EPTI am an extremely bad broodwar (and sc2 player). On good days I succeed in keeping my money below 1000 5 minutes into the game, but those days are exceedingly rare. I do however like to watch the Korean games and try to understand what's happening. I might never even come close to doing that, but I'd like to get a clearer picture of what runs through the minds of the players. It makes watching more fun for me. So I was hoping the upper-level players (and to be upper-level to me you'd have to be like D+) could weigh in and answer some of the questions I have. But beware, my questions very rarely stop >_>. Hopefully it will also be of use to the other people out there like me trying to better understand the game. There may be some things that are beyond my grasp since I do not have good playing experience, but I hope that's not too frequent.
With that out of the way, let's move on to the game that puzzled me today.
Kal vs Horang2, the ace match of Hite vs. STX. Presumably Kal was sent out instead of Calm or Shuttle because STX assumed that Leta would be playing the ace match for Hite. Leta seems to do well overall, certainly better than the other members of Hite. However, he has a bit of a weak matchup in his TvP, as he usually sits around the 50% mark there. Therefore the natural move for hite would be to send out Horang2, who at least statistically is a strong PvP player.
This decision isn't quite so obvious since the map obviously plays a huge factor. Unfortuantely the spare data on Grand Line means nothing and my map analysis is quite subpar, so I really can't make many useful comments on the decision. At the very least I can say that TvP looks difficult given the large distance to a third gas and the several million paths that a protoss army can take around the map.
Here's a random question that just popped into my head: do the islands on this map favor protoss or terran in a PvT? Could it be the third gas that the terran is looking for? After all, since double armory play requires a fast starport for the upgrades, couldn't you fit a dropship into that, use it to harass, and try to claim the island? This sounds slightly suspicious, as a competent protoss player probably will not simply allow you to turret up an island for free.
Looking at pro games, the most recent map to feature islands was Ultimatum (staying away from Outsider since that's so weird), which had a total of 6 TvP games (3-3, found here: Games). Looking at them, one was Flash vs Best, where Best attempted to take his third on the island and died to Flash's massive 6 factory timing push. Terran won again in Hwasin vs Guemchi, although according to the LR thread Guemchi succeeded in flying a shuttle full of 3 dts into turrets. Probably can throw that one out. I can't quite work out what happened in Stork vs Canata (the final game that Terran won). It looks like Stork attempted some sort of 2base carrier, which Canata scouted and rolled over. In the games where Protoss emerged victorious, Kal beat Hwasin on a failed BBS, Best beat Canata using 12nex -> 2base reaver -> 2base carrier and used this to take the island. Finally, Best took out ForGG after a failed midgame biomech push. Not sure how much can be gleaned from such a game. All in all, it looks like the protoss controlled the islands despite the extra incentive that Ultimatum offers for Terran to get a dropship - the cliffable natural. But I digress ...
Back to the game that I actually was planning on asking questions about, Kal vs Horang2. Whether or not it was the right decision given the map, I cannot comment further on. So we have a PvP, and Horang2, the supposed PvP expert, chooses a rather unorthodox opening: He expands after one gate.
Even though I do not know much, there is something I'd like to point out here. Progamers are not stupid. In fact, they are a lot better than any of us at Starcraft. We might question their in-game decisions; they are, after all, made in the heat of the moment. But when a player who was clearly selected to snipe another opens up with something unorthodox, there must be some solid thinking behind it. That's the main reason I'm so curious about it.
When thinking about fast expand PvP, two games come to mind. Neither is particularly similar to this one, but whatever. One is Stork vs Kal on Byzantium 2 (Link). Quick game summary: Kal tries to 14nex and defend with cannons, Stork builds a proxy robotics facility and walks a reaver to Kal's front, outranging the cannons and easily winning. The other game was Bisu vs Jangbi on Destination (Link), where Bisu expanded after one gate and one core (and a really sweet manner pylon). He held off Jangbi's 2gate robo pressure using the bridges of Destination suffering some probe losses, but appeared to emerge with the economic advantage. Horang2's nexus was after one gate, and Grand Line does not appear to have any bridges for his defense.
This is what Horang2 knows: He knows that the center is buildable, so proxy gates are a possibility. On top of that I think this makes a crossmap scout just as likely as a traditional scout. Again, since the center is buildable, he also knows that what Stork did to Kal (and what Kal ended up doing to him) is a possibility - to build production facilities in the middle of the map that negates the reinforcement time.
From this perspective, the luck factor of the game turned out reasonably well for Horang2. I don't think he expected some super-late scout from Kal on a 4-player map with a buildable center, and he even got cross positions. Yes he was scouted first, but with a cross map spawn and a traditional scout you're most likely to get scouted second anyway.
Horang2's plan looked like it revolved around defense of the nexus with cannons. Nothing Kal did was out of the ordinary or brilliant; he simply made lots of units. He went up to 3 gateways fairly quickly, added a robo and observer far sooner than he needed (his observer spotted Horang2's citadel of adun warping in), and won easily anyway. Horang2's plan was clearly dt-proof, but it did not appear to hold off much else. His 3 gates (naturally) came slower than Kal's so he simply did not have enough time to produce an equivalent army. He put up something like 4 total cannons which are not too effective against dragoons and were even shot at while they were warping in. Even a proxy robo, as in the Stork-Kal game, seems like it would've led to an easy victory.
So what was Horang2's plan? Maybe the early scout threw him off and the few seconds that Kal had to deviate from his normal build really did cost him the game. Does anyone have any experience with this 1-gate expand strategy? It does not appear that it gains any benefit from the map, or the opponent (Like Flash used to be known for scouting late, etc.). I'm pretty lost on this and I was wondering if anyone had an explanation.
Thanks in advance for any help you guys might have.
