|
Copying builds from progamers is a good way to focus on practicing mechanics only, seeing as you're doing a preset list of things that you already understand has a high level of efficiency.
What do I mean by this? When a progamer does a build in a tournament, you can assume that they've done a large amount of practice concerning making this build efficient, testing it against various strategies, and making sure that the plan it goes into in the mid and late game makes sense.
At this point, when you copy it, you don't have to worry that what you're doing solves these problems because someone who spends a lot more time than you playing this game has already answered these questions, and you can focus on your execution and mechanics, which is the most important part of improvement.
A "standard" build, at least in my opinion, is one that satisfies the previous three requirements: it has to be efficient in achieving its goal, it has to work against the various extremes your opponent can do, and it has to have a mid and lategame plan that makes sense and is an easy transition.
It makes the most sense to copy these standard builds because they excel in the most general of cases, which you will be facing 90% of the time.
How, then, do we look at a build to see if it's "standard" or not?
A pretty easy test is to see if a progamer does this build against multiple opponents on different maps with little change. This implies he's not metagaming the matchup he's in or map he's in. This isn't useful when dealing with a smaller number of games, but given the opportunity this is the easiest way to answer this question.
An example of a build I would copy (and actually did, I do this build in 90% of my TvPs), is from this game between Kawaiirice and WeekendfOu on Antiga Shipyard. I don't know if Kawaii came up with this build, but this is the rep I stole it from essentially.
The build (if you don't feel like watching the rep):
+ Show Spoiler + 1 rax fe
bunker at natural
2 more barracks
double gas
tech lab on one barracks -> stim
ebay -> +1 weps
factory
third gas
reactor and tech lab on barracks -> combat on second tech lab
reactor on factory when it finishes + starport
double medivac once starport is done
then move out and take a third behind
Let's look at why this is standard (and thus copyable):
Is it efficient?
If you watch the replay, you'll notice that all of the essential upgrades (combat, stim, +1 weps) and medivacs are all done at around the same time (though stim is earlier). This is the mark of an efficient build done to exploit a timing.
Is it solid against extremes?
This replay is against a fairly standard robo opener after a quick expand, but protoss has a lot of options! Things like 4gates, 6gates, 3 gate robo allins, DT openers...
The only way to prove this is to either see it done or do it yourself against these openers, or theorycraft.
4gate is easy, have a bunker and pull, make more bunkers once you scout it. Same goes for 3 gate robo.
6 gate and related allins require a bunch of bunkers, but you have +1 and stim (and sometimes medivacs) so it's doable.
You have a relatively early ebay so turrets are doable if you scout double gas on one base and want to protect yourself against DTs.
We saw how it does against robo openers, and templar openers require more time to tech to be efficient so the pressure will do even better.
Seems fine to me.
Does it have a good transition to a mid-lategame plan?
You take a third behind pressure, and can add whatever tech you deem necessary by the scouting your push does (i.e. what units you run into).
So this build seems "standard" by the definition I gave earlier, which is by no means exhaustive.
The only other concern for copying is difficulty of execution because, after all, we don't have the mechanics that progamers do.
However, a lot of people put way too much emphasis on this fact in explanations of why people shouldn't copy these builds.
It's true that you won't be able to execute as well, but the fact that a build requires a sizable amount of mechanics means that you'll be put in situations where you have to improve mechanics to win. Think of it like resistance training for your mechanics.
Moreover, a good standard build will put you in mid-lategame situations more often (think of the third constraint), and it's fairly common knowledge that the later a game goes the more mechanically is required of a player. Again the "resistance training" factor.
+ Show Spoiler +I wrote this blog because I keep seeing people say that people shouldn't copy progamers's builds, and that's simply false. I attempted to explain why here as well as give a framework for determining which builds are good to copy.
 
|
I'd say some of this info would be a bit of common knowledge, but still, 5/5.
Even though not all styles and builds are applicable to all players, people in any league below masters are probably better off copying builds done by the pros. Didn't we all do the same in BW?
|
Problem is that there's WAY more thought in Pro builds than a random gamer will understand. Sure, 1rax fe is standard, getting more raxes, awesome. The general gameplan sounds good and is of course based on the standard flow any decent Terran should know. What makes it more difficult to copy is all the minor thing the Pro is doing because of things he know/scouts which you as a random Joe doesn't. Even if you see him do a similar play in several games, you will definitely see minor variations... but how minor are they to the pro? Do they make or break the build?
This make pro builds hard to copy, the builds are optimized to be as effective as possible for what they are intended to do, but just because you watch the games doesn't mean you know what this intention is. You can pick up the timings, but not always what makes those timings viable and what forces transitions. There's also the huge factor that pros build their gameplans with their mechanics in mind, their timing might work because they have just enough minerals and gas to pull it off, while someone with lesser mechanics might be lagging behind and screwing it up (for a simple example, check out Tangs aggressive builds which usually put a strain on mechanics to get the optimal timing).
I'd say people should stay away from copying pros if they are less than master or at least high diamond. Stick to guides which let you know about the important factors of the matchup, then learn a standard build. Play it enough, improve it to your liking, THEN look at pros using something similar and integrate what you see (for example, I use 2base muta opening vT like 90% of other zergs, but I've started to mess around with late gas early four queens after I saw Sheth use it to completely deny hellions).
|
It probably does depend on the person. One approach involves taking a "pro build" and then keep trying to execute it and hammer away for quite a while, another involves just doing something with knowledge of the general features of the matchup and a safe build that doesn't require the most crisp execution, taking things from the pros when something seems to be consistently going wrong to see how they fix it.
The one thing though is that if someone keeps trying to execute a build taken from a pro, they will learn at least a good amount of the logic by experience, even if it takes a while (sometimes a long while). However, this also means that in taking builds from pros, multiple games should generally be taken to identify common aspects of the build, otherwise a move specifically done in fear of an all-in may just be taken as standard when it actually just sucks against everything else (a little extreme there, this normally applies to smaller deviations)
|
On February 28 2012 07:36 Tobberoth wrote:
Problem is that there's WAY more thought in Pro builds than a random gamer will understand. Sure, 1rax fe is standard, getting more raxes, awesome. The general gameplan sounds good and is of course based on the standard flow any decent Terran should know. What makes it more difficult to copy is all the minor thing the Pro is doing because of things he know/scouts which you as a random Joe doesn't. Even if you see him do a similar play in several games, you will definitely see minor variations... but how minor are they to the pro? Do they make or break the build?
This make pro builds hard to copy, the builds are optimized to be as effective as possible for what they are intended to do, but just because you watch the games doesn't mean you know what this intention is. You can pick up the timings, but not always what makes those timings viable and what forces transitions. There's also the huge factor that pros build their gameplans with their mechanics in mind, their timing might work because they have just enough minerals and gas to pull it off, while someone with lesser mechanics might be lagging behind and screwing it up (for a simple example, check out Tangs aggressive builds which usually put a strain on mechanics to get the optimal timing).
I'd say people should stay away from copying pros if they are less than master or at least high diamond. Stick to guides which let you know about the important factors of the matchup, then learn a standard build. Play it enough, improve it to your liking, THEN look at pros using something similar and integrate what you see (for example, I use 2base muta opening vT like 90% of other zergs, but I've started to mess around with late gas early four queens after I saw Sheth use it to completely deny hellions).
I sort of solve for this by saying that a good copyable build should have a mid-lategame plan in mind, and in order to copy it (using what I talk about) it at least requires some knowledge of what the plan is.
The build done in the game I link isn't that difficult to comprehend: he gets a good amount of tech up at similar timings and moves out. There's a lot of nuance a lower league player wouldn't get as to why that works (things like how the weapon upgrade synergizes with the large marine count to create ludicrous DPS, how the timing is usually just before a large amount of AoE is out in a lot of cases, etc.), but you don't need to know this nuance in order to copy it.
The reason lower level players may want to copy pro builds I sort of mention, it removes a lot of the decision making calculus so you can focus purely on mechanics without worrying about deciding what to do. In short, what you need to do is already set out so you can focus on execution (which is definitionally mechanics).
|
This looks like a great build. Thanks for the replay.
|
|
|
On February 28 2012 07:36 Tobberoth wrote:
Problem is that there's WAY more thought in Pro builds than a random gamer will understand. Sure, 1rax fe is standard, getting more raxes, awesome. The general gameplan sounds good and is of course based on the standard flow any decent Terran should know. What makes it more difficult to copy is all the minor thing the Pro is doing because of things he know/scouts which you as a random Joe doesn't. Even if you see him do a similar play in several games, you will definitely see minor variations... but how minor are they to the pro? Do they make or break the build?
This make pro builds hard to copy, the builds are optimized to be as effective as possible for what they are intended to do, but just because you watch the games doesn't mean you know what this intention is. You can pick up the timings, but not always what makes those timings viable and what forces transitions. There's also the huge factor that pros build their gameplans with their mechanics in mind, their timing might work because they have just enough minerals and gas to pull it off, while someone with lesser mechanics might be lagging behind and screwing it up (for a simple example, check out Tangs aggressive builds which usually put a strain on mechanics to get the optimal timing).
I'd say people should stay away from copying pros if they are less than master or at least high diamond. Stick to guides which let you know about the important factors of the matchup, then learn a standard build. Play it enough, improve it to your liking, THEN look at pros using something similar and integrate what you see (for example, I use 2base muta opening vT like 90% of other zergs, but I've started to mess around with late gas early four queens after I saw Sheth use it to completely deny hellions).
While it is indeed true that copying a build without any idea of the mid-late game plan and possible transitios is quite silly, but that can be solved by just watching enough replays. As the OP stated (and I agree), for a build to be standard, it has to be solid, be used against several opponents on several maps and been done and tested well enough for the player to have enough reps to see it in action and see what do the pros react to.
Once you have a general idea of what does the build do and how players react to the enemy, you can start working on it yourself, see what are you having problems with, look for a rep to see a good way to react to it, keep working on it, etc.
|
On February 28 2012 06:50 Sc1pio wrote:
A pretty easy test is to see if a progamer does this build against multiple opponents on different maps with little change. This implies he's not metagaming the matchup he's in or map he's in. This isn't useful when dealing with a smaller number of games, but given the opportunity this is the easiest way to answer this question.
I think you meant to use the word "mindgame" instead of "metagame", common misconception.
|
|
|
|
|
|
EPT
