Quote from Eastman on January 30, 2006, 07:13:04 PM
Discussion flourishes as individuals gain respect and notoreity through consistent progress and tournament performance. That's how this site works: you trust the opinions of the people you've come to know as solid players, and solid testers/deckbuilders.
This quote outlines what I view as a disturbing attitude towards the format. If it was some random "n00b" with 5 or 6 posts asking for help with a Sui build in the Improvement Forum, it wouldn't merit debate. But, that quote is from a "Vintage Adept" posting in the closed forum. Since rebukes by full members haven't been forthcoming, I have to assume that it outlines a broader view held my many Adepts/full members regarding TheManaDrain.com and its use for advancing and promoting vintage. Not only is the view horrifically flawed due to its reliance on the assumption "solid" players/deckbuilders have useful opinions often enough that they should be defered to, it stifles the only substantial opportunity this format has for significant advancement.
Vintage has relatively few major tournaments, the payout is low, and we have a small player base. Not only that, all tier I/II decks have real overall win/loss ratios no greater/lower than 55/45. As I said in a forum dedicated to discussing the impact/usefulness of mathematical modeling of the vintage environment, few trials with low payout and little competition lead to slow progress. The math itself isn't necessary for understanding this: if 20 people go to a tournament with Stax builds different by only one card, and after two rounds the winner is decided, it won't be very clear if the win was caused by that card or if the pilot just got lucky. Even if that card is Gray Ogre, it has a low probability of diluting the deck enough to make the impact of its presence clear in such a short time.
The immediate critical response is clear, "But Grey Ogre has no place in Stax, we know that. Our
experience and expertise are sufficient to say that cutting a Workshop for Grey Ogre is a bad idea." Ok. I'll buy that, but my point is more relevant when the impact of a card is less clear, for instance putting a random Extract into 5 color stax. Yeah, it's probably a bad idea, overall. But, can you prove it? Sure, it may suck if it's not in your opening hand...but how many cards will you really see against storm-combo if you aren't winning anyways? How much will that one Extract (which could be pure crap 99.99% of the time) dilute the deck as compared to how many times it will randomly take out the opponent's one Darksteel Colossus, or one Burning Wish? If we assume that every opening hand has a 5 color mana source, and one-third (I know that's bullshit) of the decks out there have a single win condition, going first you have about a 3% of randomly winning by casting it turn one. Will adding it dilute your win percentage a full 3% in that environment? Sure your intuition may say yes or no, but why should anyone trust you over testing it?
Testing is where the problems start appearing. Nobody, and I mean nobody, plays enough games to really know how much of an impact switching any single support card with a reasonable alternative has on their deck. That statement is a bit out there, but its underpinnings aren't really controversial. Tier I/II decks (play-skill controlled) really don't vary much in their overall win percentages. When a deck starts winning even 55% of the time against non-mirror matches, most people switch to it. Note Ravager Affinity back in pre-banning Darksteel type 2. Since we can safely assume that most decks have overall win ratios in the 47/53 to 53/47 range, it would be surprising to learn that changing...say...a Vampiric Tutor to an Imperial Seal changed that number by more than .3% (assuming you see 20 cards/game, and having the Vampiric vs. the Seal actually matters a whole 1/10 of the time). To see that kind of change clearly, you'd have to play several thousand games...and record them...and work the stats. Seal, while strictly worse than Vampiric, would have a minimal impact on actual outcomes. In fact, Grey Ogre in that slot should only lose you the game ~3% of the time (20 cards/game, it makes you lose 1/10 of the time you draw it in situations where a meta choice would instead have saved you). That difference is only clear in 3-4 hundred games.
The canned response, "You're actually arguing
for intuition and analytical assessment. If we know a card is "strictly better," we'll sub it in and win more." My response: but you see, you're only a neural network. Sure you may be a very complicated and well-trained neural network, but your opinions and insight are completely limited by how well-trained you are. The fact is that when we have a game like magic that changes so regularly, even "experts" are ridiculously undertrained. We can't really tell if Extract would help or hurt Stax. You may say it would hurt it, and I'd be inclined to agree...but those are only guesses. The very best logical, rules, and guesstimations that anyone here has cannot soundly predict that impact. I feel
very safe suggesting that we'd need to see a few hundred games to really know that answer.
You're probably wondering, "So...where is he going with this? Uncertainty, sure...but...hurting the format? Go test
your decks
yourself."Â The simple fact is that in order to advance the format: we need a faster rate of mutation, we need more trials/selection events, and we need a reward for "successful evolution."Â The best way of approaching this requires a cooperative approach to mutation and testing, while pointing out "strictly better" choices such as Vampiric Tutor over Imperial Seal is a genuine use of expertise, moving a Stax deck running a random Extract to the Improvement Forum is not. (I don't know if this has ever happened, to be honest...Has it?)Â In fact, if everyone on here went out of there way to play 2-3 bad decks from the "Improvement" Forum every day, and gave useful and constructive feedback (towards the deck, the creator is likely an idiot) we'd have a better knowledge base for future development, a more rapidly developping format, and because of that we'd have a better format.
Why spends months mastering a given deck (5C Stax, for example) when the real chances of it being ridiculously supoptimal are so high? Ridiculously suboptimal, btw, means about .5 in its overall win percentage. Why not instead contribute to evolving the format? If Grey Ogretron happens to have a 54% win condition when fully developped, how would be know? Vintage has traditionally developped slower than any other format...we've overlooked many solid cards and decks, traditionally, and the latency period before useful concepts/builds are exploited is still unacceptably long.
By the way, any rebutal focusing on team secrecy as being useful requires the assumption that the posters are actually on teams. Personally, team secrecy can only be useful when a limited group can actually out-innovate everyone else. If that's actually true, it may be time to start asking why.
