NewMutator wrote:I don't know how a game will play out, which is why I'm skeptical that a "best" trade can exist.
The same thing struck a nerve when I read GKGs article last year (or whenever it was he posted it) as well, but I've softened on it a bit lately after participating in and observing a few more tournaments.
My thinking used to go that rock, paper, scissors balance implies hard counters, and Myth definitely has these unit trade-offs. Thus, one would think, for any given trade you should be able to construct a "counter" trade that shifts the selections towards the hard/soft counters for the reference trade, and therefore these is no "optimal trade" for every situation.
However, there are a number of problems with that line of reasoning that have become apparent to me over the past few years:
1) While Myth definitely has hard counters, it doesn't really have pure "rock, paper, scissors" balance in that it's not really cyclical in a lot of cases. Units like Trow, fetch, HG and so on are strictly better than some other units. i.e. there's not really a situation in which a thrall is going to do better than a myrk. Thus the only question is, is a myrk really worth as much a 4 thrall in this situation? The key is that the answer to that isn't really dependent on what the other teams trade is to a large degree... it's more dependent on the map.
2) The ability to choose when and where to fight is extremely valuable. Thus, fast units are really important, and you can't really substitute "counters" to them or you'll just get picked apart bit by bit. This is doubly-true since one place that Bungie strays from rock, paper, scissors balance a bit is that the fast units are generally very powerful melee units as well, excepting perhaps ghol (w/o pus) and spiders.
So really, I'd argue that trades are primarily determined by the map (how open it is, how many units you need to control it effectively), obviously the game type and to a large extent, the game time (i.e. how much time you have to screw around and do damage). To a large extent the game time is fairly standardized at this point based on the game mode, so it really just is map/mode dependent.
I think beyond that the other stuff that you mention - distribution, team-play, etc. - is obviously all very important, but the fundamental point about there existing "counter" builds to the optimal one I just don't think is true any more. And it's not just an argument related to probability and ability to handle unknown enemy trades, I don't think even perfect information would matter much. I mean, GKG basically has posted his strategy for trades and such and it was out there before the last MWC, so any teams playing against him pretty much knew what he was going to do, but still I didn't see any really solid "counter-trades".
At this point the burden of proof is squarely on the shoulders of those trying to prove that trades really are a sort of rock, paper, scissors balance. I tried for a while to prove that (to myself and others), but I no longer think it's the case.
Interesting discussion in any case, and was fun reading your article again grim, although I do recall reading it a few years back.
Milk Man wrote:Also I'd like to point out that, while I agree with the article, noob grim doesn't seem to know that back in the b.net days there was controversy over whether or not custom forms where ethical or chetz, which is why a lot of top level players, especially unrankers, still weren't using them.
Yeah this was always sort of surprising to me since Bungie very clearly made custom formations (and observer constants) work online, whereas any other modifications result in out-of-sync. Why would they have put so much effort into synchronizing formations in multiplayer if it wasn't intended? I'm surprised there was anyone to whom that was not obvious even at the start.
grim wrote:When you don't really fear/care about losing, you don't strive hard enough to win.
That's obviously true, but it's also pretty stupid when people bring it up as an argument of... well... anything. It's fundamentally just an excuse for sucking, of which there are many
"My past self could have beaten your current self" is a typical vacuous, unprovable statement. It's just an attempt to shift the performance metric towards "best per unit care" or "best per unit time played" vs just "best", and a stupid one at that. But basic human psychology compels people to try and justify failure externally rather than taking responsibility, and that's especially true for people who derive a lot of their sense of self-importance from the matter at hand (i.e. Myth skills), so those are usually the ones that you'll see making those arguments