|
When I first started playing and getting into Starcraft 2 the most omnipresent and universal piece of advice I was always given was learn standard builds. I personally cringe at the word and idea of "builds" now (almost as much as Day9) but back then I just did as I was told without even so much as a grain of salt. Needless to say their advice did help me get to understand the fundamental mechanics of the game better. But now as I look back at the actual advice I was given I'm starting to question the overall validity of it.
When people suggested that I 'just play standard' and would proceed to (helpfully of course) show these 'standard' openings I was really thankful at the time and I still am now. But as I play more and more and get more and more into the act of scouting (as it used to be a bad habit of mine to forget or some cases even consciously omit scouting up until I finally realised how important it really was) I've stumbled across a very disturbing thought.
What if the advice I was given has a timestamp on it? What if it came with a limited time warranty?What if I learned everything for nothing? What if all that time I put into learning these openings is going to be made worthless?
Why would I be worrying about these things? Why should anyone? Because of a daunting paradox that crossed my mind that I at first shrugged off but became increasingly perplexed about:
If a strategy works, it can't work, because it worked.
If 'A' wins, 'A' loses, because 'A' won.
Logically speaking, the idea of playing standard in and of itself should be a complete fallacy of logic, a complete impossibility. To play standard, is to play in such a manner that has had a high ratio of success on a widespread level. But if such a strategy has gotten that degree of success and attention and universal awareness, then by that same token and point in time it should already be useless by virtue of how predictable and used it is in the first place because by then countermeasures and the obviousness and unoriginality of it should negative its potentcy.
If people know it's coming, why should it work? Obviously human error, and micro and involved, but dislocating those pseudorandom factors what really is behind this all? Why does 4 gate still work? Why do standard openings or playstyle exist? Because logically, they shouldn't because by the the definition of standard defies logic itself. Something can't be both well known and unexpected. And if it's something is well known, it is (by a competent and knowing player) prepared for. So why are any particular strategies or styles even able to reach any respectable levels of notoriety or wide stream use? Is Starcraft defying the very concept of logic? Or is there some deep twist to the metagame that I am either overlooking or we have not fully explored yet?
Because as it stands this is my hypothesis of what will eventually happen to Starcraft 2 in terms of what we call the metagame:
The metagame will cycle between 'norms' and diverging definitions of 'standard' play until eventually the rate at which these cycles occur becomes so rapid that the community will be unable to fully assimilate a universal code of what is standard. Afterall, a chain is only as strong as it's weakest link and as the cycle speed increases information wont trickle down and be assimilated fast or well enough for the tiers below the professional level. And at that point, everyone will break off and the metagame will become a world of utter stylistic randomness on a complete player by player basis. Little consistency if any will exist at any level of play and it will be wars of blind counters versus blind counters until eventually it is utter randomness and completely dependant on micro and extremely small scale mechanics.
So with that being said, should I or any player even bother trying to follow and keep up with the latest 'trends' of what is considered 'standard'? Or should we prepare now for what may be an inevitable randomness anyways and become our own player, play to our own style, and swear off standard and conventional play for good? If so, does this mean we might all end up in a world of TLO lookalikes?
Any thoughts and feedback on this is appreciated and I'm interested to see what you guys think about this paradox. Again sorry if this is a bit rusty, this is my first blog here. 
 
|
You're ignoring the fact that the majority of games aren't build order wins or losses. I guess some are, it's arguable, but most games are won based on player skill (macro/micro) more than their chosen build order.
"Standard" builds are typically decent vs. almost every other build the opponent has to offer. Therefore, if the player playing "standard" plays better than his opponent, he should most of the time win. That's why it's standard, it's not risky, but it's reward isn't particularly large.
Your signature says you're gold, especially at your level, you should feel the difference in skill, as opposed to build order (no offense intended!), even at master's you can tell when you lost to someone with better macro than you.
"Non-standard" build orders might just eke out a small edge under certain conditions however, which is why build orders are looked into so much.
EDIT: To continue, if players could ever play perfect (they can't) then yes, your scenario would be the case. But Starcraft 2 is a real-time strategy game where it's physically impossible to manage the game perfectly, which is what makes it fun.
|
I definitely see what you're saying about the human error aspect of it. I've seen tangible differences in the ability of my opponents in terms of micro battles if I go with a 4 gate opening for example. As of late it does not even feel like it's the opening winning me the game, but my micro which makes the opening work.
But on the flip side of things in as close to a perfect vacuum environment that is humanly possible at the moment, pros do not often make the silly mistakes that lower tier players would. And I have seen that more and more is becoming 'acceptable' and 'standard' now as the pros explore the realms of possibilities. And that means the idea of what is 'standard' will get watered down until it reaches what I was talking about in the first place, stylistic, preference-based randomness where there is little to no consistency. I can already imagine casters saying: "Looks like Kiwikaki is going for the standard 1 gate into mothership rush opening and Idra is reacting by going for a standard triple expand mass queen into lair style for those mass overseers to endlessly contaminate and prevent the mothership," Like you would hear casters literally go nuts about motherships but now its like "...and it looks like hes going for a mothership," almost as casually as a golf sportscaster. And I think that this nature of more and more becoming commonplace is a sign that inevitable randomness of stylistic and preference-based nature may eventually reign over the metagame at all levels.
|
I think that a "standard" strategy becomes standard because it is useful vs. a variety of other strategies and that it's so reliable even if the opponent knows its coming it still works. For example, in BW Flash's build during 09-10(or 08-09) was super popular, and everyone was using it. Even if the protosses knew it was coming, they still had a hard time to deal with it. Eventually, Bisu came up with a way to beat it, but if you do it correctly it can still be very effective. The reason it is standard is because it is just so good at being good vs. anything if executed properly. Sure there are strategies which could beat it, but mostly you just had to outplay it. That is what I think a standard strategy is, a strategy that is reliable and, with proper scouting, can be useful vs. anything.
|
Standard isn't just 1 build order, but a group of them, such that there is no build that counters the group as a whole. This is why something can be well known yet unaccounted for in game--it is simply impossible to have an advantage vs. everything. (Assuming the game is balanced.) That doesn't mean the crazy builds you talk about are ever going to become standard, though. Just because the current group of standard builds is larger now than it was earlier doesn't mean the number of standard builds will continue to grow. I personally think the group will shrink, as people realize they are using builds which are simply inferior versions of other builds intended to accomplish the same thing.
The real reason you are encouraged to learn standard play, however, isn't because it'll win you the most games, but because (as was stated) it usually leads you into a playable midgame, even if you are a behind often, where your skill can lead you to victory. So even if the builds you know are no longer standard, that doesn't mean they are worthless, it just means they are suboptimal.
|
No0n:
I think that a "standard" strategy becomes standard because it is useful vs. a variety of other strategies and that it's so reliable even if the opponent knows its coming it still works. For example, in BW Flash's build during 09-10(or 08-09) was super popular, and everyone was using it. Even if the protosses knew it was coming, they still had a hard time to deal with it. Eventually, Bisu came up with a way to beat it, but if you do it correctly it can still be very effective. The reason it is standard is because it is just so good at being good vs. anything if executed properly. Sure there are strategies which could beat it, but mostly you just had to outplay it. That is what I think a standard strategy is, a strategy that is reliable and, with proper scouting, can be useful vs. anything.
Thats very true. So basically you expect rather, than a sea of randomness and infinite diversity with little consistency, you expect everyone to blob together one extremely flexible openings thats equally good and bad against every single possibility?
huameng:
Standard isn't just 1 build order, but a group of them, such that there is no build that counters the group as a whole. This is why something can be well known yet unaccounted for in game--it is simply impossible to have an advantage vs. everything. (Assuming the game is balanced.) That doesn't mean the crazy builds you talk about are ever going to become standard, though. Just because the current group of standard builds is larger now than it was earlier doesn't mean the number of standard builds will continue to grow. I personally think the group will shrink, as people realize they are using builds which are simply inferior versions of other builds intended to accomplish the same thing.
The real reason you are encouraged to learn standard play, however, isn't because it'll win you the most games, but because (as was stated) it usually leads you into a playable midgame, even if you are a behind often, where your skill can lead you to victory. So even if the builds you know are no longer standard, that doesn't mean they are worthless, it just means they are suboptimal.
So you're also learning in the same direciton? You think the end result of the metagame will eventually reach a point where each race has one super openings thats equally good and bad against every single possibility and rely completly on reaction and scouting? And that it won't actually be moreso just an ocean of random variations and blind counters?
|
It's standard BECAUSE it's known and still works. Those "counters" you're talking about usually fail when scouted because they rely on being unexpected.
Besides, I think Starcraft: BW would be the perfect empirical counter to your argument. :/ Look at 3 hatch muta ZvT. It's been the standard since savior era; it's still the standard, and it still works. Sure you can "counter" it with an ayumi-build or something, but if it's scouted, the counter can still be deflected and fail.
|
What I think is that things will remain similar as they are now. I don't think only one extremely good strategy will ever exist so that it is stable vs. everything, things will change and new standards will appear, but I do not think that it will be an ocean of randomness and blind counters OR a singularity of 1 strategy each race.
|
ghrur:
It's standard BECAUSE it's known and still works. Those "counters" you're talking about usually fail when scouted because they rely on being unexpected.
Besides, I think Starcraft: BW would be the perfect empirical counter to your argument. :/ Look at 3 hatch muta ZvT. It's been the standard since savior era; it's still the standard, and it still works. Sure you can "counter" it with an ayumi-build or something, but if it's scouted, the counter can still be deflected and fail.
So basically you're saying the same as them? When the metagame reaches it's end(?) then we will be left with one blob of a super opening thats equally good and bad against everything that is completely reactive and scouting is absolutely crucial to your survival where you could be surgically taken out by the 'extremes' that rely on secretiveness to kill you off?
|
No0n:
What I think is that things will remain similar as they are now. I don't think only one extremely good strategy will ever exist so that it is stable vs. everything, things will change and new standards will appear, but I do not think that it will be an ocean of randomness and blind counters OR a singularity of 1 strategy each race.
I see. So you think it'll hit a stage where it's inbetween those two extreme sides of what the metagame could become? Interesting.
|
On March 20 2011 16:58 FODDER~ wrote:Show nested quote +
ghrur:
It's standard BECAUSE it's known and still works. Those "counters" you're talking about usually fail when scouted because they rely on being unexpected.
Besides, I think Starcraft: BW would be the perfect empirical counter to your argument. :/ Look at 3 hatch muta ZvT. It's been the standard since savior era; it's still the standard, and it still works. Sure you can "counter" it with an ayumi-build or something, but if it's scouted, the counter can still be deflected and fail.
So basically you're saying the same as them? When the metagame reaches it's end(?) then we will be left with one blob of a super opening thats equally good and bad against everything that is completely reactive and scouting is absolutely crucial to your survival where you could be surgically taken out by the 'extremes' that rely on secretiveness to kill you off?
Well, yeah. Empirical evidence supports it. :/
|
o.O
3 hatch muta ZvT is pretty standard, but one day a progamer named Jaedong changed it and added 2 hatch muta. Another workable opening is 3 hatch lurker. Theres a lot of "standards" to choose from. I chose 3 that came up into my mind first. I don't know... but it seems that there is a balance...
|
ghrur:
Well, yeah. Empirical evidence supports it. :/
That's true. The predecessor is a great way to try and predict how Starcraft 2 could turn out. But the past is not always a 100% accurate inclination of the future. There's always somewhat of am an ambiguous ambience to it. I guess I'll be looking forward to seeing what we decide is the super blob openings thats good versus everything equally when we reach that point. I wonder how far away from reaching that point we are (if thats even what will happen of course).
|
ideally, things should end up similar to brood war where there's one or two very standard builds per matchup, and a bunch of other builds that are intended to take your oponent by surprise and throw him off his game.
In bw maybe the only exception to this is ZvZ... since its somewhat rock-paper-scissors (though you could argue that overpool is viable for all situations)
I think "perfect" metagame will either end in one standard build, or a rock-paper-scissors match which is what will happen if there is no way to scout and react in time to variations in strategy. (for example, spawn positions are too close, or a build can effectively deny any scouting at all). Certain maps can also have their own standard builds.
Still, sc2 is still a long way from being "solved" with 2 expansions in the making and the inevitable plethora of balance patches to come.
|
Gak2:
ideally, things should end up similar to brood war where there's one or two very standard builds per matchup, and a bunch of other builds that are intended to take your oponent by surprise and throw him off his game.
In bw maybe the only exception to this is ZvZ... since its somewhat rock-paper-scissors (though you could argue that overpool is viable for all situations)
I think "perfect" metagame will either end in one standard build, or a rock-paper-scissors match which is what will happen if there is no way to scout and react in time to variations in strategy. (for example, spawn positions are too close, or a build can effectively deny any scouting at all). Certain maps can also have their own standard builds.
Still, sc2 is still a long way from being "solved" with 2 expansions in the making and the inevitable plethora of balance patches to come.
So gradually as time goes by the priority of scouting will only increase even more I can only imagine if Starcraft 2 does go down that road. How do you forsee players then maintaining map control and vision in the future sufficient for this level of play by that point? Could you even forsee players making things such as auxiliery robos simply for constant observer production in the future or other extremes to maintain maximum servailance and vision?
|
i find it interesting that noone spoke about "standard" play in bw. It was always strategy1 > strategy2, or different openings and what not, but no strategy was defined as "standard play". Mabye a bunch of different openings that was considered safe togheter created "standard play", but not one single strategy.
|
i think the reason its useful to give bad/simple advice to really really low level players is you dont want to "blow their mind"
if you tell a person to be creative when they've never made more than 5 workers in their life, how are they gonna know where to start? keep things as simple as possible while people improve their basic mechanics. like day9 says, if your not in diamond you will win if you just make more shit than the other guy, and that means better mechanics.
more on your point of "playing standard" i think you have to look at it differently. idra plays the most standard game in the world. but most players atleast on the NA ladder have so much worse mechanics than him, even knowing EXACTLY what hes gonna do they cant stop it, even at the 4k masters level "having more shit" is the best tactic. now obviously if he goes mass air and you cant shoot up, doesnt matter how much stuff you have, but thats what playing standard is all about.
muta ling bane, marine tank, these "standards" do atleast ok, against 'everything' out there. theres so much room for style, and mechanical skill, even with the standards of play im sure idras muta ling bane is alot more scary than mine
|
turdburgler:
i think the reason its useful to give bad/simple advice to really really low level players is you dont want to "blow their mind"
if you tell a person to be creative when they've never made more than 5 workers in their life, how are they gonna know where to start? keep things as simple as possible while people improve their basic mechanics. like day9 says, if your not in diamond you will win if you just make more shit than the other guy, and that means better mechanics.
more on your point of "playing standard" i think you have to look at it differently. idra plays the most standard game in the world. but most players atleast on the NA ladder have so much worse mechanics than him, even knowing EXACTLY what hes gonna do they cant stop it, even at the 4k masters level "having more shit" is the best tactic. now obviously if he goes mass air and you cant shoot up, doesnt matter how much stuff you have, but thats what playing standard is all about.
muta ling bane, marine tank, these "standards" do atleast ok, against 'everything' out there. theres so much room for style, and mechanical skill, even with the standards of play im sure idras muta ling bane is alot more scary than mine
So what you see is us our narrowing our idea of standard play to a limited amount of relatively safe openings versus everything. That aren't paticularly stong in one way, but more so a super reactive blob of an opening that heavily relies on scouting. But at the same time you say you believe that variations will exist due to stylistic choices, and actual skill level. So wouldn't that mean that (with such a large variations of skills and styles out there) that it would break down into inconceivably random variations anyways? Where Idra would approach one thing one way, and Fruitseller another. And eventually wont it be so wattered down you can't even distinguish at first glance that they are going for the same openings (in their minds) since they will by that point have different opinions and beliefs of what works best (both in their eyes overall, and for them personally) towards the execution of a given opening?
|
im saying that if you have a solid plan, where you cant die in the first 10 minutes you will have time to improve. if you arent in master league, forget everything you think you know about reactive play and scouting, it just doesnt matter.
pick your build, for this example, im gonna go with marine marauder. practice the hell out of it, make the most marines and marauders you can. then once you have the first 5-10 minutes of your game mapped out, start applying earlier pressure, practice keeping your macro on top of things while still attacking. once you can do this, you can start examining replays and say; if i attack now, im really far ahead of him, i shouldnt of waited, or conversely i need to wait for my x to be done before i push.
at this point you can easily be in platinum with 0 game knowledge and only a very basic idea of timing. now you can start thinking about scouting. using our terran example, you need to either poke their front or scan, and factor in losing that scv/marine/mule into your build order. now you can finally start adding reactions to your play.
if youve been staying mm till 10 mins so far, then finally adding tanks and w/e else, now you can start looking to react, banelings? bunkers? get the tanks faster. and the cycle of improving your macro while having this reaction available continues.
this is how a 'pro' improves, how many protoss 'pros' were '4gaters' till x time ago, but finally started playing a legit game. thats because theres SO much to learn before you even look at reactions and stuff. huk for one was known as a 4gater for a long time, and now he has over9000 fans.
the overall point is, if you know you arent good, dont get bogged down in y counters x, if i see this i need to do this. dont worry about 10 variations on your opener, just improve a very basic build, and then when you are comfortable slowly add in tiny variations.
|
turdburgler:
im saying that if you have a solid plan, where you cant die in the first 10 minutes you will have time to improve. if you arent in master league, forget everything you think you know about reactive play and scouting, it just doesnt matter.
pick your build, for this example, im gonna go with marine marauder. practice the hell out of it, make the most marines and marauders you can. then once you have the first 5-10 minutes of your game mapped out, start applying earlier pressure, practice keeping your macro on top of things while still attacking. once you can do this, you can start examining replays and say; if i attack now, im really far ahead of him, i shouldnt of waited, or conversely i need to wait for my x to be done before i push.
at this point you can easily be in platinum with 0 game knowledge and only a very basic idea of timing. now you can start thinking about scouting. using our terran example, you need to either poke their front or scan, and factor in losing that scv/marine/mule into your build order. now you can finally start adding reactions to your play.
if youve been staying mm till 10 mins so far, then finally adding tanks and w/e else, now you can start looking to react, banelings? bunkers? get the tanks faster. and the cycle of improving your macro while having this reaction available continues.
this is how a 'pro' improves, how many protoss 'pros' were '4gaters' till x time ago, but finally started playing a legit game. thats because theres SO much to learn before you even look at reactions and stuff. huk for one was known as a 4gater for a long time, and now he has over9000 fans.
the overall point is, if you know you arent good, dont get bogged down in y counters x, if i see this i need to do this. dont worry about 10 variations on your opener, just improve a very basic build, and then when you are comfortable slowly add in tiny variations.
So would you suggest something along the lines for any player below the Diamond league to just open with a 4 gate opening every single game, on every single map, and in every single matchup until it's literally embedded into your brain? If so, how could someone ever walk away from all they know? Wouldn't they be too used to 4 gate every game to branch out thereafter? I mean I can't imagine (what some would call cheese) having many transferable applications to the 'standard openings' of this moment if they tried to switch over after a long time of only practiceing a single gimmick all in opening for monthes on end. Because I've been experimenting with so many variations of 4 gate and it seems no matter what, you literally almost always have to do substantial dammage on the borderline of outright winning as a bare minimum, or you just are too far behind and lose an uphill fight from that point on.
|
you think building a robo bay is harder to do than learning to macro?
im not saying embed 4gate into your mind so you can never do anything else, im saying use 4gate, because its simple, for as long as it takes for you to master all the other aspects of the game, before you care about reacting and scouting
|
On March 20 2011 16:51 FODDER~ wrote:
So you're also learning in the same direciton? You think the end result of the metagame will eventually reach a point where each race has one super openings thats equally good and bad against every single possibility and rely completly on reaction and scouting? And that it won't actually be moreso just an ocean of random variations and blind counters?
What, no, how did you get that from my post? I think a particular race's strategy on a particular map vs. a particular race will converge to a group of builds, which is impossible to do better than 50% against. The individual builds aren't 50% vs each opposing build, but the group (with weighting, e.g. 80% 2rax and 20% 14cc) together does 50% or better vs every possible opposing group.
|
turdburgler:
im not saying embed 4gate into your mind so you can never do anything else, im saying use 4gate, because its simple, for as long as it takes for you to master all the other aspects of the game, before you care about reacting and scouting
And how do you know you've mastered the other aspects? Or you've been on a long lucky streak, or everyone has been off the entire time? Those are the kickers... there is so much room for descrepency and misread results...
huameng:
What, no, how did you get that from my post? I think a particular race's strategy on a particular map vs. a particular race will converge to a group of builds, which is impossible to do better than 50% against. The individual builds aren't 50% vs each opposing build, but the group (with weighting, e.g. 80% 2rax and 20% 14cc) together does 50% or better vs every possible opposing group.
Oh I gotcha'. Accidentally conjoined your comments with another and got your opinions mixed up. Good insight though.
|
Belgium6774 Posts
You're abstracting too much. The argument works on a theoretical level, but practically there's just so many other factors in this game to account for that you can't merely talk about buildorders
|
Id say the strangest paradox are my parents. They both work at the same hospital.
|
I always viewed "standard" as a simpler way of saying "least fucked in the worst-case scenario".
In BW, I would always 9 hatch, in every matchup. This is because I had roughly the economy of a 9 pool, with the lings coming out at about the same time as a 12 pool, speed/lair coming at around the same time as a 12 pool speed/lair, and a 2nd hatch + earlier creep than a 12 hatch.
There was no situation where I was utterly fucked beyond belief. There were numerous timing windows available that my opponents could abuse, regardless of the opening they chose, however, since I knew what those timing windows were, I also had to learn how to deal with it.
This allowed me to get into the midgame in every game on close to even footing. From that point, I tried to play as if I was better than my opponent, and make up for that disadvantage I put myself in.
This was solely for when I was laddering. I still practiced more common openings.
Because of this, when I was in a specific match, I could tailor my openings to face what I expected from my opponent, and since I was used to fighting off worse abuses of timing windows that I left myself open to on a regular basis, it made it easier to deal with them in games that actually counted (I didn't, and still don't, take laddering seriously), since I did not leave as big of windows open in those games.
|
For such a silly idea that is so very wrong, you sure are good at articulating yourself.
|
The OP is completely right in one sense. It is all conveniently explained in another recent blog here.
The link goes to a blog with scans of the notes from the Starcraft professor from Berkeley (?) at a special lecture in Chicago. He explains something that progamers and veteran starcraft enthusiasts already know but never phrased into words.
Just read the notes and you'll be vindicated. Also, "standard" in colloquial SC2 speak just means any one of the builds that are accepted to be solid and safe in the current metagame.
|
Calgary25990 Posts
On March 23 2011 23:50 tyCe wrote:The OP is completely right in one sense. It is all conveniently explained in another recent blog here. The link goes to a blog with scans of the notes from the Starcraft professor from Berkeley (?) at a special lecture in Chicago. He explains something that progamers and veteran starcraft enthusiasts already know but never phrased into words. Just read the notes and you'll be vindicated. Also, "standard" in colloquial SC2 speak just means any one of the builds that are accepted to be solid and safe in the current metagame.
He's not a professor lol. He was a second year student at that time.
|
intrigue
Washington, D.C9934 Posts
oh dear. you are way overthinking this. the best way to resolve your crisis is to go play a ton of games on ladder, and then laugh at this thread in year.
|
intrigue:
oh dear. you are way overthinking this. the best way to resolve your crisis is to go play a ton of games on ladder, and then laugh at this thread in year.
I already do play a lot.
Though you are right, it's always good to have a look past to the past every once and a while and laugh at how oblivious you used to be about things earlier on. But I enjoy theorizing and philosophy so I don't see what's so bad about analyzing metagame a little bit more than most people bother to. 
|
|
|
|
|
|
EPT
