On February 23 2013 03:27 TheRabidDeer wrote: Does anybody have any stats on the best players from bw in like, 2003? I mean even then, that is 5 years experience with BW, but I imagine itll be pretty different than stats from like 2008+
What you have to realize is that everyone didn't lose the 10+ years of learning to analyze RTS games they gained from BW, AOE, WaR3, ect.
When BW came out maximizing efficiency in macro and builds wasn't nearly at the level it was at when SC2 hit the scene. In the first 6 months of Sc2 we had thread after thread, of mathematical breakdowns of each race and it's optimization. We had lengthy analysis of maximizing economy, zerg expansion and pool timing were broken down scientifically before sc2 beta ended, terran expansion and mule timing also became set in stone very early (the only change now is we often see expansion 1st builds) but these were also popular in BW. Keep in mind, it took BW 5 years to reach that point in the meta, where it took Sc2 less than 2 years (and that is a big deal considering all of the terrain negating and crazy all ins that exist in Sc2 compared to BW.
I think BW was a virgin experience, no one really knew how to play an RTS when the original starcraft came out, the RTS games before it didn't have the complexity in econ or the unit diversity to warrant the depth of understanding that starcraft requires at top levels.
Any new mainstream RTS that ever is released, will be analyzed and understood in it's first year, better than BW was understood in it's first 5. Just look how long BW was out when players like iLuvoov rose up, think about how elementary his "revolutionary macro play" seems by today's standards. He started 1 rax expanding against zerg, and it changed everything, it took another 5 years after that when flash came along and started CC first all the time in BW. Starcraft 2 when through these changes in under 2 years, with constant and drastic balance changes.
Yes, this is true. This is also why we are currently at a point where we are close to the probably 2009 area of BW in terms of game knowledge (in WoL). However, a year ago even we were much much lower in skill.
Also, this rapid rapid change that we have undergone is another reason things seem so volatile. In BW, players saw a year of dominance. It took a lot of time for people to figure out what to do, then get it known throughout all of the players to be able to consistently beat them. Just look at the stats posted about boxer/iloveoov/Nada. They had stellar YEARS then fell off a bit as people learned how to play against them. Now compare that to SC2, people had stellar SEASONS then started to fall off. To go 4-5 months with a 70% win rate is the equivalent of a dominant year in BW just because of how quickly information and strategy travels.
Once the game is truly figured out, only then will we see a Flash in SC2. Flash was so dominant because he knew everything about the game, had great play and mechanics and everything. You cant have a dominant force when things are still changing so quickly.
True starcraft fans will never be the very best at game prediction as no matter how 'good' you are passion will always cloud your judgement. You are human after all.
However if you take purely gambling/profiting and objective approach then prediction become much easier. Then given you have inside information regarding player conditions, health and other internal factors that would affect the outcome of a games and taking everything into account in making a weight bet. Wasn't there some person who was showing off their pinnacle account statement with winnings over 10K+ net profits over the course of a month? They probably lost it all now as it was much easier to predict back in the day but its possible!
I wonder if anyone would like to do math, and do some probability calculations.
If you were to take 3100 coins, and each on of them 'makes a prediction' on every liquibet match (by being flipped). What is the highest guessing percentage we can expect? ^^
Federer is a bad example because domination like the current top 4 in tennis is quite rare. If you look at women's tennis or any other sport you will see that it's quite random and has lots of variance too..
Sc2 is fairly luck based make it quite variable i think but there are also some other factors.. Competition is tight and the way invites and ranking works you often have players of nearly equal skill meeting eachother. In other sports with set competitions like soccer or any american sport this is not the case at all because the teams vary much more in resources etc. Also the formats in sc2 are not really aimed at reducing this variability. There is almost no protection for higher ranked players like there is in other sports through seeding, bye's etc. There is hardly any large competition style formats either where you can offset losses easily. Also a single match is usually just BO3 with each game being fairly luck dependant, a tennis match in important tournaments is BO5 and each set is has many points each basically unrelated so just being a bit better causes a high win chance overall..
Basically you can protect higher tiers as much as you want by just making it longer series, adding protection for winners, playing more in competition form etc. The question is do you want this.. As long as the names are recognizable I think sc2 is in a fine spot. Tennis is too predictable for men's at the moment making it very boring. Stuff like poker and magic is too random with different guys winning all the time.. Sc2 is quite in the middle of those, at bit more random than champions league soccer i'd say but in that area. I would favor a slightly longer format for GSL though just so the top players don't get knocked out so randomly so often as they do now.. The game itself will just stay quite random but I think that's just the nature of the beast with hidden information playing such a big role as do other random things you don't have exact control over...
For example the lure of soccer is greatly in it's randomness I think, almost any game even the underdog has a chance and even during the game it often is still very much possible for both teams to improve their result within minutes making it so interesting to keep watching..
On February 23 2013 05:24 niteReloaded wrote: I wonder if anyone would like to do math, and do some probability calculations.
If you were to take 3100 coins, and each on of them 'makes a prediction' on every liquibet match (by being flipped). What is the highest guessing percentage we can expect? ^^
On February 23 2013 05:24 niteReloaded wrote: I wonder if anyone would like to do math, and do some probability calculations.
If you were to take 3100 coins, and each on of them 'makes a prediction' on every liquibet match (by being flipped). What is the highest guessing percentage we can expect? ^^
Just going to quickly throw out an idea that may have already been mentioned earlier in the thread - only read about half of it.
Is it not possible that at least part of the reason we see less consistently dominant players in SC2 as opposed to BW is that the game is constantly evolving (literally) through patches. BWs last even remotely significant patch was almost a decade ago iirc. Flash dominated in BW long after the game had 'officially' entered its final version. It was entirely up to other players to work out how to beat his style. That can take time. In SC2 if a player/race enters a period of dominance it has been, on at least a few occasions I can think of, ended abruptly/arguably prematurely by a balance 'fix'.
5 rax reaper was obviously ludicrously strong vs zerg early on in SC2s lifetime. It was patched. Does anyone honestly think that if this patch was not introduced zerg players of today would STILL be facing 5 rax reaper every game and find themselves completely oblivious as to how to stop it? I feel this gives far too little respect to the abilities of hard working professional players. It, and many other styles/builds/units were/are certainly very, very powerful - even overly so. That doesnt mean there isnt a way to stop it that can be found eventually. (Note here: I am not entirely sure that 5 rax reaper or any other patched strategy could have 100% been figured out - I do think its reasonable to suggest that theres at least a chance it could have been though. Theres some pretty smart people playing BW/SC2.)
BW has gone through many balance 'flips' in matchups when a new style has been figured out - even sometimes lasting for a year or more. A common misconception expressed when discussing BW (as opposed to SC2) is that all the matchups are completely balanced. This simply isnt the case in a way thats relevant to SC2s history. If you look at the BW matchups over 5 years or so, sure - the %s all come out to roughly 50% give or take a few % depending on the matchup. What you dont see is that at any point of time in BWs history generally each matchup is roughly a 60/40 in one direction or at some times even more skewed. Someone pioneering comes up with a new way to play a matchup or two, does very well for a period of time, other people of his race begin to copy him at which point the other races playerpools start to really go into the lab to figure out ways to beat these styles as theyre now running into them every game vs that race. Over time some progress is made and eventually pioneering player #2 comes along with the last bit of the puzzle, dethroning pioneering player #1 and we repeat.
In SC2s history, pioneering player #1 comes along, does really well and is then 'fixed' a couple of months or less down the line and they fade into obscurity. The playerpool resets with everyone having to learn/figure out fresh builds again and random new 'best of the current moment' players pop up and do well for a tournament or two. A large part of their success is the knowledge reset of the patch, so they too fade into obscurity when their new tricks are figured out within a couple of weeks. Other random people win the occasional tournament due to having little bits of knowledge other people dont have yet but it only works for one significant tournament because then youre studied and what you were doing really isnt that special/difficult to counter so long as you know about its existence. Eventually a potential Flash/iloveoov/savior equivalent comes along with something very powerful that is not easily countered within 1-2 weeks. They get fixed. Repeat. <1 month dominance periods by the occasional player with the rest of the time filled in with random dare I say flavour of the month players winning random tournaments.
SC2 has always felt pretty disappointing in many respects personally but I think its progress as a truly interesting strategy game has been stifled to at least some degree by zealous over patching. In a sufficiently complex game with interesting mechanics almost anything should be counterable. If 5 rax reaper or other builds are literally uncounterable the problem doesnt exist because of the strategy itself but in the lack of viable options given to the opposing player in general. If things like 5 rax reaper cause significant problems its due to the game not being complex/objectively 'interesting' enough. If thats the case with SC2 (Im not completely sure it is), fix the game overall. Dont just hotfix little parts of it and most definitely dont patch just because a bunch of people are whining about being unable to use their old strategies when a metagame shift occurs. The only strategy that should be nerfed directly through a patch is something that happens very, very early into the game (as in pre 2-3 minutes) such that the opponent has barely any options at this point and none of them work. The old Starcraft1 150 mineral pools leading to hilariously early 4 pools are a decent example in that they were essentially uncounterable by any strategy that didnt lose to everything else and they would happen at a stage in the game where utilizing scouting or anything remotely clever was not possible. By the time you get to strategies kicking in at the 5 minute point the amount of positions any player can be in with regards to economy/army/tech are VAST in both BW and SC2 and it should be up to the players to work out the position that works favorably while not putting themselves at a disadvantage vs other builds/styles. This isnt allowed to happen in SC2 for the most part and I think its a shame.
This has probably been pointed out but everyone comparing it to lifetime sports records of times is being foolish since teams are constantly changing. I'm sure there was a 3 year period where some football team had a gigantic win-rate. In this 3 year period of SC2 no one has stayed above the low 60s.
Isn't the assumption that blind picking on Liquibet would give you an average of 50% false, since for the GSL group picks, you have to pick the two players coming out of the group, giving you odds of 1 over (4 choose 2) or 1/6.
The inconsistency happens because if you lose a big fight in sc2 you lose the game in most cases, while in BW it doesn't necessarily mean that. Also, stuff dies/moves too fast and a lot of maps are/feel smaller compared to BW, while the chokes feel too big. No high ground/vision advantage invalidates a lot of positional/strategic/macro/defensive play as well. The unit composition countering might be too important as well. Also it's a lot easier to all-in.
Making a small mistake can cause a lot more damage in SC2 than BW I feel, in general.
On February 23 2013 07:46 BrokenMirage wrote: Isn't the assumption that blind picking on Liquibet would give you an average of 50% false, since for the GSL group picks, you have to pick the two players coming out of the group, giving you odds of 1 over (4 choose 2) or 1/6.
The expectancy value of points you get when picking at random will still be 50% of the maximum amount.
I'm not sure that's correct. Each group seems to be worth 1 point if you get it right. At random I think you have a 1/6th chance of getting the group right. So if you only bet on the GSL groups blind then you would be at about 16.6%?
I'm pretty sure most of the counter points to this line of thinking have already been raised.
But let me just say one thing: StarCraft CANNOT be compared to traditional sports. SC2 has an evolving meta-game influenced by many agents: Blizzard constantly changes the balance of the game by tweaking units. Maps change. And of course, optimal builds change based on those changes.
Basketball and football only change slightly insofar as players change; but players are only ever slightly better or worse than one another. Even 20 Michael Jordans per season does not change the game that much.
It's a huge leap of logic/category mistake to suggest the predictability of a game like this should be any different than what it is.
On February 23 2013 07:46 BrokenMirage wrote: Isn't the assumption that blind picking on Liquibet would give you an average of 50% false, since for the GSL group picks, you have to pick the two players coming out of the group, giving you odds of 1 over (4 choose 2) or 1/6.
The expectancy value of points you get when picking at random will still be 50% of the maximum amount.
I'm not sure that's correct. Each group seems to be worth 1 point if you get it right. At random I think you have a 1/6th chance of getting the group right. So if you only bet on the GSL groups blind then you would be at about 16.6%?
Well, that would be a very odd way of calculating the percentage of correct liquibets. My guess is that guessing two players correctly to advance contributes double that of guessing one player correctly.
If you only get one player right, you didn't get it right, and you don't get any points.