck. For example, if Witch/Chapel (or god forbid something like Mint/Fool's Gold) is available as an opener, what 4/3 opening is 'good enough' to still have a competitive/non-luck-based match? That's somewhat subjective, certainly, and based on the opinions of the players involved.
Treating card draw as effective treasure amounts is an acceptable approach only when you consider Big Money strategies, that is, when you ignore almost everything Dominion is really about.
An example game where I trash down to only 6 cards after T8, at which point my only buying power comes from a Market and a Count. I proceed to gain a whole bunch of stuff (including Duchies for VP) by developing Fortresses, and the only real "buying power" I eventually amass is but an unfortunate side-effect of playing a Pillage each turn, for Spoils are not much better than Curses in a Double-Tactician engine. If you'd do a buying-power/turn analysis, my strategy would be judged as really awful, yet it worked out pretty good
Another example, here I go for some early buying power because I want the Forge, but then I trash down to only $4 buying power halfway T12, after which I bring it back up to $8. I could actually build it quite a bit further, say, to $16. A "buying power" analysis would conclude that this is better, but it's not: due to the lack of +buy, 16 is as good as 8. Furthermore, such an analysis would also say that a strategy that does not use 2 Forges to trash Fortresses into Provinces but does get to 8 a little faster is better, yet it clearly isn't.
Ultimately, I'm interested in (once again) building a method of determining opener 'strength' into my card picker program, to provide an option for minimizing the impact of the 4/3 vs. 5/2 split on the game 'decisions'. I'm not entirely sure if attacks and heavy trashing can be adequately evaluated in terms of opening strength, however.
Early buying power negatively correlates with a strategy's long-term strength when you consider reasonable strategies only (e.g. not things like opening double Curse).
Money graph isalso in dominiate...
Chapel/Witch, for example, doesn't visibly start stomping BM until around turn 10. Hmm..
You can choose turn 4 there specifically, it looks like?
I don't think one of them does this by default, if you are willing to set up dominiate i guess it's about 10 lines of code to et it do what you want. But it's not that easy to set up without expirience, i don't know your background, i could make a branch that does what you want
Will it be able to output deck percentage breakdowns, though? Dominiate's output (like other simulators) is focused on win/loss ratios, but I actually want to see the deck buying power (averaged is fine) at the fourth-turn state.
Yeah, no trasher needed. NV-Bridge is self-sufficient.
Now, where am I going with this? Well, two places. First, this is evidence to me that Goko's rating system sucks.
Although a few of the AIs advantages are a bit brutal (wishing well, for example, it can just pick the most likely card left in its deck, which it can always know precisely).
Here's a very basic formula I asked the engineer who programed the stats:
rating = μ - (2 * σ)
Where μ is the mean of your estimated skill level and σ is one standard deviation from that mean.
Dominion's initial values for μ and σ are:
μ = 5500
σ = 2250
That does not mean TrueSkill. I mean, without any method of how you come up with those things, it could be basically anything. Actually 'Elo-like' is more information than this, and more or less satisfying - I wouldn't expect them to give us the exact algorithm, though of course it would be nice.
When I was younger I wanted to be an actor. Then I realized Patrick Stewart had basically already made all future acting irrelevant.
Using the uh "fun set of 10" generator with default settings seems like fair game for rating. Engines will be more common I guess? But like, let's say Joe uses the generator because he wants engines that he's good at, and Sam says all-Prosperity for me (in this new world of that not forcing Colony), and Fred says who cares, shut up and deal. I don't feel like now the pro board is a joke (unless the rating system sucks as some people think, but that's a separate issue). You have to be pretty good to crush people consistently in whichever of these formats.
The Goko card picker, as is, will throw up Monument super often, almost always has about three Attacks in it, and almost never has Pirate Ship.
the 0-10000 point rating scale is too large (no other game or sport uses a rating system of anything remotely like that interval).
OK, so I agree with your other points, but what does the scale matter? I mean, would it be a better rating system if it only displayed the first three digits of that number making it a "0-100" rating scale instead? I bet internally it's always a floating-point number anyway, it's not like isotropic is storing them as integers!
We know that an auto-match feature is in the works (hopefully at least reasonably similar to the one used on Isotropic), however hand in hand with an auto-match feature to pair players with 'similar' ratings up is ensuring that said 'similar' ratings actually mean something.. something which I have been concerned about for some time. Basically, the rating system Goko implemented is very bad. First, it does not appear to increase in accuracy of ratings (in any reasonable fashion) as number of matches increases (if anything, the system gets less and less accurate as number of matches increases). Second, the K-factor (or whatever mechanism they use to calculate points won/lost) they chose is *much* too high.. ratings fluctuate wildly (part of the inaccuracy of #1), which may be partly related to; third, the 0-10000 point rating scale is too large (no other game or sport uses a rating system of anything remotely like that interval). Fourth, the developers seem to be treating the rating system like a point accumulation system rather than a rating system. As an example, when the rating system recently broke (it temporarily reverted players to zero rating after any match), the developers awarded all affected players a 'free' 300-rating-point bonus, over-inflating ratings and adding more total points into the overall point 'pool', further decreasing the accuracy of the system.
