Why are Starcraft 2 tournaments so unpredictable? - Page 11

On February 03 2012 18:33 Hinanawi wrote:


Irrelevant data, it's like comparing your old high school's chess club winrates with international grandmasters' winrates.

If/when more than a couple actual good players start coming to SC2, expect those players' winrates to narrow down closer to 50%.

Remember kids, the best way to lie is with statistics.

On February 03 2012 18:33 Hinanawi wrote:


Irrelevant data, it's like comparing your old high school's chess club winrates with international grandmasters' winrates.

If/when more than a couple actual good players start coming to SC2, expect those players' winrates to narrow down closer to 50%.

Remember kids, the best way to lie is with statistics.

On February 03 2012 16:39 tree.hugger wrote:
Sc2 feels unpredictable for a few reasons, but I don't think they have a lot to do with how dumb and coin-flippy most match-ups can be. Even though Sc2 is basically designed to never allow any player to come back from a deficit, some players are simply better at getting that early lead than others, and so they do better. When you compare the volatility just by looking at the names, it doesn't really look more than BW.

To me, the crucial difference is that we don't have proleague running in the background. With proleague, we saw good players before they did well in the starleagues, because teams gained confidence in them, sent them out to lose a few times, and gradually they matured, got more comfortable, and then they did well in individual stuff. It almost never was the other way around. In Sc2 in Korea, we don't have that kind of system. GOM treats team competition like another GSL, and while the teams take it seriously, it just doesn't happen often enough to give us a glimpse and regular skill check on a bunch of players. And because individual Sc2 games can be so stupid all the time, often you don't even feel as though you got a good read on a player, because the game consisted of some dumb immortal all-in, or a 1/1/1.

In Europe however, the Sc2 scene is much more predictable, because most top pros still play weekly cups, and so we're constantly getting a reference of their skill level compared to that of other players. We can see up and coming players as they emerge, because they start beating better and better people every week. We can see who will never be any good because they just stay the same. We can see who is falling off. Weekly cups are basically the equivalent of proleague. It doesn't need to be a team competiton, just a regular way of assessing a player's skill. In Korea, with all these new tournaments coming out, we have a better sense of who is good in Korea than we did before. Players like TSL_Symbol, for example, have been able to prove they have redeeming qualities, despite being terrible in Code B.

So while Sc2 is a pretty silly game in pretty much every respect compared to it's predecessor, and way less balanced and rewarding than tennis, the difficulty of predicting Korean sc2 is mostly because we just don't see these players very often.

On February 03 2012 18:44 lorkac wrote:


See Wildmoon! What did I say? Numbers are obviously irrelevant--especially when it shows that I'm wrong about my opinions. The only acceptable numbers are whatever shows I'm right.

On February 03 2012 18:49 lorkac wrote:


You're obviously right!

When we compare winrates amongst top players in their respective fields to show similarities in the consistency of the winrates in those respective fields--they are obviously random and arbitrary comparisons like mountain ranges to hand guns

On February 03 2012 18:54 Hinanawi wrote:


You should really lay off the sarcasm, it just makes you sound like a douche. The point was that statistics can lie, as they obviously are here. The reason the comparison is invalid is because the current 'best players' were players that were good in BW and took advantage of the immaturity of the game and stomping bad foreigners (and bad Koreans) to inflate their win ratio. Flash's 72% is about a thousand times more impressive than MVP's 74% or whatever, because of the quality of his opponents, and therefore lends more legitimacy to the argument of volatility (or lack thereof) in the game.

Every pro in this thread is agreeing on some level that SC2 is more volatile and less skill-based than BW, but you've got 'statistics' and sarcasm on your side, and you're not going to look any further into it than that.

On February 03 2012 18:54 Hinanawi wrote:


You should really lay off the sarcasm, it just makes you sound like a douche. The point was that statistics can lie, as they obviously are here. The reason the comparison is invalid is because the current 'best players' were players that were good in BW and took advantage of the immaturity of the game and stomping bad foreigners (and bad Koreans) to inflate their win ratio. Flash's 72% is about a thousand times more impressive than MVP's 74% or whatever, because of the quality of his opponents, and therefore lends more legitimacy to the argument of volatility (or lack thereof) in the game.

Every pro in this thread is agreeing on some level that SC2 is more volatile and less skill-based than BW, but you've got 'statistics' and sarcasm on your side, and you're not going to look any further into it than that.

On February 03 2012 19:17 karpo wrote:


He didn't say he trusted the statistics fully. There's a difference between acknowledging a similarity in winrates and being all "STATISTICS LIE, totally irrelevant!".

Sure i think most will agree that Flash winrate are more impressive, considering the competition, but it still shows that SC2 doesn't have as volatile a setting as many claim. If that volatility will be worse if better players comes along is nothing more than speculation at this point.

On February 03 2012 03:55 Micket wrote:
People can call the game shit and still enjoy it. They truly retire from the game when it does become unfun. Idra actually wrote his letter of resignation out at the beginning of 2011, when he thought the game simply wasn't fun anymore. Playing steppes, delta quadrant, jungle basin, does that to you.

The vibe I get from most pros is that they don't enjoy sc2 all that much and treat it more like a job. I just haven't seen any sc2 pros except perhaps Nani, Idra and Huk that have showed the raw passion of what people like day 9 and Artosis had for BW. Players like Ret and Tyler just seem unmotivated most of the time, and probably practise half as much sc2 than they used to for bw.

On February 03 2012 04:04 Bagration wrote:


We all play the game: the game of life. It's a hell of a game, inherently imbalanced, but we play it anyway. It's a coinflip, and the situation we are born in could easily put us in a auto loss situation. Billions of people have jobs that they hate, but what can they do?

Sad to see idra post that, though he has had a rough week.

On February 03 2012 01:16 naut1c wrote:
I don't think it's unpredictable, i think it is as predictable as it should be. Maybe PvP and ZvZ is too fragile, but everything else is really ok.

You can never predict the outcome of a match 100%, but you can nearly always predict a good chance of the outcome because of the following factors:
- their skill-levels
- recent results overall
- matchup results in particular
- recent results against of the same player
- results on map against the same race
- recent activity (training amount)
- daily mood

The higher the 'best of - n' value goes, the more definitive the prediction value gehts. But you can never be 100% sure, and that is GOOD, and it is Entertaining, otherwise it would be JUST boring.

It is on the same predictable-level like many other sports.

The point is none of these women are marriable.

Pictures don't tell much about how attractive someone really is.

How can you know whether you're attracted to a person or not until you know

-Their religion, and respective piety
-Their family and family history
-Their socio-economic class
-Their artistic and literary talents
-Their sensitivity, romanticism and sentimentalism
-Their patriotism, provincialism, and rootedness in their native culture
-Their cooking, cleaning and other domestic abilities
-Their mental and emotional stability
-Their historical attitude and preferences regarding bourgeois family life
-Their racial origins (if not already clear)
-Their feminine virtues (virginity, modesty, etc)

The above listed constitute 95% of the qualifications for any marriable and therefore attractive girl.

On February 02 2012 23:36 IdrA wrote:
cuz its a shit game

On February 03 2012 19:17 karpo wrote:


He didn't say he trusted the statistics fully. There's a difference between acknowledging a similarity in winrates and being all "STATISTICS LIE, totally irrelevant!".

Sure i think most will agree that Flash winrate are more impressive, considering the competition, but it still shows that SC2 doesn't have as volatile a setting as many claim. If that volatility will be worse if better players comes along is nothing more than speculation at this point.

On February 03 2012 19:48 s4life wrote:


If I am playing against scrubs all the time and somehow I get a 90% w.r. it doesn't mean the game isn't volatile, it just means that my win rate is inflated. The proof of the game's volatility is that there is a minimal chance of Flash losing in a round of 32 in a OSL, but MPV, MC and Nestea losing in a round of 16 or 32 happened multiple times.

On February 03 2012 19:24 lorkac wrote:


Statistically speaking, the top players in SC2 are more stable than their broodwar counterparts. It's only by a few percentage points, and its from a smaller sample size. But if you want to get to brass tacks and count out all games played in their respective fields, the top tier in SC2 is dominating their fields much more than top BW players are dominating theirs.

Whether their dominance "counts" or is seen as qualitatively valid is subjective. The numbers don't lie as numbers.