I was nonplussed when GomTV didn't use a normal round robin format for the group stages and I was sure that playing one-off games instead of best of X matches was a mistake, but the system has grown on me. The modified round robin eliminates useless games and avoids lengthy tiebreakers, while the single match setup means there's a constantly shifting set of players and matchups.
As we swing into the later stages of the tournament, there are some incredible matchups to look forward to in the quarterfinals and beyond, plus some interesting storylines heading into the relegation matches with the best Code A players.
Race Breakdown
| Round | Terran | Zerg | Protoss |
|---|---|---|---|
| Round of 32 | 14 (44%) | 9 (28%) | 9 (28%) |
| Round of 16 | 9 (56%) | 4 (25%) | 3 (19%) |
| Quarterfinals | 4 (50%) | 2 (25%) | 2 (25%) |
As with all GSL tournaments, there's a lot of Terrans. Two Protoss and two Zerg makes this a reasonably diverse quarterfinals by historical standards, but whether or not that will continue into the semifinals remains to be seen.
Team Breakdown
| Team | Round of 32 | Round of 16 | Quarterfinals |
|---|---|---|---|
| PrimeWE | 6 | 1 | 1 |
| StarTale | 1 | 1 | 0 |
| oGs | 8 | 4 | 1 |
| IM | 2 | 2 | 2 |
| fOu | 3 | 2 | 1 |
| Liquid | 1 | 1 | 1 |
| EG | 1 | 1 | 1 |
| SlayerS | 2 | 1 | 0 |
| TSL | 4 | 2 | 1 |
| ZeNEX | 3 | 0 | 0 |
| None | 1 | 1 | 0 |
The biggest surprise is the what a rough time PrimeWE had in the opening round. With no players eligible for promotion from Code A, they are in danger from slipping from their perch as the second place team in number of Code S players. Most of the smaller teams did better.
Evaluating the Relegation Threat
Players who finished third in the opening group stage will have two chances to avoid relegation, for if they lose they can play a fourth place finisher to stay in. Here's an attempt to group the 16 players threatened with relegation by their danger of going down:
Low: HongUnPrimeWE, LeenockfOu, oGsEnsnare, oGsInCa, TSLFruitDealer, ZeNEXKyrix
All of these players are third place finishers who are widely believed to be strong players. Unlikely to lose against the Code A player, if they do they still would be favored against almost every fourth place finisher.
Medium: PoltPrimeWE, SlayerSLegalMinD
Also third place finishers, Polt and LegalMinD aren't quite at the level of the players I categorized under Low. Depending on how things shake out they could possibly be the underdog in both their matches.
High: CheckPrimeWE, MakaPrimeWE, oGsTheWinD, TSLRain
Since they finished fourth and have only one shot to stay up, these four have to be nervous. I've never been very impressed by Rain despite his GSL 3 run, but the rest have been impressive in the past. Although their play has fallen off somewhat as of late, they are still good players that Code A hopefuls will want to avoid.
Very High: AnyproPrimeWE, oGsjookTo, sanZenith, ZeNEXLiveForever
Of these players, jookTo probably has the best chance of avoiding relegation. The reason he got Code S in the first place was his consistent qualification across all three 2010 tournaments, so you know he does well against mediocre players. The other three players got to Code S based on their results in GSL 1 and haven't done much since. Apparently the Code A tournament finalists get to pick their opponents. I'd be shocked if sanZenith and ZeNEXLiveForever don't find themselves picked. The only way I could see it going differently would be if one of the finalists feels they are weak against Protoss and particularly strong against Zerg, in which case oGsjookTo might be the choice. If they want a Terran opponent and the other finalist took LiveForever, I guess it's just barely possible they could go with TSLRain. But no, I think sanZenith and LiveForever will get their numbers called.
Predictions
At the risk...actually, the certainty...of looking foolish, here are my picks for the quarterfinals:
choyafOu (P) vs. IMNesTea (Z) - IMNesTea. The easiest call to make by a huge margin. He's dominated his opponents so far. choya only got this far by getting drawn into by far the weakest group.
IMmvp (T) vs. TSLTrickster (P) - IMmvp. mvp has been getting better and better, and this doesn't seem to be an auspicious time for Protoss given their results in the group stages.
EGIdrA (Z) vs. LiquidJinro (T) - EGIdrA. Yes, Jinro beat him in the round of 32, but in a best of 5 I'll take IdrA on the grounds that ZvT is his best matchup. This will be fascinating since both players are so similar in their preferred play style. Incidentally, although it's a little frustrating they have to play each other, it does mean for the second tournament in a row there will be a foreigner in the semifinals.
MarineKingPrimeWE (T) vs. oGsNaDa (T) - oGsNaDa. I really like MarineKing, but NaDa has looked more impressive to me lately. Although MarineKing is more versatile than people give him credit for, it's true that he's best with infantry, and there seems to be a trend toward mech play in TvT lately.
The group stage setup is a little confusing at first, but if you think about it from GOM's point of view, it makes a lot of sense.
ReplyDeleteWhat they want:
1) Consistent Game numbers
2) Make every game count
3) High level of play across all of Code S
The seeded-Robin system (for lack of a better name) gives:
1) a slight advantage to the more consistent player, since winning Game 1 is huge
2) Eliminates most throw-away or "if I die, my teammate gets through" games. Also makes it much more like the NCAA basketball tournament where winning is everything.
3) Ensures you play at least 2 different players, thus making it far less likely a "bad" player gets through a group.
You could see this effect going on during the Game 5/6 in Ro32 for Code S. All of those games were tense and fought hard. Which is good for GomTV and good for SC2 fans in general.
Yeah. It's interesting for me as someone who (unlike a lot of SC2 fans) grew up watching normal sports to see how different the competition structures are in e-sports. I think it's because the games tend to be shorter and somewhat more variable to the best analogue (tennis, IMO) so you get double elimination tournaments like MLG, this modified group system, and the bizarre teamplay formats of Starcraft 1.
ReplyDeleteWell, even in the more common American sports, the issues that GOM was trying to avoid come up. Was in a little league tournament where we beat, in the loser's bracket, the team that sent us there. This issue is why MLG has that extended series "issue" that crops up. (And to save that discussion for another time, there's some merit to what they're trying to do, but I don't think it's a good thing, in total)
ReplyDeleteTennis is a pretty solid analogy. Play a "Set" well is going 6-3 or 6-2. 6-0 really isn't expected, because you should win your Service "Games" and lose your Returning games. But, the Tie-Breakers are very different in Tennis at the end of a Set.
I'm also of the impression there's several different tournament formats in Chess. (I notice the Chess/SC relation comes up a lot on the TeamLiquid forums, so figured I'd mention it). There's different ways of running the Chess clock and best of 1/3/5 and round robin setups, so it's not terribly settled there either.
I think once people get used to the GSL Round-Robin system, they'll come to like it. And when they get a better map pool. As we saw in the Ro32 Group H, if Jinro doesn't misplace his wall, there's no way he'd risk losing on Jungle Basin to a Zerg.