|
I really enjoyed watching today's Code S group. "But there was so much cheese/all-in!" you might say. What others might see as cheesy though, I see as brilliantly planned, strategic play. No, this isn't me using "strategic play" as a euphemism for cheese as someone else once did. These games were actually well-planned out by the winner. Let me explain.
Supernova vs MC Game 3
MC didn't get detection and lost to cloaked banshees. Terrible game, Supernova cheesed and got lucky, right? Well, I don't think so. At least, it's not that simple. To understand Supernova's build, consider two key pieces of scouting information he showed MC:- MC's scouting probe was blocked by a barracks, 2 supply depot wall. Players savvy in the PvT matchup may know that if you scout the opponent on your first try (which is guaranteed on GSL Antiga Shipyard because of forced cross positions), your probe arrives at the Terran's base when a standard gas opening doesn't have enough minerals to build the second supply depot. So if you see that second depot, you know that Terran didn't build a refinery yet.
- He built a bunker on the low ground, on top of the ramp leading to his natural. MC saw this with his zealot/stalker poke, thus "confirming" that Supernova expanded.
What build does this look like? Both pieces of information suggest the Terran is going for the most popular, standard macro opening in the current metagame: 1 rax FE with no gas. If I were in MC's position, I would've made the same read. Any Protoss player who didn't find Supernova's smurf and look at his match history would've made the same read.
![[image loading]](http://i.imgur.com/C5Uio.png)
How can this not be a no gas fast expand?
Once Supernova gave away these two pieces of information to fake the gasless FE, he correctly predicted MC's response of 1 gate FE into 4 gates with no robo. How did Supernova know this was coming? I believe he made his prediction based on the following logic:
- During Homestory Cup 4, this was MC's go to build against Terran's gasless FE. It is how he demolished Poyo and Cloud when they both attempted to play macro games against him. Supernova could've watched HSC4 and learned MC's tendency to play this way, but he probably also knew that MC likes to play this way, being on the same team.
- When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo. Gasless FE is usually followed up by either medivac/stim tech(the standard TvP build) or by a fast 3rd CC, neither of which a robo is immediately useful against. A robo with observers kills Protoss' ability to pressure before Terran's stimpack is done researching, and also delays Protoss' 3rd base unnecessarily. At best the 3 gate robo followup comes out even against standard medivac play, and is terrible against fast 3rd CC. 4 gate, on the other hand, puts immense pressure on Terran between 8 and 9 minutes. Terran is especially vulnerable during this period because the standard medivac build has neither stim nor medivacs at this time, and fast 3rd CC delays those even further. This combined with the fact that Terran fears a potential 6 gate all-in means that he must make more bunkers and pull many SCVs off of mining to get ready to repair. Until Terran has stim and medivacs (or ghosts if he went for that route), it is not safe to sell the bunkers and send SCVs back to mining. Once the 9 minute mark is reached and Terran has stim, Protoss simply retreats his units and takes a 3rd base with the resources saved from not getting any tech or upgrades. All in all, the Terran is in a worse position economically than he would be if Protoss didn't pressure.
When all is said and done, sure enough, MC fell for the bait and went for his tried and true 4 gate pressure. Supernova, knowing MC's play style and the fact that the current metagame favors 4 gate against Terran's gasless FE, mindgamed MC by going for cloaked banshees disguised as a gasless FE. With cloak vs no detection, it was an easy win. To win with a cheese against unsafe play is one thing, and happens all the time on ladder. But to win with a cheese after tricking your opponent into thinking he's safe? Now that's something special.
ForGG vs MC Game 1
It looks like your average, run of the mill macro game where one side wins after a good engagement. Having just lost to Supernova, MC learned his lesson: no matter what reads you make from scouting, unless you actually see the expansion, Terran can always add gases and switch to cloaked banshees! In this game ForGG showed MC a gasless opening, but not the CC itself. MC, having learned from his mistake the previous game, went for 3 gate robo instead of 4 gate to follow his 1 gate FE, just in case ForGG would pull the same trick Supernova did. ForGG's build? Fast 3rd CC before gas. If gasless FE into medivacs is the soft counter to 1 gate FE into 3 gate robo, then gasless 3rd CC is the hard counter. ForGG took such a big lead just from the choices of build orders that he lost 2 full medivacs for nothing, and still rolled over MC with his massive supply advantage. Did ForGG intentionally manipulate MC into going 3 gate robo, knowing how 4 gate turned out against Supernova? Maybe. Maybe not. But I'd give a former MSL champion the benefit of the doubt.
![[image loading]](http://i.imgur.com/eH5AR.png)
Game timer 13:30. This is what happens when you try to play robo vs fast 3rd CC
MC vs Leenock Game 3
At a first glance, MC's build doesn't make any sense. Why do a 7 gate all-in with 12 sentries at the last second, when you could make 8 sentries much earlier and still have just as many force fields? In fact, why do such an outdated build as 7 gate anyway? I believe MC's success resulted from a clever trick taking advantage of how unpopular 7 gate is nowadays. Consider the following 3 builds that are similar to each other, all popular in the current metagame:- 4 gate +1 zealot pressure. This build exploits the trend of Zergs expanding twice without gas to skip any sentries or additional cannons beyond the first and sending 8 zealots with +1 weapons to the Zerg's 3rd base at 8 minutes. This pressure hits at an awkward timing for Zerg, forcing them to make many lings and produce roaches earlier than they would like. Meanwhile at home, Protoss has a large gas build-up that eases the transition into either a macro game(see Hero) or a 2 base all-in timing (see Naniwa).
- 4 gate +1 zealot pressure with void ray support. This is just like the last build, but with 7 zealots and a void ray instead of 8 zealots, at almost the same timing. Upgraded zealots kill spores, queens, and lings while void rays kill roaches, so this build is much more likely to kill the Zerg's 3rd, but worse at transitioning out. Whereas the 8 zealot version punishes late roaches, this one is good against early roaches and punishes late creep spread.
- 4 gate +1 zealot pressure with DTs. Same idea again, except with DTs instead of void rays. You kill spore crawlers with the zealots and focus down the 3rd with DTs to follow up.
What do all of these builds have in common? They all pressure with +1 zealots at 8 minutes, all require Zerg to make roaches to survive, and the latter 2 require Zerg to make spores. Oh, and they're all way more popular, and considered better builds, than the 7 gate. Meanwhile, 7 gate tends to build up a high sentry count early for more force field energy, hit later, and also come with an armor upgrade. Now if you're Leenock and you scout the following with an overlord and a ling at the watchtower, what build do you think MC's doing?
![[image loading]](http://i.imgur.com/gt3Oi.png)
4 gates
![[image loading]](http://i.imgur.com/sdMLf.png)
+1 zealots
Yeah, that's a no brainer. It's some form of the popular 4 gate +1 zealot pressure, and if Leenock watches a lot of MC's games, he knows the DT variant is MC's favorite one. Especially in foreign tournaments, MC loves the DT followup. Maybe the timing was off, but remember that MC blocked Leenock's hatchery with a pylon, which throws off all timings and I doubt Leenock could've calculated the timing difference on the spot. Also, jetlag and tilt from losing to a baneling bust could have caused MC to mess up his execution. Now consider how Zerg is supposed to play against these builds:- Against all 4 gate +1 variations, Zerg must make an early roach warren and start making pure units at around 50-55 drones.
- Against the void ray and DT variations, Zerg must make spore crawlers.
- Against a 7 gate all-in, Zerg has more time to build drones and wants to have burrow for his roaches. Hydras also help.
Guess what Leenock had when the 7 gate hit? 55 drones, a spore crawler, enough roach/ling to hold off 4 gate +1 pressure, and no burrow or hydra den. By building 3 gateways simultaneously in the same area scouted by an overlord, showing only zealots, and delaying sentry warp-ins until the last second, MC fooled Leenock into thinking he was doing his 4 gate +1 zealot pressure into DT build. Leenock prepared for the DT build, but the defense against that is inappropriate for holding off the 7 gate all-in that MC was actually doing. Some of you may have noticed that MC did the same trick against Nerchio in HSC4. Once again a player created an illusion in the mind of his opponent, and won the mindgame.
Three oGs players today were able to take advantage of trends in the metagame and knowledge of players' past tendencies to disguise their builds, inducing their opponents into sub-optimal responses to their real strategy. These games, despite looking like simple, uninteresting, or even "cheesy" games to the untrained eye, were actually an incredible display of the strategic depth found at the Code S level. If you have read all the way here, I hope you now have a deeper appreciation of these games than you did before.
 
|
Great read! Didn't even watch yesterday's games but it sure does give a whole lot more explanation than the recommended game polls and the LR thread lol.
|
|
|
United Kingdom38153 Posts
Really great read Iamke!
You're defintely on the money with the Nova vs MC analysis, from their interviews today:
Q: In your game against oGsMC, you used Banshees in every match.
A: That is how I established my concept. I pressed on in that sense because oGsMC uses a build that often excludes accounting for concealed banshees. I think it went according the planned scenario.
Particularly loved the insight into MC vs Leenock G3 too since not playing either race much often means I just completely overlook the subtleties of various build order decisions.
|
The second depot is actually very very late in a normal gasless 1 rax FE. As far as I understand, you want to plot down the second CC well before your second depot.
|
Someone with reasoning. Incredible.
|
Thanks for this! Helped explain some stuff I did not understand.
|
|
|
I thought that the form of 7g that MC did was relatively popular as a followup to +1 zealot, Artosis might have made that popular by writing the b/o on his blog.
|
Baller thread. Now I'm def gonna go watch these.
|
Definitely gained a much deeper appreciation for the builds and the wins now!
|
United States10094 Posts
amazing read. great analysis.
|
iamke55, you are a gentleman and a scholar. We need more in-depth analysis like this!
|
This is a great writeup and great to see some high level analysis! Just to comment on the first 2 games above - I'm a bit unconvinced though...
Although HSC is a decent tournament, Poyo and Cloud are hardly top-tier terrans. I would be interested to see 1-gate FE into 4-gate pressure vs 1-rax FE against decent terrans. If I recall correctly, MC did this 4-gate pressure against MMA in the GSL Nov Code S ro32 group stages and got badly trashed. Basically, what I'm trying to say is that 4-gate pressure is not that great vs 1-gate FE anyways and MC was better off going for 3-gate robo. However, MC could be trying to exploit the fact that Supernova likes to go 1-rax FE into double gas, and hence the 4-gate pressure should be able to outright kill that.
I'm not convinced that 3-gate robo fares badly against 3 CC before gas. Won't a quick 2-base colossus fare well or outright win against it? 3 CC before gas means that the terran will be marine heavy and colossus will do well against it regardless of whether the protoss does a timing push or not. At worse, the protoss can get that 3rd base up before the terran feels safe to land the 3rd CC anyways. Perhaps forGG exploits the fact that MC likes to go gateway heavy?
I'm not that good in ZvP so I was puzzled about leenock's defensive strategy. I thought he scouted the 7-gates but I wasn't paying too much attention. Thus, I was surprised that he built only 1 spinecrawler initially at the 3rd when I thought he should've built much much more. The analysis clears alot of things up.
|
Really nice analysis of those games, good read.
|
Agreed, thanks for putting things in perspective. The casters more or less shrugged it off as one player failing or the other, but they didn't explain the tricks quite as well. We're about to reach another level of starcraft :D
|
Nice analysis ! Good job !
|
This is a brilliant post. All of the micro/cheese builds that came out today were well-planned or the result of smart scouting. Code S is really upping the ante on the wrinkles to strategic play.
|
Great article, thank you.
|
Wow! This is amazing. I'll never look at "random" cheese the same way ever again!
|
I disagreed with the game ratings TL gave for alot of these reasons. Nice writeup.
|
Amazing analysis. I would never be able to see this if I were to watch the games live. In fact, I think most casters will never comment on this when the games are streaming live. Hope to see more of these in the future (hopefully with one example where a zerg player wins ^^)!
|
On January 11 2012 09:26 chenchen wrote:
The second depot is actually very very late in a normal gasless 1 rax FE. As far as I understand, you want to plot down the second CC well before your second depot.
That version is more economical, however gives away what you are doing entirely and is therefore possibly inferior. Fast second depot is at the very least not rare.
|
Beautiful analysis of these games. Looking at the "Overall Recommended" polls for these games I was pretty disappointed. Seems like people only want to see macro games, refusing to acknowledge the amount of intelligence and skill it takes to fake a build while executing something that directly counters the response to the fake build. I love the way Supernova played, showing that he plays to win, not to entertain.
|
wow thanks for this, i even more enjoy the game now
|
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience.
|
amazing read. that is a great analysis of the metagame!
|
Amazing post. Groups A and B have been some of the best SC2 I've watched in a long time, this season is going to be epic!
|
Nice write-up. I slightly disagree with the ZvP analysis though - I think MC punished Leenock for going fast lair. You talk about 7 gate, and I'm sure you are talking about the later, 4 gas sentry heavy 7 gate rather than the 7 gate all-in with cut probes and no upgrades (similar to 6 gate +1 which cuts probes, and has no gas at natural, only 2 gas in total). But basically it's the same idea - Leenock saw just the 3 extra gateways being added (even made spores because he obviously sensed tech of some sort), so 'knew' it wasn't a gateway all-in but some sort of 4 gate pressure with tech or macro behind it. He went lair because he thought he was safe, and he wasn't. Can't make that lair too early or you die to gateway pressure. Against 4 gate pressure, sure you can get lair, and you don't need 100% unit production after 50-55 drones, just ~70% unit production (against 6 gate+1 or 7 gate with no gas is when you go 100% unit production). Basically, 6 gate +1 or 7 gate no gas all-ins kill you if you go lair, and 4 gate pressure or tech openings (SG/DT) it's safe to go lair.
Anyways, I was surprised how much you wrote about builds in TvP. You make this game look like BW, with how people talk about builds in that match-up. I don't feel ZvX is anywhere near as complex as you seemed to make out TvP, maybe it's just Zerg doesn't need build orders as much (maybe just about third timing and "make X% units to deal with such push"), but I definitely thought that was cool.
good job.
|
|
|
Aotearoa39261 Posts
On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience.
I guess poker is just coinflips to you too.
|
On January 11 2012 14:01 Plexa wrote:Show nested quote +On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience. I guess poker is just coinflips to you too.
I guess you can't deny that "the current trend is x, so I will assume my opponent will be preparing for x so I will do Y instead and hopefully win" is strategy of a sort, but I remain highly skeptical of the OP's claim that this is demonstrative of any sort of strategical brilliance or that spectators should be enthralled by the intelligence of these player's decisions. Isn't the fact that you can sum up each of these games in one sentence each (leenock thought MC was X because of current metagame trends but instead he Y, GG) indicative of something lacking? Compared to macro games, where the decisions would take pages and pages to explicate, there is something clearly unsatisfying about matches decided solely by BO mindgames.
|
|
|
Made me wanna get a subscription from GOM tv to watch all the code s madness! Really good read though definitely worth my time!
|
Really great post. Imagine how great it would be to have casters explaining this at the time... (Some day, I hope, when pros have retired and so forth, we'll have such casters. No one casting now can do it with any reliability.)
|
On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience.
These games were coinflips because the players chose them to be. For instance, if MC had gone 1-gate FE into 3-gate robo against supernova, he should've easily won the game.
|
This is brilliant. This is the sort of finesse that made BW such a fantastic game. I actually didn't notice it, and assumed that SuperNoVa was a cheesy run-of-the-mill Terran and that most of Group B games weren't that great.
Fantastic.
|
Well written! I was cognizant of the mind games at work with Supernova and MC, but not the ForGG v MC and MC v Leenock. Very interesting to read, thanks for the writeup <3
|
You explained everything very well...and is absolutely horrifying for the future of sc2. This shows the vest worst side of sc2. There is no way to just play safe and reactive at the highest level. Builds counter builds in a giant cycle of mindfuckery, and the game becomes more a game of tricking one another (which is cool to a point) then it does a game of pure skill.
I actually think this is the reason Protoss cant seem to find a single "Hero" as Terran and Zerg have been able to. All of Protosses matchups are just too coinflippy and weird
|
How do people even pick up on things like this?
Any Protoss player who didn't find Supernova's smurf and look at his match history would've made the same read.
Are the sc2 pros really that into mind fucking eachother? I suppose group B was the group of death and desperate times call for desperate measures, but that's some serious dedication ^_^
|
wow great writing sad to have missed the games
|
This really helped me a lot. Both the talk about 3gate robo v 4gate follow up and the variations on the 4gate followup.
|
We need you commenting live for the GSL. Awesome analysis.
|
Thanks for writing this up. The word cheese is thrown around way too liberally. It is disrespectful to the player's skill I think.
|
I watched the games where MC lost and remember thinking "shoot, he's learned from the last game but he's playing a different guy now..."
Very good analysis! This reminds of me some old Final Edits (like God of War and Clutch Engage). Feels like it should be part of the Day 3 recap.
|
More please this is not ENOUGH I NEED MORE ANALYSIS
Good though
|
These statements was made in the OP and I would appreciate if pro-level players can enlighten me on them:
If gasless FE into medivacs is the soft counter to 1 gate FE into 3 gate robo, then gasless 3rd CC is the hard counter. What happens if the Protoss goes 2-base Colossus or immortal bust as a response to gasless 3rd CC? Lets say the Protoss goes Colossus, he should at the very worst be able to pressure safely and take a 3rd whilst the Terran cannot safely shift the OC to the 3rd.
When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo. I remember MC doing this 4-gate pressure against MMA in GSL Nov Code S ro32. Is the quoted statement true? MMA held the 4-gate pressure easily and won comfortably. I contend that 4-gate pressure is not that great against 1-rax FE into 3-rax. However, I do agree that 4-gate pressure will very likely kill a terran who plays greedy (e.g. 1-rax FE into double-gas).
|
Australia18228 Posts
Another thing to include is that, MC prefers taking a third and playing a macro game on Daybreak, so it would've hard for Leenock to assume he would all-in in a map he prefers to macro in.
On January 11 2012 15:16 Azzur wrote:These statements was made in the OP and I would appreciate if pro-level players can enlighten me on them: Show nested quote +When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo. I remember MC doing this 4-gate pressure against MMA in GSL Nov Code S ro32. Is the quoted statement true? MMA held the 4-gate pressure easily and won comfortably. I contend that 4-gate pressure is not that great against 1-rax FE into 3-rax. However, I do agree that 4-gate pressure will very likely kill a terran who plays greedy (e.g. 1-rax FE into double-gas).
Didn't MMA have three to four bunkers though? MC lost that game because he ended committing to that attack with 2-3 rounds of zealot warpins and didn't have enough to contest the third or take his own after it failed.
|
This isnt a good thing lol.
This is stupid. Yes they played it masterfully, but it's still just a cheap trick. This shouldnt win games, I'm starting to see why alot of top players always spout the term 'cointoss'
|
|
|
On January 11 2012 15:29 pjw wrote:
This isnt a good thing lol.
This is stupid. Yes they played it masterfully, but it's still just a cheap trick. This shouldnt win games, I'm starting to see why alot of top players always spout the term 'cointoss'
Disagree. Starcraft is a strategy game, not a mechanics challenge. If you can successfully play with your opponent's head, you deserve the win.
|
Even if ForGG lost, I'm glad some epic mindgames were busted out by ForGG. Love that guy.
|
On January 11 2012 15:40 Bengui wrote:Show nested quote +On January 11 2012 15:29 pjw wrote:
This isnt a good thing lol.
This is stupid. Yes they played it masterfully, but it's still just a cheap trick. This shouldnt win games, I'm starting to see why alot of top players always spout the term 'cointoss'
Disagree. Starcraft is a strategy game, not a mechanics challenge. If you can successfully play with your opponent's head, you deserve the win.
pwj, youtube 'Casey fake FFE' on youtube and then come back and tell me that such clever tricks don't deserve a victory. =)
|
On January 11 2012 14:34 Bonkarooni wrote:
You explained everything very well...and is absolutely horrifying for the future of sc2. This shows the vest worst side of sc2. There is no way to just play safe and reactive at the highest level. Builds counter builds in a giant cycle of mindfuckery, and the game becomes more a game of tricking one another (which is cool to a point) then it does a game of pure skill.
I actually think this is the reason Protoss cant seem to find a single "Hero" as Terran and Zerg have been able to. All of Protosses matchups are just too coinflippy and weird
You could conclude that
OR
You could notice that these counter builds are getting more subtle as players have been slowly learning safe builds.
Mindgames like this flesh out the game a bit, and through experience they stand a chance of refining their play further so as not to fall for fake expansions, etc.
|
Thank you very much for this article! I like to believe that there really isn't much random cheese anymore. I feel that especially with this new GSL format, that every build was calculated and prepared before hand (for the most part).
|
As amazing as this post is it just reaffirms how truly bad I am at this game :D
I guess it is motivating too, so much to learn!
|
every thing here is correct altho supernova build has some hints to it
The way you are hinted to know there is no expand in the back of the bunker is that when dooing the second scout(the stalker scout) you see there are only 4 marines.That means he might be going for a 2 rax with 2 gas and really fast addons, but this also means some more marines and a second bunker.Sometimes you feel things are not what they seem to be, so if mc would see there are 4 marines over there for way way to much time i would just pass by the bunker to make sure there is no cc, cancel the 4th gate(or keep it, it dosn't matter really) and get a robo.
I really really hope mc knows this cause it makes me sad when he loses like this =.='
|
But you can actually pressure with a 3gate robo (especially vs a fast 3rd cc), if you get blink, theres quite a few blink pressure builds popular on korea right now
|
Great post, I like it and I liked the cheese and strategy too!
|
Love your work. Was trying to explain the same thing to a lot of people lastnight who were vomiting abuse towards SuperNoVa.
|
this is sherlock holmes stuffs. nice work.
|
Really nice, thank you for the analysis. I could never have figured anything out myself.
|
I fucking love talk like this. We need more of it on this forum. This is the type of thing that reminds me why I love Starcraft.
|
Great post, you've really increased my enjoyment of those games and I hope I pick up something like that later.
Not everything is as it seems.
|
very well written, double thumbs up to you sir
|
Good read! Im gonna crono my two observers even more now 
|
On January 11 2012 09:26 chenchen wrote:
The second depot is actually very very late in a normal gasless 1 rax FE. As far as I understand, you want to plot down the second CC well before your second depot.
Wha? No it's not. If you did CC before 2nd depot you would get supply blocked at 19 supply and have a very hard time holding off early pressure. You can even just look at the basic build order in Liquipedia if you're unsure:
http://wiki.teamliquid.net/starcraft2/1_Rax_FE
|
Really nice post!!
The supernova vs MC one was obvious. I am honestly surprised the "show no gas + 1 bunker on low ground" is not used more ofter, since it's so easy and harmless to do.
The ForGG vs MC one was delicious. Fast 3rd CC against a player known not to forgive greed requires well a well planned strat and balls of steel.
The MC vs Leenock one was astonishing. As a matter of fact I play zerg and I deal with those timings all the time and yes, 5 zealots moving across map, no sentries, +1 done is definately a 4gate +1.
I mean the other two games were mindgames. This one was deception at it's best. Extremelly well disguised attack. When watching live I was mad with Leenock's greed who made only 1 spine. Now I realised how it looked from Leenock's perspective... and it looked nothing like an all-in. Congrats to MC!!
|
|
|
On January 11 2012 15:43 0neder wrote:Show nested quote +On January 11 2012 15:40 Bengui wrote:On January 11 2012 15:29 pjw wrote:
This isnt a good thing lol.
This is stupid. Yes they played it masterfully, but it's still just a cheap trick. This shouldnt win games, I'm starting to see why alot of top players always spout the term 'cointoss'
Disagree. Starcraft is a strategy game, not a mechanics challenge. If you can successfully play with your opponent's head, you deserve the win. pwj, youtube 'Casey fake FFE' on youtube and then come back and tell me that such clever tricks don't deserve a victory. =)
The first time it deserves a victory.
When stuff like this is repeated over and over and over again then it will just be a cointoss.
It's not even that smart, it's essentially baiting.
Seriously, just because this guy made a pretty post people are lapping this stuff up as some sort of amazing feat. This happens multiple times in every quake duel on a decent level. The only difference is a game isnt always decided after one wrong guess.
I really hope this kind of thing eventually fades out for more methodical play(see supernova vs Leenock) as not only is it boring to watch, it's miserable to see clearly more talented players get 'metagamed'
|
I wish I knew enough about the game to watch them and think like this. I don't sit there and cry like some people about cheese and all-ins and stuff like that but it would be cool to see the game like this.
|
This is gold. Great analisys. Thank you.
|
On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience.
You know, all competitions in life are all about the mind games. Fighting games, card games, physical sports, and even ye olde fist fight can be broken down to a mindgame, or that "coinflip" you so like to use.
|
Great post, was a real insight to TvP!
|
On the supernova vs MC game 3, (i play protoss)
Ive come across a few terrans who fake an expo with a bunker on the low ground.
They usually die to my 1 gate expo into 3 gate pressure(does get a bit close sometimes)
if i remember correctly after warpgate finished all Supernova had was 1 bunker with 4 marines on the low ground.
all his other production buildings were constructing addons.
At this point i thought MC's pressure would come, bust the bunker and he would throw down a robo asap after that, but for whatever reason it was slower.. (i remember MC walking units across the map instead of a proxy pylon)
Protoss's need to tweak there builds so pressure comes before cloaked banshees if they havent actually seen a CC going down. Any sort of pressure will either bust the bunker or notify you that something is wrong. No other units & no scvs pulled, at this point you must throw down a Robo asap
|
this was a very well thought out write up
good job
|
My mind is blown
That was great
|
More, please. One of the best reads in a long time.
|
5 stars this is amazingly well thought out
|
Well written and interesting to read, thanks.
|
On January 11 2012 14:06 sickoota wrote:Show nested quote +On January 11 2012 14:01 Plexa wrote:On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience. I guess poker is just coinflips to you too. I guess you can't deny that "the current trend is x, so I will assume my opponent will be preparing for x so I will do Y instead and hopefully win" is strategy of a sort, but I remain highly skeptical of the OP's claim that this is demonstrative of any sort of strategical brilliance or that spectators should be enthralled by the intelligence of these player's decisions. Isn't the fact that you can sum up each of these games in one sentence each (leenock thought MC was X because of current metagame trends but instead he Y, GG) indicative of something lacking? Compared to macro games, where the decisions would take pages and pages to explicate, there is something clearly unsatisfying about matches decided solely by BO mindgames.
this post explains very well how I sometimes feel about sc2. both as a spectator and as a player. nice post.
Surely everyone can agree that the best games require some back and forth action right?
|
Those are some great points that've been pointed out. props
|
Eye-opening indeed, very nice post.
|
Very nice post. I liked most the supernova game felt sorry for MC in that game.
I wish zerg had oportunity to make such mindgames in his matchups. All feel same macro up, defend all pressure, win in lategame with superior production. You cant rly make cute moves like this :/
|
Loved that Supernova expand feint, such a smart move, great run-down
|
Great read, though I do still feel for Leenock. It seems harder for Zerg to use clever mindgames to surprise your opponent and sweep a quick win. Only time it's possible is an all-in vs forge fe I think.
|
Good writeup.
Most of the mind games made sense during the match.
The important thing that lower level players like us need to remember is that it's VERY rare for a pro in a tournament to just blindly execute one of these builds with no thought process behind it.
People like to say coin flip, but look at the stats.
Outside of the top 3-5 players in the world, win rates eventually approach 50% as the game is played out.
If you see an edge because of a map or a build by your opponent or a weakness in the meta game, and you think your build has a say 65% chance of working, it's actually quite a sound bet.
|
On January 11 2012 21:56 Veriol wrote:
Very nice post. I liked most the supernova game felt sorry for MC in that game.
I wish zerg had oportunity to make such mindgames in his matchups. All feel same macro up, defend all pressure, win in lategame with superior production. You cant rly make cute moves like this :/
I would disagree that all zerg can do is hold pressure go macro and try win later. Zerg have both drops and nydus worm and they nearly never use them even though both are very powerful, zerg insist on playing macro because its proven to work really well I just feel that at some stage double infestor or infestor baneling drops to kill off whole mineral lines and nydus flanks have to happen. Even the old style +2 attack baneling drop is really effective if your opponent doesnt react quickly enough. I just feel like zerg puts themselves in this macro box where they have to have more bases at all times when really having more units positioned better is a really good way to play also.
|
Also great OP nice to see more than just complaints about balance being posted and to have actual strategy discussed
|
On January 11 2012 14:39 Blitzkrieg0 wrote:
Are the sc2 pros really that into mind fucking eachother? I suppose group B was the group of death and desperate times call for desperate measures, but that's some serious dedication ^_^
Not sure if this has been said yet, but there were 3 oGs members in that group, so I think it's safe to assume they're far more familiar with each others' tendencies and playstyles than most players.
Nice analysis, I also quite liked MCs proxy Stargate on Metropolis after he realized exactly what build ForGG was going for.
|
I definitely have a deeper understanding after reading all that. I did not even slightly catch how MC had been tricked into Banshee cloak play but your analysis is dead-on.
|
This is great. Thank you.
Now I will have to go back to those game and see them again just to appreciate these mind tricks :-D
|
Good read, and thanks for the insight, let me help you out with how to use the word metagame.
On January 11 2012 09:11 iamke55 wrote:
standard macro opening in the current metagame: 1 rax FE with no gas.
Supernova, knowing MC's play style and the fact that the current metagame favors 4 gate against Terran's gasless FE, mindgamed MC by going for cloaked banshees disguised as a gasless FE.
all popular in the current metagame:[list][*]4 gate +1 zealot pressure. This build exploits the trend of Zergs expanding twice without gas to skip any sentries or additional cannons beyond the first and sending 8 zealots with +1 weapons to the Zerg's 3rd base at 8 minutes.
Three oGs players today were able to take advantage of trends in the metagame and knowledge of players' past tendencies to disguise their builds, inducing their opponents into sub-optimal responses to their real strategy.
The first metagame should be replaced with Trend in strategy.
(metagame has to have a specific opponent in mind, there is no general metagame that works for the general public, because not everybody follows the same trend on whatever match up we are talking about, to apply metagame one must know something about the opponent, knowing the trend of what the general public would think in that particular match up doesn't imply what your opponent might do, it only implies what he knows.)
On the second sentence the word metagame should be replaced with Trend in strategy, and the word mindgamed should be replaced with metagame.
(Supernova can use his knowledge of MC's play style to metagame MC, as he knows how MC would react. He can't use the current trends of tvp to metagame MC, as he has no idea whether or not MC would follow the current trends of tvp if Supernova does not have any knowledge of MC's play style.)
On the 3rd sentence the word metagame should be replaced with Trend in strategy
(again, popular trend can only be interchangeable with metagame if the particular opponent's style follows strictly with the popular trend with little to no deviation.)
On the 4th sentence it should have just been trends in strategy, instead of trends in the metagame.
( you almost got it, you have half of the phase right, and then you put metagame in it and screw everything up. a;ldfjaldfjadfkja;dklfjakl;dfj)
|
I wish that all write-ups were this in-depth and informative.
|
Nice write up but i think your view of scouting is alittle bit off, or perception of scouting" very much so in tvp since that opening can be any of the following gasless FE or 111 or cloackbanshe to mech or just a 5-6 rax allin.
it was just simply bad play from mc's part i belive since he easy could have gotten the scout just buffer bunker with a stalker/zealot and run by with stalker/probe
|
How can people think mindgames are a bad thing?..... Hell, the majority of series in individual leagues in broodwar are based around mindgames much more heavily then is used in SC2 currently. Mind games are gooood, it adds a completely different level to the game where you have to actually think and not be a mechanical monkey.
|
Nice read! Unfortunately I haven't seen the games but I would love to read more analyses like this.
|
Waa this was an awesome read.
My PvZ will never be the same again.
Prepare for rape EU zergs!!!!
|
SoCal8908 Posts
im kinda curious if supernova and forgg planned those strategies in order to mess with MC - think about how intelligent that would be (albeit frustrating)
show MC that ogs terrans are capable and willing to switch into cloaked banshees and force him to play safely. then with the knowledge he'll play safely, take a quick third the next with NO gas.
might have been more viable in the old format, but interesting to think about nonetheless
|
sick read. glad to see that players are finally intelligently utilizing metagame tactics.
|
One of the best blogs I have read in a while. Thanks for that.
|
nice catches
you should make theese more often : )
|
On January 11 2012 14:24 Azzur wrote:Show nested quote +On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience. These games were coinflips because the players chose them to be. For instance, if MC had gone 1-gate FE into 3-gate robo against supernova, he should've easily won the game.
So MC does the build he did, ha automatically he loses, he does the other build he automatically wins, by your definition. Isnt that pretty much the definition of Supernova flipping a coin to see if he wins?
|
really great analysis. Keep it up :D
|
A++++ Will read again!
Really great analysis, as someone who doesn't get these small things, or doesn't pick them up during the game, I appreciate it
|
On January 12 2012 01:01 Typhus wrote:Show nested quote +On January 11 2012 14:24 Azzur wrote:On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience. These games were coinflips because the players chose them to be. For instance, if MC had gone 1-gate FE into 3-gate robo against supernova, he should've easily won the game. So MC does the build he did, ha automatically he loses, he does the other build he automatically wins, by your definition. Isnt that pretty much the definition of Supernova flipping a coin to see if he wins?
No.... If MC did 3gate + robo after being fed all the information that it was a 1rax expand then he would of been playing completely differently to what he ALWAYS does. Supernova used knowledge of MC's/Protoss tendencies to fool him. Not a coinflip.
I think a really great example of using knowledge of someones tendencies was Ganzi's game in Code A against Lucky on Crossfire(http://www.gomtv.net/2011gslsponsors7/vod/66573). Ganzi built a proxy depot/rax literally in the natural of the zerg and won - this would not of worked against anyone other than Lucky since he obviously watched alot of Lucky's games and found the position that he doesn't scout. In the winners interview(http://www.teamliquid.net/forum/viewmessage.php?topic_id=292447) he said " After the next match was determined, then I checked all his VODs and I saw his habits that were hard to fix. His scouting pattern was always so I chose that build. Most people said that it would get scouted but I was sure that it would work.".
|
Great analysis for the PvTs! I think you are slightly off on MC vs Leenock PvZ. All you need is 4 gates and a chrono cybercore (which you'd get for a 7 gate anyway) to impose the threat of a +1 4 gate zealot warp. Most players make early sentries too, just to save energy and be safe from cheese. Just by getting the infrastructure of quick 4 gates, you can scare the shit out of the zerg like what happened to Leenock, because he doesn't know if you have 2-3 probes around the map laying down pylons, especially if you contest the watchtowers like MC probably did.
|
Excellent writeup, it's impressive that you remember so many of MC's games, not just from GSL. HSC had so many games that it's hard to remember them all.
|
Thank you for letting us know cloaked banchee is not cheese. Doh...
|
I wish there were more articles like this in the strategy forum. I only saw this because of Reddit. Does analysis like this go on in the blogs section all the time? Is it not allowed in the strategy forum?
|
wow O_O very interesting read, learned few things from this
|
Awesome post, great analysis.
|
On January 12 2012 01:01 Typhus wrote:Show nested quote +On January 11 2012 14:24 Azzur wrote:On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience. These games were coinflips because the players chose them to be. For instance, if MC had gone 1-gate FE into 3-gate robo against supernova, he should've easily won the game. So MC does the build he did, ha automatically he loses, he does the other build he automatically wins, by your definition. Isnt that pretty much the definition of Supernova flipping a coin to see if he wins?
You misunderstood my statement - I'm claiming that the Supernova vs MC game was a coinflip only because the players chose for it to be one. Yes, Supernova coinflipped and MC in return coinflipped and lost. Instead, MC had a perfectly good non-coinflip build which is good against many things but he chose to not use it. The OP highlighted that the 3-gate robo is hard countered by no-gas 3CC but I disagreed with that assessment. In return, Supernova had a perfectly good non coinflip build as well (1-rax FE).
Thus, SC2 doesn't have to be a coinflip game unless the players chose to indulge in it.
|
On January 11 2012 09:26 chenchen wrote:
The second depot is actually very very late in a normal gasless 1 rax FE. As far as I understand, you want to plot down the second CC well before your second depot.
You can but you can also put down the depot to deny the scout. But as a Protoss player you shouldn't use that second depot as a reason why the Terran is not getting gas. PuMa walls off with a second Supply Depot all the time so that Protoss players never know what he's sending at them. But still, it's a fairly good way to metagame your opponent.
|
Seems like nothing but poker strats/coin flipping hoping it works out in your favor ;/
|
On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience.
I disagree. I think that as a spectator, obviously seeing intense macro games is great, but watching a clever build unfold is always interesting. When you watch BitByBit play it has very little strategy and is more along the lines of All-in with a follow-up transition of another All-in.
Also, as far as strategy games go, "coinflips" (at least at this level) are less along the lines of coinflips and more along the lines of well-thought out strategies. That's why Flash is one of the greatest progamers of all time, because the "risks" he takes are based off of a plethora of information outside of the game which he had gathered. Limiting your opponent's scouting and only showing them what you want them to see may seem like a "horrid spectator experience," but for the more indepth fans of both StarCraft and strategy, those subtleties make the games highly enjoyable.
EDIT:
Strategy is all about risks especially when you're playing in a limited information environment. Also players taking risks makes for exciting games.
|
Great write-up. This should be in the strategy forum imo.
|
Amazing article, loved it. I learned so much from this, thanks!
|
|
|
On January 11 2012 09:26 chenchen wrote:
The second depot is actually very very late in a normal gasless 1 rax FE. As far as I understand, you want to plot down the second CC well before your second depot.
A gasless 1 rax fe is not the safest build in the universe by showing your build early you give your opponent the chance to react to it optimally. I have done the cc before second depot a ton and even on my level it lead to me encountering an absurd amount of all ins because my opponents had all the information they needed. Walling off early with a second depot forces your opponent to either take a risk or to play safe which makes your 1 rax expand a lot stronger because Protoss will probably end up with a lower probe count and a slower third base. You can still do the faster cc variant on big maps where it's harder to scout but especially on pro level you have to be extremely comfortable to always do it like this in important matches.
|
The cloaked banshee one was pretty smart. Its crazy how much difference it makes to know your opponent even before starting a game.
|
Honestly, this feels like just dressing up a weighted coin flip. Yes, you can research all sorts of things to weigh the coin so it favors a certain outcome, but in the end, a coin is still being flipped.
|
This blog is so awesome, normally i always skip the really deep strategic articles since i can't play SC2 on my pc, so i'm a pure watcher. But this blog is so indepth and clear that it really grabbed me.
Seems like OGS spends alot of time studying the opponents.
|
Mind games cause the biggest face palms and I love watching them go down. Poor MC though, I'd go as far as saying supernova planned out this "find muck" in consideration of MC being jet lagged. They are team mates, so who else better to toy with? Especially when you know they are exhausted o_O?
|
Great post, the bunker in front of the natural is a great choice as it denies scouting and is generally indicative of a fast cc. Nice read.
|
Good read; well reasoned.
|
On January 12 2012 08:08 pikagrue wrote:
Honestly, this feels like just dressing up a weighted coin flip. Yes, you can research all sorts of things to weigh the coin so it favors a certain outcome, but in the end, a coin is still being flipped.
Isn't that what all games are, unless they're like chess and have perfect information and turn-based?
Sure, you might have really good micro. Sure, your strategic mind has picked up on his patterns and how he moves his army. But there's always that chance that he does a random a-move when you're not expecting it and your marines melt to banelings.
Sure, you think he's going to drop your main. He has to, right? That's what he's teching for, right? You put part of your army to defend because there's a 90% chance of a drop happening. But the opponent, he doesn't give a fuck. He marches right into your third and kills it before you can respond. That too is possible.
Skill and preparation are about weighting the coin, right? You're never going to have a guaranteed victory. All you can do is give yourself the best chance possible.
|
Great write up. However, I do resent you show that MC was the person to get mind-gamed in 2/3 of the examples. -.-
|
I clap for thee! I love this stuff.
|
Very nice read. You have an understanding of strategy that many casters dont even have. SC2 is a complicated game once you include the meta game and so helping more people to understand it will no doubt make them enjoy watching these amazing players duke it out more
|
Is this Kwark in disguise?
Great analysis!
|
Wow this is some high level analysis. Nice write up.
|
|
|
|
|
On January 11 2012 15:16 Azzur wrote:These statements was made in the OP and I would appreciate if pro-level players can enlighten me on them: Show nested quote +If gasless FE into medivacs is the soft counter to 1 gate FE into 3 gate robo, then gasless 3rd CC is the hard counter. What happens if the Protoss goes 2-base Colossus or immortal bust as a response to gasless 3rd CC? Lets say the Protoss goes Colossus, he should at the very worst be able to pressure safely and take a 3rd whilst the Terran cannot safely shift the OC to the 3rd. Show nested quote +When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo. I remember MC doing this 4-gate pressure against MMA in GSL Nov Code S ro32. Is the quoted statement true? MMA held the 4-gate pressure easily and won comfortably. I contend that 4-gate pressure is not that great against 1-rax FE into 3-rax. However, I do agree that 4-gate pressure will very likely kill a terran who plays greedy (e.g. 1-rax FE into double-gas).
I think Mc did it wrong in the game game against MMA where he wanted to win there but lost a lot of units. If you look at his game against Clound he just tested the water and when he found out the terran was playing standard then he retreated and did not lose anything.
|
On January 11 2012 21:56 Veriol wrote:
Very nice post. I liked most the supernova game felt sorry for MC in that game.
I wish zerg had oportunity to make such mindgames in his matchups. All feel same macro up, defend all pressure, win in lategame with superior production. You cant rly make cute moves like this :/
I think you can. The nestea and Mvp game3 is some kind of mind game there in my view.
|
GOMtv just linked to this article on facebook! Nice work. Seems like they appreciate it as well.
|
So I think we all can learn one thing from these games and that is SC1 pro gamers are pretty good at mind game.
btw for those who think zerg can not mind game, just think about how fast zerg can swith its tech comparing to other two.
Zerg surely has the ability to mind game, just give some time and we will see more. GSL codeS is getting much better.
|
On January 12 2012 15:01 MementoMori wrote:
GOMtv just linked to this article on facebook! Nice work. Seems like they appreciate it as well.
They just wish foreigners can understand this game more like Korean do.
|
Thanks for this. I'm sure the players appreciate you mentioning some of the subtlety to the games as well.
|
|
|
i think mc made a poor read on supernova, actually. as it has been said before, if terran is going 1 rax no gas FE his second depot should be far more delayed. although a lot of players do delay their cc until after the depot, i think mc should not have - or possibly didn't - assume there was an expo until he saw the bunker at the ramp. but even then, i think that bunker timing was just too early 1 rax CC > 3-4 rax to be confirmed. a probe could have gone up the ramp far before his zealot+stalker and confirmed that the bunker went down super fast. if mc had scouted the bunker timing he may have been able to get a better read on the situation and played safer by getting up his detection.
in summation, both the depot and the bunker were too early for supernova to realistically be doing an efficient 1 rax CC opening, and i believe it was mc's mistake for not realizing that.
|
On January 12 2012 16:12 megapants wrote:
i think mc made a poor read on supernova, actually. as it has been said before, if terran is going 1 rax no gas FE his second depot should be far more delayed. although a lot of players do delay their cc until after the depot, i think mc should not have - or possibly didn't - assume there was an expo until he saw the bunker at the ramp. but even then, i think that bunker timing was just too early 1 rax CC > 3-4 rax to be confirmed. a probe could have gone up the ramp far before his zealot+stalker and confirmed that the bunker went down super fast. if mc had scouted the bunker timing he may have been able to get a better read on the situation and played safer by getting up his detection.
in summation, both the depot and the bunker were too early for supernova to realistically be doing an efficient 1 rax CC opening, and i believe it was mc's mistake for not realizing that.
He was tired and jetlaged. He played well enough in that situation.
|
Awesome insight. I was wondering if you could do in depth analysis of some more of these mind games if you see them. I only picked up on the 4gate by MC but that was easy due to the blatant overlord scout
|
The first mindgame revolves a lot around 4 gate being able to pressure terran, forcing him into disadvantageous precauteons. I don`t play protoss, so I`d be really thankful, if anyone could explain just in a few words, why exactly can`t the same (albeit somewhat weaker) pressure be applied by "only" 3 gates (+ a robo). I mean, woudn`t that be still very strong against fast 3rd CC (which is the whole point - to punish the greed)?
|
great read man, love to see if the players actually thought this way!
|
On January 11 2012 17:56 pjw wrote:Show nested quote +On January 11 2012 15:43 0neder wrote:On January 11 2012 15:40 Bengui wrote:On January 11 2012 15:29 pjw wrote:
This isnt a good thing lol.
This is stupid. Yes they played it masterfully, but it's still just a cheap trick. This shouldnt win games, I'm starting to see why alot of top players always spout the term 'cointoss'
Disagree. Starcraft is a strategy game, not a mechanics challenge. If you can successfully play with your opponent's head, you deserve the win. pwj, youtube 'Casey fake FFE' on youtube and then come back and tell me that such clever tricks don't deserve a victory. =) The first time it deserves a victory. When stuff like this is repeated over and over and over again then it will just be a cointoss. It's not even that smart, it's essentially baiting. Seriously, just because this guy made a pretty post people are lapping this stuff up as some sort of amazing feat. This happens multiple times in every quake duel on a decent level. The only difference is a game isnt always decided after one wrong guess. I really hope this kind of thing eventually fades out for more methodical play(see supernova vs Leenock) as not only is it boring to watch, it's miserable to see clearly more talented players get 'metagamed'
I doubt it will fall off, because as trends change people will change up their builds and new builds will create more exploitable weaknesses. I think this type of play is just fine, if you can find a weakness in someones build and play why not exploit it? Its up to the other player to be good enough to react before its too late, and if they react correctly and quickly enough, assuming the exploit was not totally all in, then you will have your back and forth macro game.
|
great writeup, and I agree. I especially liked when Nestea busted MVP's greedy 3 cc build!
|
On January 12 2012 09:27 Redmark wrote:Show nested quote +On January 12 2012 08:08 pikagrue wrote:
Honestly, this feels like just dressing up a weighted coin flip. Yes, you can research all sorts of things to weigh the coin so it favors a certain outcome, but in the end, a coin is still being flipped. Isn't that what all games are, unless they're like chess and have perfect information and turn-based? Sure, you might have really good micro. Sure, your strategic mind has picked up on his patterns and how he moves his army. But there's always that chance that he does a random a-move when you're not expecting it and your marines melt to banelings. Sure, you think he's going to drop your main. He has to, right? That's what he's teching for, right? You put part of your army to defend because there's a 90% chance of a drop happening. But the opponent, he doesn't give a fuck. He marches right into your third and kills it before you can respond. That too is possible. Skill and preparation are about weighting the coin, right? You're never going to have a guaranteed victory. All you can do is give yourself the best chance possible.
No because in BW and hopefully in SC2 in the future, micro skill can overcome this kind of thing. While mindgames especially in long series and teamgames are fun and interesting if they completely decide games and so often it's going to get stale.
|
On January 13 2012 03:13 infinity2k9 wrote:Show nested quote +On January 12 2012 09:27 Redmark wrote:On January 12 2012 08:08 pikagrue wrote:
Honestly, this feels like just dressing up a weighted coin flip. Yes, you can research all sorts of things to weigh the coin so it favors a certain outcome, but in the end, a coin is still being flipped. Isn't that what all games are, unless they're like chess and have perfect information and turn-based? Sure, you might have really good micro. Sure, your strategic mind has picked up on his patterns and how he moves his army. But there's always that chance that he does a random a-move when you're not expecting it and your marines melt to banelings. Sure, you think he's going to drop your main. He has to, right? That's what he's teching for, right? You put part of your army to defend because there's a 90% chance of a drop happening. But the opponent, he doesn't give a fuck. He marches right into your third and kills it before you can respond. That too is possible. Skill and preparation are about weighting the coin, right? You're never going to have a guaranteed victory. All you can do is give yourself the best chance possible. No because in BW and hopefully in SC2 in the future, micro skill can overcome this kind of thing. While mindgames especially in long series and teamgames are fun and interesting if they completely decide games and so often it's going to get stale.
I hope so.
BO loss into a-move happens so much even at top level play... micro needs to have more of an effect outside of the first two seconds of a battle.
|
On January 13 2012 03:13 infinity2k9 wrote:Show nested quote +On January 12 2012 09:27 Redmark wrote:On January 12 2012 08:08 pikagrue wrote:
Honestly, this feels like just dressing up a weighted coin flip. Yes, you can research all sorts of things to weigh the coin so it favors a certain outcome, but in the end, a coin is still being flipped. Isn't that what all games are, unless they're like chess and have perfect information and turn-based? Sure, you might have really good micro. Sure, your strategic mind has picked up on his patterns and how he moves his army. But there's always that chance that he does a random a-move when you're not expecting it and your marines melt to banelings. Sure, you think he's going to drop your main. He has to, right? That's what he's teching for, right? You put part of your army to defend because there's a 90% chance of a drop happening. But the opponent, he doesn't give a fuck. He marches right into your third and kills it before you can respond. That too is possible. Skill and preparation are about weighting the coin, right? You're never going to have a guaranteed victory. All you can do is give yourself the best chance possible. No because in BW and hopefully in SC2 in the future, micro skill can overcome this kind of thing. While mindgames especially in long series and teamgames are fun and interesting if they completely decide games and so often it's going to get stale.
I agree that mind games should not define a game or lead to a win outright. If anything, they should allow a player to gain an advantage which they can then capitalize through solid play.
|
|
|
United States8476 Posts
MC's scouting probe was blocked by a barracks, 2 supply depot wall. Players savvy in the PvT matchup may know that if you scout the opponent on your first try (which is guaranteed on GSL Antiga Shipyard because of forced cross positions), your probe arrives at the Terran's base when a standard gas opening doesn't have enough minerals to build the second supply depot. So if you see that second depot, you know that Terran didn't build a refinery yet.
Disagree here. When I see a fast supply depot like that, it tells me my opponent doesn't want me to know if he has gas or not.
When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo. Gasless FE is usually followed up by either medivac/stim tech(the standard TvP build) or by a fast 3rd CC, neither of which a robo is immediately useful against. A robo with observers kills Protoss' ability to pressure before Terran's stimpack is done researching, and also delays Protoss' 3rd base unnecessarily. At best the 3 gate robo followup comes out even against standard medivac play, and is terrible against fast 3rd CC. 4 gate, on the other hand, puts immense pressure on Terran between 8 and 9 minutes. Terran is especially vulnerable during this period because the standard medivac build has neither stim nor medivacs at this time, and fast 3rd CC delays those even further. This combined with the fact that Terran fears a potential 6 gate all-in means that he must make more bunkers and pull many SCVs off of mining to get ready to repair. Until Terran has stim and medivacs (or ghosts if he went for that route), it is not safe to sell the bunkers and send SCVs back to mining. Once the 9 minute mark is reached and Terran has stim, Protoss simply retreats his units and takes a 3rd base with the resources saved from not getting any tech or upgrades. All in all, the Terran is in a worse position economically than he would be if Protoss didn't pressure
I highly disagree with most of the stuff in this section. Basically, you haven't convinced me at all that 1 gate FE into 4 gate is strictly superior to 1 gate FE into 3 gate robo vs 1 rax cc. If you've read a lot of my strategy forum posts, you'll find that I'm a huge anti-fan of blanket statements like "When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo." Every single "greedy" build that Terran can do including fast cloaked banshee and 3rd CC can hold the standard 4 gate pressure with 2-3 bunkers and a decent scv pull. In addition, a safe terran has to prepare for a 4 gate pressure even if you're doing 3 gate robo. A really fast observer does have its uses, being able to scout for such non-superstandard play such as ghost pushes, fast 3rd cc, cloaked banshee, or marine tank pushes. Then, the observers become useful for fending off drops. With a 4 gate opening, you're relying on your push for scouting, which not always paint as clear a picture. As for the potential 6 gate allin, a 3 gate robo player can do an even stronger 6 gate immortal allin against terran; MC himself has shown many variations of such a build.
If gasless FE into medivacs is the soft counter to 1 gate FE into 3 gate robo, then gasless 3rd CC is the hard counter. Pretty much disagree for the same reason I disagreed in the above point. At least with 3 gate robo, you can scout the terran's 3rd cc in time and start your nexus accordingly. With 4 gate, you might just stay on 2 base for too long. You can hold off any early pressure from toss if you play it well. The big disadvantage with going fast 3rd CC is that you have to kind of blindly prepare for everything to be safe and be very creative with your scouting, either with hidden SCVs or burning scans. That is, you have to prepare extra bunkers for vs 6 gates, fast starport for colossi allins, and eng bays to not fall behind versus double upgrades.
|
On January 13 2012 03:13 infinity2k9 wrote:Show nested quote +On January 12 2012 09:27 Redmark wrote:On January 12 2012 08:08 pikagrue wrote:
Honestly, this feels like just dressing up a weighted coin flip. Yes, you can research all sorts of things to weigh the coin so it favors a certain outcome, but in the end, a coin is still being flipped. Isn't that what all games are, unless they're like chess and have perfect information and turn-based? Sure, you might have really good micro. Sure, your strategic mind has picked up on his patterns and how he moves his army. But there's always that chance that he does a random a-move when you're not expecting it and your marines melt to banelings. Sure, you think he's going to drop your main. He has to, right? That's what he's teching for, right? You put part of your army to defend because there's a 90% chance of a drop happening. But the opponent, he doesn't give a fuck. He marches right into your third and kills it before you can respond. That too is possible. Skill and preparation are about weighting the coin, right? You're never going to have a guaranteed victory. All you can do is give yourself the best chance possible. No because in BW and hopefully in SC2 in the future, micro skill can overcome this kind of thing. While mindgames especially in long series and teamgames are fun and interesting if they completely decide games and so often it's going to get stale.
This is delusional talk. Mindgames still decide games ALL the time in BW.
|
I was annoyed at so many calling jjaki a cheeser and stuff like that, glad someone points out how badly he was mind gaming his teammate. Also nice explanations for the rest--I miss this stuff. I'm glad MC was able to make it out in the end. I'd be sad if his HSC casting got him knocked out of the RO32.
|
On January 13 2012 10:20 NrGmonk wrote:Show nested quote +MC's scouting probe was blocked by a barracks, 2 supply depot wall. Players savvy in the PvT matchup may know that if you scout the opponent on your first try (which is guaranteed on GSL Antiga Shipyard because of forced cross positions), your probe arrives at the Terran's base when a standard gas opening doesn't have enough minerals to build the second supply depot. So if you see that second depot, you know that Terran didn't build a refinery yet. Disagree here. When I see a fast supply depot like that, it tells me my opponent doesn't want me to know if he has gas or not. Show nested quote +When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo. Gasless FE is usually followed up by either medivac/stim tech(the standard TvP build) or by a fast 3rd CC, neither of which a robo is immediately useful against. A robo with observers kills Protoss' ability to pressure before Terran's stimpack is done researching, and also delays Protoss' 3rd base unnecessarily. At best the 3 gate robo followup comes out even against standard medivac play, and is terrible against fast 3rd CC. 4 gate, on the other hand, puts immense pressure on Terran between 8 and 9 minutes. Terran is especially vulnerable during this period because the standard medivac build has neither stim nor medivacs at this time, and fast 3rd CC delays those even further. This combined with the fact that Terran fears a potential 6 gate all-in means that he must make more bunkers and pull many SCVs off of mining to get ready to repair. Until Terran has stim and medivacs (or ghosts if he went for that route), it is not safe to sell the bunkers and send SCVs back to mining. Once the 9 minute mark is reached and Terran has stim, Protoss simply retreats his units and takes a 3rd base with the resources saved from not getting any tech or upgrades. All in all, the Terran is in a worse position economically than he would be if Protoss didn't pressure I highly disagree with most of the stuff in this section. Basically, you haven't convinced me at all that 1 gate FE into 4 gate is strictly superior to 1 gate FE into 3 gate robo vs 1 rax cc. If you've read a lot of my strategy forum posts, you'll find that I'm a huge anti-fan of blanket statements like "When the Protoss knows Terran is opening with gasless FE and takes his own fast expo, 4 gate pressure is simply a better build than 3 gates with a robo." Every single "greedy" build that Terran can do including fast cloaked banshee and 3rd CC can hold the standard 4 gate pressure with 2-3 bunkers and a decent scv pull. In addition, a safe terran has to prepare for a 4 gate pressure even if you're doing 3 gate robo. A really fast observer does have its uses, being able to scout for such non-superstandard play such as ghost pushes, fast 3rd cc, cloaked banshee, or marine tank pushes. Then, the observers become useful for fending off drops. With a 4 gate opening, you're relying on your push for scouting, which not always paint as clear a picture. As for the potential 6 gate allin, a 3 gate robo player can do an even stronger 6 gate immortal allin against terran; MC himself has shown many variations of such a build. Show nested quote +If gasless FE into medivacs is the soft counter to 1 gate FE into 3 gate robo, then gasless 3rd CC is the hard counter. Pretty much disagree for the same reason I disagreed in the above point. At least with 3 gate robo, you can scout the terran's 3rd cc in time and start your nexus accordingly. With 4 gate, you might just stay on 2 base for too long. You can hold off any early pressure from toss if you play it well. The big disadvantage with going fast 3rd CC is that you have to kind of blindly prepare for everything to be safe and be very creative with your scouting, either with hidden SCVs or burning scans. That is, you have to prepare extra bunkers for vs 6 gates, fast starport for colossi allins, and eng bays to not fall behind versus double upgrades.
Whether or not Terran wants you to know about his gas doesn't change the fact that he can't afford both a refinery and a supply depot that early, and that this is a common sign of 1 rax FE. Of course, nobody commits to a read off of just the supply depot, but when you also see the low ground bunker, that's when you're almost certain there's a CC behind it.
About the rest, you must be playing the safest Terrans in the world if they play the same against 3 gate robo and 4 gate. I don't know how you can deny that 4 gate forces more bunkers and SCVs pulled. I even have a replay of you playing against NrGAvoid who made 1 bunker and didn't pull SCVs because you never went to his side of the map despite seeing his expansion CC. MC did the greediest possible 3 gate robo in his game against ForGG, taking his third at 9 minutes just like he does when using 4 gates. He still faced a 30+ supply disadvantage and lost after killing 2 full medivacs for free. Compare to his game vs Cloud at HSC4, where he donated a nexus, 4 zealots, and his entire group of stalkers over the course of the game and still won through pure macro. These are mistakes you would never get away with against a Terran who's free to do whatever he wants off of 1 bunker. MC played way better against ForGG on Bel'Shir Beach than he did against Cloud on Daybreak, but lost mainly because ForGG had 50 more supply than Cloud at any given point in the game. If you want to see how much harder it is to play against a greedy Terran, I have a friend you could play against who never makes bunkers.
|
Just too damn smart, great and eyeopening read.
|
United States8476 Posts
On January 13 2012 15:19 iamke55 wrote:
Whether or not Terran wants you to know about his gas doesn't change the fact that he can't afford both a refinery and a supply depot that early, and that this is a common sign of 1 rax FE. Of course, nobody commits to a read off of just the supply depot, but when you also see the low ground bunker, that's when you're almost certain there's a CC behind it.
You can make both that early as Terran. If you couldn't then how do you explain how Supernova made both too early? You just have to delay your OC a bit or go 16 OC.
About the rest, you must be playing the safest Terrans in the world if they play the same against 3 gate robo and 4 gate. I don't know how you can deny that 4 gate forces more bunkers and SCVs pulled. I even have a replay of you playing against NrGAvoid who made 1 bunker and didn't pull SCVs because you never went to his side of the map despite seeing his expansion CC. MC did the greediest possible 3 gate robo in his game against ForGG, taking his third at 9 minutes just like he does when using 4 gates. He still faced a 30+ supply disadvantage and lost after killing 2 full medivacs for free. Compare to his game vs Cloud at HSC4, where he donated a nexus, 4 zealots, and his entire group of stalkers over the course of the game and still won through pure macro. These are mistakes you would never get away with against a Terran who's free to do whatever he wants off of 1 bunker. MC played way better against ForGG on Bel'Shir Beach than he did against Cloud on Daybreak, but lost mainly because ForGG had 50 more supply than Cloud at any given point in the game. If you want to see how much harder it is to play against a greedy Terran, I have a friend you could play against who never makes bunkers.
Using my own replays as an example is hax. We were both playing the same strats over and over for a long time, so he knew he could be greedy with me. My point is that as a terran you have to account for 4 gate to be completely safe, because it's really hard to tell the difference between 3 gate robo and 4 gate until the attack actually comes. Any terran who goes fast banshee or 3 cc will put up 2+ bunkers if they're being extra safe. In fact, there was a thread about how LiquidHero was being too aggressive with strats like his 4 gates against Puma and that's why he lost the NASL finals.
In the particular game vs ForGG, MC knew for a fact that ForGG had not gas for a very long time. If ForGG had actually gone cloaked banshees, they would have been more than a minute late, which would have given enough time for MC to do his 4 gate pressure and then add a robo and base trade versus the cloaked banshees. Thus, I doubt that MC went 3 gate robo largely in part because he feared 1 base cloaked banshee.
Also in the game vs ForGG, MC was only down 10 supply after killing 2 medivacs, but was whittle down to a 30 supply deficit directly after the battle, because he army was out of position and he had a unit composition disadvantage. Imo the big mistake of MC in that game was getting colossi too late. In my experience, vs a Terran who rushes 3cc, you need some type of fast AoE to stop the mass bio that will eventualyl be coming. I can't comment much about the game vs Cloud because I don't remember it.
|
On January 12 2012 05:00 Azzur wrote:Show nested quote +On January 12 2012 01:01 Typhus wrote:On January 11 2012 14:24 Azzur wrote:On January 11 2012 13:53 sickoota wrote:
You can only dress up coinflipping so much, sorry. If mindgames alone can dictate 4 out of every 5 games that is not a good place for a game to be in competitively. Denying scouting and putting a bunker on the lowground isn't some revolutionary, highly intelligent play, its just praying that dumb luck will favor you based on a couple of replays you watched of your competitor and it makes for a horrid spectator experience. These games were coinflips because the players chose them to be. For instance, if MC had gone 1-gate FE into 3-gate robo against supernova, he should've easily won the game. So MC does the build he did, ha automatically he loses, he does the other build he automatically wins, by your definition. Isnt that pretty much the definition of Supernova flipping a coin to see if he wins? You misunderstood my statement - I'm claiming that the Supernova vs MC game was a coinflip only because the players chose for it to be one. Yes, Supernova coinflipped and MC in return coinflipped and lost. Instead, MC had a perfectly good non-coinflip build which is good against many things but he chose to not use it. The OP highlighted that the 3-gate robo is hard countered by no-gas 3CC but I disagreed with that assessment. In return, Supernova had a perfectly good non coinflip build as well (1-rax FE). Thus, SC2 doesn't have to be a coinflip game unless the players chose to indulge in it.
You disagree with 3-gate robo being hard countered by fast 3rd CC, I don't see any justification besides you just saying you disagree with it while OP shows an example of MC, one of the best if not the best macro Protoss, being down 30 supply by midgame even after killing off 2 fully loaded medievacs.
|
This kind of stuff needs to find its way into the official recaps.
|
Wow...that was an amazing read...thank you!
|
Love this article. Two thumbs up for looking beneath the obvious builds to the complex mindgames that go on in the highest level of games.
|
Good read. I'm going offtopic here a little but i always wondered what the hotkey was for 'Vision' to appear, ![[image loading]](http://i.imgur.com/9Oh1u.jpg) . If anyone can answer me, will be much appreciated ^^
|
This is a really cool post. I always suspected that the sloppier, perhaps cheesier games that people roll their eyes at have more complicated stuff going on that nobody catches, but I'm not good enough to see them myself, so I'm glad someone's illuminating these details for me. I hope more analysis like this is done in the future.
|
On January 15 2012 05:23 ZONTICKSZ wrote:Good read. I'm going offtopic here a little but i always wondered what the hotkey was for 'Vision' to appear, . If anyone can answer me, will be much appreciated ^^
Select a unit or building of the player whose vision you want to see and hold V.
|
This is a nice post for people who may not have gotten to see the GSL games, but isn't this an exact regurgitation of Artosis' and Tasteless' live commentary/analysis of the games?
|
Moral of the story is simple. Until a player realizes the fact that he is playing against a real human and not an AI that hammers out specific stuff at specific timings for specific maps, he will forever be stuck at a mediocre level. Intellectual "reading" into a play without considering that an opponent can decieve you will leave you vulnerable. You got to develop an instinct and not rely on the "If I see A then option 1 is the proper response" thinking. Sometimes overthinking will be your downfall. Never assume.
|
On January 15 2012 07:03 Dalguno wrote:Show nested quote +On January 15 2012 05:23 ZONTICKSZ wrote:Good read. I'm going offtopic here a little but i always wondered what the hotkey was for 'Vision' to appear, . If anyone can answer me, will be much appreciated ^^ Select a unit or building of the player whose vision you want to see and hold V.
Or just press 1 or 2.
|
Great article sir, well written, good read, thanks
|
thanks for the analysis wish there was more quality content like this on TL
|
United States17042 Posts
On January 15 2012 14:15 HardMacro wrote:Show nested quote +On January 15 2012 07:03 Dalguno wrote:On January 15 2012 05:23 ZONTICKSZ wrote:Good read. I'm going offtopic here a little but i always wondered what the hotkey was for 'Vision' to appear, . If anyone can answer me, will be much appreciated ^^ Select a unit or building of the player whose vision you want to see and hold V. Or just press 1 or 2.
simple questions simple answers thread
|
Absolutely wonderful and informative read.
Thanks a bunch! <3
|
They're all cheese. I'm not impressed... /sarcasm
This blog was a very great read. A very good study of how early game plays snowballs to a win. Once the metagame shifts from a safe/risky early game builds to a more refined mid game transitions, this game will be more impressive.
I really can't wait for these kind of plays to be applied to the mid/late game transitions once the progamers' understanding elevates.
I just realized that, in my opinion, it took about a year and a half for these trends/metagame to develop, while during the early years of broodwar, we saw the same development in 2-3 years. This analysis really shows how the ex-BW players in SC2 have applied their experience and understanding of the game's depths.
I'm so hyped for what's to come! :D
|
On January 13 2012 20:40 NrGmonk wrote:Show nested quote +On January 13 2012 15:19 iamke55 wrote:
Whether or not Terran wants you to know about his gas doesn't change the fact that he can't afford both a refinery and a supply depot that early, and that this is a common sign of 1 rax FE. Of course, nobody commits to a read off of just the supply depot, but when you also see the low ground bunker, that's when you're almost certain there's a CC behind it. You can make both that early as Terran. If you couldn't then how do you explain how Supernova made both too early? You just have to delay your OC a bit or go 16 OC. Show nested quote +About the rest, you must be playing the safest Terrans in the world if they play the same against 3 gate robo and 4 gate. I don't know how you can deny that 4 gate forces more bunkers and SCVs pulled. I even have a replay of you playing against NrGAvoid who made 1 bunker and didn't pull SCVs because you never went to his side of the map despite seeing his expansion CC. MC did the greediest possible 3 gate robo in his game against ForGG, taking his third at 9 minutes just like he does when using 4 gates. He still faced a 30+ supply disadvantage and lost after killing 2 full medivacs for free. Compare to his game vs Cloud at HSC4, where he donated a nexus, 4 zealots, and his entire group of stalkers over the course of the game and still won through pure macro. These are mistakes you would never get away with against a Terran who's free to do whatever he wants off of 1 bunker. MC played way better against ForGG on Bel'Shir Beach than he did against Cloud on Daybreak, but lost mainly because ForGG had 50 more supply than Cloud at any given point in the game. If you want to see how much harder it is to play against a greedy Terran, I have a friend you could play against who never makes bunkers. Using my own replays as an example is hax. We were both playing the same strats over and over for a long time, so he knew he could be greedy with me. My point is that as a terran you have to account for 4 gate to be completely safe, because it's really hard to tell the difference between 3 gate robo and 4 gate until the attack actually comes. Any terran who goes fast banshee or 3 cc will put up 2+ bunkers if they're being extra safe. In fact, there was a thread about how LiquidHero was being too aggressive with strats like his 4 gates against Puma and that's why he lost the NASL finals. In the particular game vs ForGG, MC knew for a fact that ForGG had not gas for a very long time. If ForGG had actually gone cloaked banshees, they would have been more than a minute late, which would have given enough time for MC to do his 4 gate pressure and then add a robo and base trade versus the cloaked banshees. Thus, I doubt that MC went 3 gate robo largely in part because he feared 1 base cloaked banshee. Also in the game vs ForGG, MC was only down 10 supply after killing 2 medivacs, but was whittle down to a 30 supply deficit directly after the battle, because he army was out of position and he had a unit composition disadvantage. Imo the big mistake of MC in that game was getting colossi too late. In my experience, vs a Terran who rushes 3cc, you need some type of fast AoE to stop the mass bio that will eventualyl be coming. I can't comment much about the game vs Cloud because I don't remember it.
I went back and watched the game. I think the only real reason that MC didnt expect it to be a depot blocking information, and instead a early depot for a gasless FE is that his probe didnt go up the ramp immediately, and instead glitched out and went beside it. By the time he went up the ramp, 1 to 2 seconds later, the 2nd scv was gone and the scv making the barracks was working on the supply depot. IMO that might of made him believe the depot was at a natural timing, rather than seeing it started while the barracks was still building.
|
|
|
|
|
|
|
|
EPT
