|
Looking at the ELO charts of SC1 and SC2 players, it is visible that the chart of SC2 has less dynamics - the scores of the best players are close to those of the worst players, when compared to the chart of BW.
Different formula? different structure of tournaments or number of players(this one seems unlikely given low ranking players having equal ELO in both games), meaning the charts are mismatched? or is the diehard brood war fans' criticism of lack of skill curve/ceiling in sc2 really true? is the brood war pro player skill ladder really more dynamic than that of sc2?
![[image loading]](http://i.imgur.com/eAwaE.png)
While it is true that SC2 seems to have the dominant player - MVP - his Elo is not nearly as much differentiated from the bottom end players than that of BW dominating players Bisu and Flash. And perhaps more significantly, the top 20 players of BW seem to have a consistently bigger score over the low ranking players, compared to those of SC2.
  
|
Why is it that when I read
More proof of the elephant?
by someone named
BroodWarHD
I knew what was coming 
Do we really have to rehash the same arguments over and over again? The problem with using your example is that for a player to consistently dominate the game has to be relatively stable. This has not been the case with the constant patching and the newness of the game. It just hasn't been developed yet.
|
|
|
|
|
Ah yeah my bad, I wasn't thinking.
|
No it's just a matter of dilatation of the ELO ranks the more games are played (since everyone starts at 2000, you gain much points when you win compared to when you play against a 2200+ opponent (say when you are S-class BW level). You can check the ELO graph restricted to a map (so there will be a lot less games played) and you will see it is much more contracted. You can also see it in the ELO of the vX matchup: they are generally 100 points lower than the general ELO.
|
On November 28 2011 09:20 Probulous wrote:Why is it that when I read by someone named I knew what was coming Do we really have to rehash the same arguments over and over again? The problem with using your example is that for a player to consistently dominate the game has to be relatively stable. This has not been the case with the constant patching and the newness of the game. It just hasn't been developed yet.
is it? i would think a stable environment with less information asymmetry would imply more similar conditions for players to perform, leading to more equal results compared to the fluctuations and changes of sc2.
and is it a rehashed argument? the data in the chart is less than a hour old, surely not rehashed by itself, as the concept of its statistics based presentation would imply, compared to the more descriptive/winrate based style intrigue used in his article.
|
Stunning ad-hoc analysis based on eye-balling an excel chart. -_-
|
Top BW players play many more games than the other 90%, due to more PL appearances, having more games in individual leagues, etc.
|
Yeah BW rocks, SC2 sucks. Who cares man, SC2 is fun to watch and is competitive, or at least so it seems.
As more games are played in SC2 by each players, ELO ratings will start dispersing, I think.
|
gondolin, isnt that argument only relevant in very low sample sizes? given the 100s of games both league players have played, it should no longer be relevant, would it?
|
No-one cares.
User was warned for this post
|
He isn't arguing the volume of games of individual players but the duration of time elapsed since the games release.
|
What good can possibly come out of this?
|
On November 28 2011 09:28 BroodWarHD wrote:Show nested quote +On November 28 2011 09:20 Probulous wrote:Why is it that when I read More proof of the elephant? by someone named BroodWarHD I knew what was coming Do we really have to rehash the same arguments over and over again? The problem with using your example is that for a player to consistently dominate the game has to be relatively stable. This has not been the case with the constant patching and the newness of the game. It just hasn't been developed yet. is it? i would think a stable environment with less information asymmetry would imply more similar conditions for players to perform, leading to more equal results compared to the fluctuations and changes of sc2. and is it a rehashed argument? the data in the chart is less than a hour old, surely not rehashed by itself, as the concept of its statistics based presentation would imply, compared to the more descriptive/winrate based style intrigue used in his article.
Yes, it would. As you say you would expect more equal results, thus someone with an above average skill level would be able to dominate more because all other things are more equal. The problem is right now good players can lose because of that "information assymetry". One player may have been able to practice more with the new patch simply because they haven't been travelling (or something). Thus they have more exposure to the current state of the game for no reason other than an outside influence (patch). This is not what you want in a competitive environment.
What you want is a stable environment where skill and imagination lead to wins, not outside influences.
I may be misunderstanging your argument. I am no statistics expert but surely a more equal playing field allows actual skill to determine wins.
Yes the argument is rehashed because it comes down to the same thing. You are comparing different games at different stages in their development. You are always going to get different results.
|
On November 28 2011 09:41 Probe1 wrote:
He isn't arguing the volume of games of individual players but the duration of time elapsed since the games release.
probe, that would appear to be incorrect as there doesnt seem to be general elo inflation - the low ranking players still have a similar score in both games.
and to give an example of sufficient number of games - it takes ~20 games with 70% winrate to go from 2000 to2300 elo, assuming tl uses 32 maximum K value. seems plausible number in the ladders
|
I'm curious what your intentions are of posting this.
I'm a 99% Broodwar 1% SC2 spectator/player, and I love BW to death, but there's really no good discussion going to come out of this blog.
You're probably one of those people that gets all crybaby and upset about people hurting the feeling of your preferred game, then turn around and try and make all sorts of arguments as to why SC2 is terrible compared to BW.
How about instead of trying to get back at the SC2 fans that are all about "lol shitty 95 graphics" "no MBS useless game" "terrible interface and controls", we ignore them and not become our own version of them.
|
Brood war out for over a decade, SC2 out for a little over a year...my head hurts
|
No, no he was directly saying "SC2 is in it's infancy and BW is a matured game with years upon years of knowledge and experience". Or if he's not saying that- I am.
Apples and Oranges. Of course Flash with a (# of month duration depending on how big of a KT Rolster fan you are) to practice would dominate SC2 in a way we can barely comprehend. Yes, the best player in SC2 is a lot closer to the lower player than in BW. But it's to do with many factors- one of them being bw has a long and deep history and SC2 is just starting its second year.
Regardless, thank you for compiling a few statistics.
|
I don't think "dynamic" is the word you're looking for... it means changing. If anything, you could say sc2 is too dynamic- susceptible to swings/upsets.
Given that, I agree with you.
|
Australia8532 Posts
So a decade old game that has been picked over, discussed, played and strategised over for so many years provides more dominant players than a new, constantly patched, somewhat erratic and volatile game?
Shit. I should have majored in statistics
|
im with you on this one OP, youve got the right idea
|
It's just inflation. When a player retires, his rating is removed from TLPD. If the rating of a retiring player is low enough, the average rating of the active player pool will increase with his retirement. As games are played, this extra rating will eventually end up in the hands of the top players. Naturally there are many more retired BW players than retired SC2 players, so the curve will be steeper in BW.
|
On November 28 2011 10:29 bkrow wrote:
So a decade old game that has been picked over, discussed, played and strategised over for so many years provides more dominant players than a new, constantly patched, somewhat erratic and volatile game?
Shit. I should have majored in statistics
bw has almost always had dominant players, not one per week but consistently dominant ones
Plus I think it's pretty fair to say that sc2 has been more figured out in about a year than bw was for it's first 2-3 years at the very least. We have like 1000 pro's playing as a profession from the get go and more players overall, more discussion and more competitions.
|
On November 28 2011 10:34 hifriend wrote:Show nested quote +On November 28 2011 10:29 bkrow wrote:
So a decade old game that has been picked over, discussed, played and strategised over for so many years provides more dominant players than a new, constantly patched, somewhat erratic and volatile game?
Shit. I should have majored in statistics bw has almost always had dominant players, not one per week but consistently dominant ones Plus I think it's pretty fair to say that sc2 has been more figured out in about a year than bw was for it's first 2-3 years at the very least. We have like 1000 pro's playing as a profession from the get go and more players overall, more discussion and more competitions.
Try more like 4 or 5 years. It took quite a while before players really started to realize how important macroing and multitasking was. It really wasn't until Nada that builds like fac-ccing became quite standard (it was a build used before Nada, but not used most games), and in TvZ, terrans often stayed on one base until the majority of their tech tree was filled out. Really, the way we think of SC2 now was far far far more advanced than even the way we thought of Sc1 when Boxer was the dominant player.
Anyone who played both SC1 and SC2 will tell you that the skill discrepancies between the highest end players, the mid end players, and even the lower end players was a lot bigger in SC1 than it is right now in SC2, but I don't think that has anything to do with whether one is the better game or not at all. I don't even see the point in discussing it, as there's no real meaningful conclusions that you can make from that.
|
So.... there is a 30-greater ELO difference in the top 10 percent of the BW players than SC2 players. and MVP has the Same gap above the next best players than bisu has. and this is supposed to say something about BW being better than SC2? if anything it proves players in SC2 can be as dominant in sc2 as bisu is in BW and that no one is as good as flash period. also please note the huge differences in the proleague -team league ruleset. a player is less likely to be dominant in sc2 teamleague than proleague as there are easy snipers allowed in sc2, so you can pick on a players bad matchup (hurting his elo) where in proleague oftentimes you send players out in specific matchups and they only have to play the matchup they prepare for.
|
It's Elo.
The Elo rating system named after Arpad Elo, a chess grandmaster.
Not the Electric Light Orchestra rating system.
|
It's pretty standard knowledge that the nature of SC2 actively minimizes "skill" disparity. This graph is pretty good evidence of that. Please don't confuse Skill disparity" with "skill ceiling" or i'll shoot you with an e-bullet.
|
Is there even a point to this thread? Was anyone arguing that SC2 was a better game or something? Anyone who followed BW knows it was a better game. I fail to see the purpose of this thread other then to tell us what we already know.
|
On November 28 2011 10:54 InToTheWannaB wrote:
Is there even a point to this thread? Was anyone arguing that SC2 was a better game or something? Anyone who followed BW knows it was a better game. I fail to see the purpose of this thread other then to tell us what we already know.
alot of people find it cool to make fun of sc2 at every opportunity.
|
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers.
|
|
|
Well to be fair Starcraft didn't have much of a scene until Broodwar.Broodwar had its fair share of time and SC2 is barely 1 year old.The wind of change is coming.
|
If you also graphed the SC2 International Elos, it'll actually exceed the BW distribution. Does this mean SC2 have higher skill ceiling?
:D Of course not, you can't compare two set of Elo rankings that has zero interface.
|
On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers.
Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1.
On November 28 2011 11:03 Primadog wrote:
If you also graphed the SC2 International Elos, it'll actually exceed the BW distribution. Does this mean SC2 have higher skill ceiling?
:D Of course not, you can't compare two set of Elo rankings that has zero interface.
The skill ceiling point is null and always has been... this isn't at all about skill "ceiling."
|
On November 28 2011 11:04 ShadeR wrote:Show nested quote +On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1.
regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones.
but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league.
|
I don't think it's cause SC2 is worse, I think it's cause BW has been played for a decade, has been stable (gameplay/unit ability wise) and so has been developed for a decade. SC2 is new, still being patched every month or so, and so has a changing metagame and complete strategies. This means that players will rise and fall in ELO as their styles become more or less in favor (see players like TOP, MC to an extent), and not because there isn't a skill differential- the game needs to be stable for that.
|
On November 28 2011 11:04 ShadeR wrote:Show nested quote +On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. Show nested quote +On November 28 2011 11:03 Primadog wrote:
If you also graphed the SC2 International Elos, it'll actually exceed the BW distribution. Does this mean SC2 have higher skill ceiling?
:D Of course not, you can't compare two set of Elo rankings that has zero interface. The skill ceiling point is null and always has been... this isn't at all about skill "ceiling."
Plot:
![[image loading]](http://i.imgur.com/t3Cib.png)
|
On November 28 2011 11:09 PrinceXizor wrote:Show nested quote +On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league.
And you bring up bronze level tactics because professional play doesn't support you point or...?
|
On November 28 2011 09:10 BroodWarHD wrote:
or is the diehard brood war fans' criticism of lack of skill curve/ceiling in sc2 really true?
This statement right here insisted that I post here...
I'd recommend that you take more time in wording your blog post. A lack of skill ceiling would imply that the game has an endless ceiling, therefore endless cap, therefore endless potential. Pretty sure they argue that the ceiling is much smaller than that of BW, not that it has a "lack of skill ceiling" (which would imply that sc2 has more skill). Like others have said your comparison of both games does not take into consideration the length of time both games have been around.
|
i'm assuming that blue dots means SC2 and pink dots means BW right?
|
On November 28 2011 11:12 Primadog wrote:Show nested quote +On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. On November 28 2011 11:03 Primadog wrote:
If you also graphed the SC2 International Elos, it'll actually exceed the BW distribution. Does this mean SC2 have higher skill ceiling?
:D Of course not, you can't compare two set of Elo rankings that has zero interface. The skill ceiling point is null and always has been... this isn't at all about skill "ceiling." Plot:
I'm not aiming what I say at the quoted post, but merely co-opting the graph. To me, if you take out the outliers in the BW graph, then the three graphs look roughly the same to me, which would seem to prove nothing. If the only argument that survives this is that BW has more outliers at the top end of the ELO, I don't think that proves anything about the game itself. There's no direct link between the skill ceiling of a game and the quantity of outliers at the top of the rating system for that game.
The presence of a Michael Jordan, Tiger Woods, or Flash should not speak to anything other than the fact that these individuals are brilliant in their field. It certainly doesn't speak to the difficulty of Basketball, Golf, or BW, at least.
EDIT:
On November 28 2011 11:04 ShadeR wrote:Show nested quote +On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1
You assume too much. I wasn't saying to compare SC2 year 1 with BW (or SC more realistically) year 1. What I'm saying is that we should know what affects time has had on the graph of BW. If we know how the chart has evolved for BW over time, then we can perhaps begin to develop some expectations on what SC2 should and perhaps will look like over time. If BW has actually gotten less spread over time, then that's informative. If BW has gotten a larger spread over time, then that's informative too. If it's stayed relatively the same, then that's important in its own right.
You can't just throw up two sets of numbers and assume they will be equal. You need to prove WHY they should be equal first and foremost. There should not be this implicit acceptance that two completely arbitrarily pulled data sets should just happen to be the same.
|
I agree with you Takkara. The graphs themselves are completely meaningless even if two seperate Elo system are directly comparable. They're not.
It's completely disingenuous what was claimed in the OP. That can only come from either poor understanding of how a ranking system work, or intentional nitpicking of data to support his personal beliefs. This is why I re-plot the data - to point out that an extension of his logics actually counters his assertions.
|
United States7639 Posts
The highest point on the BW data is actually Bisu, not Flash. And as flamewheel pointed out in the last PR, his elo is artificially inflated due to the fact that he mostly pubstomps through PL, and does not face the tough competition found in the upper levels of individual leagues (damn you Hyvaa T-T). This is just one example of the point others have been trying to make: that you can't make direct comparisons between a record built from PL + OSL/MSL (or lack thereof) and one that is mostly just GSL.
|
ELO doesnt mean anything in SC2 because so many players play in small tournaments and get their wins there. BW had only OSL/MSL/PL so you were basically playing against players withing a certain skill range (super high)
|
Yes skill gap between top tier pros and lower tier pros is high in Brood War, and relatively low in SC2. It's high enough to not be a joke though. Don't worry too much about it.
I've watched more violations in game in Brood War, where one player just got completely pooped on than in SC2, that's for sure.
|
SC2 is all about mechanics like macro that are easily maxed out and strategy. In BW, extreme micro could highlight the difference between noobs and masters whilst also giving the masters a way to crush noobs despite a strategical disadvantage. In other words, BW gave a tool to the truly good players that the noobs couldn't use to crush them despite disadvantages. SC2 has no such tool.
Another facet of this issue is that SC2 is not where BW is, when a series of true masters have already fleshed out all facets of the game, leaving almost only mechanics to be mastered by pros. SC2 players are still learning new stuff, not to mention the patches and expansions coming out.
I suppose the biggest issue, though, is we have had none of the aforementioned "true masters" play this game. As such, the lack of skill gap is pretty understandable.
|
On November 28 2011 11:12 ShadeR wrote:Show nested quote +On November 28 2011 11:09 PrinceXizor wrote:On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league. And you bring up bronze level tactics because professional play doesn't support you point or...?
... wat? i was using BW builds. specifically pointing out that a very very common BW strategy in year 1 was the 4 pool. didn't mention sc2 builds at all. sorry.
|
OP, I think there are some folks who just don't enjoy watching SC2. I don't know if you are one of them, but this is the impression of course everyone has gotten. I've gotta admit I find SC2 boring. The reason we find it frustrating that other people like something we don't like is because on our big BW site we see tonnes of content being generated for a game we don't care about. Whether it makes us wish those users were creating content for BW or we just don't want to have see stuff for which we are not part of the demographic I don't honestly know. But let's play nice in the sandbox because there are people who enjoy SC2, for whatever reason, and maybe they'd enjoy it even if there was no element of skill difference between pros too. In the end you can't really prove to someone why they shouldn't like different things from you. You can only say why you don't like it yourself (and unless someone asked, they probably don't want to hear it).
|
United States1719 Posts
Cmon man, even as an outspoken BW fan, what you're doing isn't classy. As an earlier poster said, no good can come out of a blog like this.
|
On November 28 2011 13:33 PrinceXizor wrote:Show nested quote +On November 28 2011 11:12 ShadeR wrote:On November 28 2011 11:09 PrinceXizor wrote:On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league. And you bring up bronze level tactics because professional play doesn't support you point or...? ... wat? i was using BW builds. specifically pointing out that a very very common BW strategy in year 1 was the 4 pool. didn't mention sc2 builds at all. sorry.
Are you deliberately missing the point here? I don't know, in fact after rereading your first paragraph it becomes more and more incomprehensible, what point are you trying to make what in my post are you replying to? 4pool very common BW start in year 1? Your just making things up now....
Your just continuing to peddle the misleading comparison between early BW and early SC2.
Edit: back on topic the OP is simply an analysis of data and you can disagree with his analysis and provide your own perspective, but i just get annoyed when a parallel gets drawn between early BW and SC2, it's simply wrong.
|
Could the OP maybe add a conclusion? It seems like you have an introduction, a few questions, and then a bit of data analysis. What exactly is your purpose and what exactly does the information mean? I think that would clear up a lot of confusion.
|
On November 28 2011 14:29 ShadeR wrote:Show nested quote +On November 28 2011 13:33 PrinceXizor wrote:On November 28 2011 11:12 ShadeR wrote:On November 28 2011 11:09 PrinceXizor wrote:On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league. And you bring up bronze level tactics because professional play doesn't support you point or...? ... wat? i was using BW builds. specifically pointing out that a very very common BW strategy in year 1 was the 4 pool. didn't mention sc2 builds at all. sorry. Are you deliberately missing the point here? I don't know, in fact after rereading your first paragraph it becomes more and more incomprehensible, what point are you trying to make what in my post are you replying to? 4pool very common BW start in year 1? Your just making things up now.... Your just continuing to peddle the misleading comparison between early BW and early SC2. Edit: back on topic the OP is simply an analysis of data and you can disagree with his analysis and provide your own perspective, but i just get annoyed when a parallel gets drawn between early BW and SC2, it's simply wrong.
I said that the first three months of SC2 were like the first year of BW, in that cheesier strategies prevailed and both games had to deal with a changing metagame inn patches, how is that NOT a parallel.
in early BW, 4 pool was a dominant zerg strategy, its kind of a fact.
|
On November 28 2011 14:47 PrinceXizor wrote:Show nested quote +On November 28 2011 14:29 ShadeR wrote:On November 28 2011 13:33 PrinceXizor wrote:On November 28 2011 11:12 ShadeR wrote:On November 28 2011 11:09 PrinceXizor wrote:On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league. And you bring up bronze level tactics because professional play doesn't support you point or...? ... wat? i was using BW builds. specifically pointing out that a very very common BW strategy in year 1 was the 4 pool. didn't mention sc2 builds at all. sorry. Are you deliberately missing the point here? I don't know, in fact after rereading your first paragraph it becomes more and more incomprehensible, what point are you trying to make what in my post are you replying to? 4pool very common BW start in year 1? Your just making things up now.... Your just continuing to peddle the misleading comparison between early BW and early SC2. Edit: back on topic the OP is simply an analysis of data and you can disagree with his analysis and provide your own perspective, but i just get annoyed when a parallel gets drawn between early BW and SC2, it's simply wrong. I said that the first three months of SC2 were like the first year of BW, in that cheesier strategies prevailed and both games had to deal with a changing metagame inn patches, how is that NOT a parallel. in early BW, 4 pool was a dominant zerg strategy, its kind of a fact.
Certainly not. People knew all about and played macro games from beta.
You are also wrong about 4pool. And you have no means of proving your "fact" that 4pool was a dominant professional strategy. Your probably confused about when spawning pools were 150 minerals.
|
On November 28 2011 15:13 ShadeR wrote:Show nested quote +On November 28 2011 14:47 PrinceXizor wrote:On November 28 2011 14:29 ShadeR wrote:On November 28 2011 13:33 PrinceXizor wrote:On November 28 2011 11:12 ShadeR wrote:On November 28 2011 11:09 PrinceXizor wrote:On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league. And you bring up bronze level tactics because professional play doesn't support you point or...? ... wat? i was using BW builds. specifically pointing out that a very very common BW strategy in year 1 was the 4 pool. didn't mention sc2 builds at all. sorry. Are you deliberately missing the point here? I don't know, in fact after rereading your first paragraph it becomes more and more incomprehensible, what point are you trying to make what in my post are you replying to? 4pool very common BW start in year 1? Your just making things up now.... Your just continuing to peddle the misleading comparison between early BW and early SC2. Edit: back on topic the OP is simply an analysis of data and you can disagree with his analysis and provide your own perspective, but i just get annoyed when a parallel gets drawn between early BW and SC2, it's simply wrong. I said that the first three months of SC2 were like the first year of BW, in that cheesier strategies prevailed and both games had to deal with a changing metagame inn patches, how is that NOT a parallel. in early BW, 4 pool was a dominant zerg strategy, its kind of a fact. Certainly not. People knew all about and played macro games from beta. You are also wrong about 4pool. And you have no means of proving your "fact" that 4pool was a dominant professional strategy. Your probably confused about when spawning pools were 150 minerals.
Please rewatch Open season 1 and look at the runner ups games throughout the event.
also please realize BW came out in 1998. year one was 1998-1999. when spawning pools were 150 minerals, and 4 pools were crazy relavent to play. now realize that professional BW didn't really take off until 2001, after the release of 1.08. now realize i didn't say a damn thing about professional BW when i said year one of BW. i meant year one of brood war, the horrible imbalanced, stupid to play year one of brood war.
|
On November 28 2011 18:05 PrinceXizor wrote:Show nested quote +On November 28 2011 15:13 ShadeR wrote:On November 28 2011 14:47 PrinceXizor wrote:On November 28 2011 14:29 ShadeR wrote:On November 28 2011 13:33 PrinceXizor wrote:On November 28 2011 11:12 ShadeR wrote:On November 28 2011 11:09 PrinceXizor wrote:On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league. And you bring up bronze level tactics because professional play doesn't support you point or...? ... wat? i was using BW builds. specifically pointing out that a very very common BW strategy in year 1 was the 4 pool. didn't mention sc2 builds at all. sorry. Are you deliberately missing the point here? I don't know, in fact after rereading your first paragraph it becomes more and more incomprehensible, what point are you trying to make what in my post are you replying to? 4pool very common BW start in year 1? Your just making things up now.... Your just continuing to peddle the misleading comparison between early BW and early SC2. Edit: back on topic the OP is simply an analysis of data and you can disagree with his analysis and provide your own perspective, but i just get annoyed when a parallel gets drawn between early BW and SC2, it's simply wrong. I said that the first three months of SC2 were like the first year of BW, in that cheesier strategies prevailed and both games had to deal with a changing metagame inn patches, how is that NOT a parallel. in early BW, 4 pool was a dominant zerg strategy, its kind of a fact. Certainly not. People knew all about and played macro games from beta. You are also wrong about 4pool. And you have no means of proving your "fact" that 4pool was a dominant professional strategy. Your probably confused about when spawning pools were 150 minerals. Please rewatch Open season 1 and look at the runner ups games throughout the event. also please realize BW came out in 1998. year one was 1998-1999. when spawning pools were 150 minerals, and 4 pools were crazy relavent to play. now realize that professional BW didn't really take off until 2001, after the release of 1.08. now realize i didn't say a damn thing about professional BW when i said year one of BW. i meant year one of brood war, the horrible imbalanced, stupid to play year one of brood war.
I mentioned professional scene several replies up and only now you take the time to clarify that you weren't talking about the pro scene at all? In the context of these types of discussions the pro scene is the only one relevant. Whatever I'm done.
|
On November 28 2011 18:16 ShadeR wrote:Show nested quote +On November 28 2011 18:05 PrinceXizor wrote:On November 28 2011 15:13 ShadeR wrote:On November 28 2011 14:47 PrinceXizor wrote:On November 28 2011 14:29 ShadeR wrote:On November 28 2011 13:33 PrinceXizor wrote:On November 28 2011 11:12 ShadeR wrote:On November 28 2011 11:09 PrinceXizor wrote:On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. regardless of play style knowledge, year one and month 1-3 of BW and SC2 were really similar, alot of cheesy crap that takes no account for the late game, games always evolve this way, it's just natural. if you can win consistently with a 4 pool why would you 12 hatch until people start beating your 4 pool so much. same kind of thing happens in sc2 except it took less time to get past that point. And besides THAT year 1 BW and year 1 SC2 are very similar in that they are inconsistant in terms of the rules of the game, people kept having to adjust the patch changes in both year ones. but yeah only thing the chart showed me was that it's perfectly possible to be as dominant in sc2 as people are in bw, and that the player who has stayed out of teamleague more than anyone, has a higher ELO because he doesn't have a forced loss in his worst matchup like the teamleague currently forces on alot of very good players. but yeah the chart only proves to me that sc2 has the same potentials that BW has. so thanks op for proving it with data. and the minor difference in ELO in the top is likely due to the system of the sc2 team league. And you bring up bronze level tactics because professional play doesn't support you point or...? ... wat? i was using BW builds. specifically pointing out that a very very common BW strategy in year 1 was the 4 pool. didn't mention sc2 builds at all. sorry. Are you deliberately missing the point here? I don't know, in fact after rereading your first paragraph it becomes more and more incomprehensible, what point are you trying to make what in my post are you replying to? 4pool very common BW start in year 1? Your just making things up now.... Your just continuing to peddle the misleading comparison between early BW and early SC2. Edit: back on topic the OP is simply an analysis of data and you can disagree with his analysis and provide your own perspective, but i just get annoyed when a parallel gets drawn between early BW and SC2, it's simply wrong. I said that the first three months of SC2 were like the first year of BW, in that cheesier strategies prevailed and both games had to deal with a changing metagame inn patches, how is that NOT a parallel. in early BW, 4 pool was a dominant zerg strategy, its kind of a fact. Certainly not. People knew all about and played macro games from beta. You are also wrong about 4pool. And you have no means of proving your "fact" that 4pool was a dominant professional strategy. Your probably confused about when spawning pools were 150 minerals. Please rewatch Open season 1 and look at the runner ups games throughout the event. also please realize BW came out in 1998. year one was 1998-1999. when spawning pools were 150 minerals, and 4 pools were crazy relavent to play. now realize that professional BW didn't really take off until 2001, after the release of 1.08. now realize i didn't say a damn thing about professional BW when i said year one of BW. i meant year one of brood war, the horrible imbalanced, stupid to play year one of brood war. I mentioned professional scene several replies up and only now you take the time to clarify that you weren't talking about the pro scene at all? In the context of these types of discussions the pro scene is the only one relevant. Whatever I'm done.
i never once mentioned the professional scene, actually you didn't either, until you said that i said it. which is weird. really a good thing you are done.
|
To quote my statistics teacher: "I don't believe in a statistic I myself haven't manipulated"
This proves nothing
|
|
|
|
|
On November 28 2011 10:33 Teddyman wrote:
It's just inflation. When a player retires, his rating is removed from TLPD. If the rating of a retiring player is low enough, the average rating of the active player pool will increase with his retirement. As games are played, this extra rating will eventually end up in the hands of the top players. Naturally there are many more retired BW players than retired SC2 players, so the curve will be steeper in BW.
Isnt that a zero sum course of events? In order for someone to lose, someone else has to win. And in the end - both players will quit, the winners and the losers. The winners might have longer careers with more overall games, but to balance that off, there must be more losers than winners.
On November 28 2011 11:12 Primadog wrote:Show nested quote +On November 28 2011 11:04 ShadeR wrote:On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers. Oh please to do that was be purposely dishonest. Are you to say that SC2 began with no knowledge of all the modern BW management concepts? I'm sick and tired of people making the false equivalence between BW year 1 and SC2 year 1. On November 28 2011 11:03 Primadog wrote:
If you also graphed the SC2 International Elos, it'll actually exceed the BW distribution. Does this mean SC2 have higher skill ceiling?
:D Of course not, you can't compare two set of Elo rankings that has zero interface. The skill ceiling point is null and always has been... this isn't at all about skill "ceiling." Plot:
The international SC2 ranking does appear to have what could be called ELO inflation, but if one normalizes the SC2 ranking data to the same range of SC2 korean/BW korean, then it could be comparable. The SC2 international appears to have an overall smaller dynamic range than both the other lists.
On November 28 2011 10:56 Takkara wrote:
One major variable that is not being adjusted for in this "analysis" is the effect of the different league styles on the data. For BW we have PL, MSL, and OSL. For SC2 we have mostly GSL. The structure of PL in particular is vastly different than anything that exists in SC2. Because the ways that the gamers are amassing ELO is different, we can't directly compare the relative values of high and low ELO and assume the two games will be the same if they are equally skilled.
There's so much noise in the "data" that you can't draw any conclusions at all from what you've presented. Not the least of which you need to prove that there is a connection between larger gaps in ELO between pro players and the skill required to play the game. There's so many other, logical explanations both for why such a gap would exist and why it would not exist that it doesn't clearly follow from your chart.
Secondly you need to find a way to normalize the ELO between the two games to turn your apples and oranges comparison into an apples to apples comparison. Of course, finding this normalization is probably not easy and may not even be solvable.
Thirdly, you'd be more intellectually honest to compare such ELO data 5-10 years ago for BW to see what, if any, effects time has had on such ELO graphs with just BW. It would be important to note if BW has gotten more distinguished, less, or stayed relatively the same.
There's other issues, but this is a good start if you honestly want to try to prove something with your numbers.
Doesnt the "noise" you describe actually define what we see in the games as skill? Because that is what the players have to do, their mission to play in the conditions that the league presents them. And those who are the most successful at that, could be described as being the most skillful. IE Bisu being very skillful as he plays superbly in proleague despite lackluster individual league games.
One can solve the normalization problem by looking at the actual dynamic range of the Elo rankings, rather than the absolute values. Nonetheless, this appears to not be an important issue as the BW and SC2 korean charts appear to be in the same range of absolute values.
The age of the game and metagame development differences may be a relevant argument, but it is so hard to quantify in numbers that i cant imagine how one would include it in the analysis
|
On November 28 2011 10:44 ColdLava wrote:Show nested quote +On November 28 2011 10:34 hifriend wrote:On November 28 2011 10:29 bkrow wrote:
So a decade old game that has been picked over, discussed, played and strategised over for so many years provides more dominant players than a new, constantly patched, somewhat erratic and volatile game?
Shit. I should have majored in statistics bw has almost always had dominant players, not one per week but consistently dominant ones Plus I think it's pretty fair to say that sc2 has been more figured out in about a year than bw was for it's first 2-3 years at the very least. We have like 1000 pro's playing as a profession from the get go and more players overall, more discussion and more competitions. Try more like 4 or 5 years. It took quite a while before players really started to realize how important macroing and multitasking was. It really wasn't until Nada that builds like fac-ccing became quite standard (it was a build used before Nada, but not used most games), and in TvZ, terrans often stayed on one base until the majority of their tech tree was filled out. Really, the way we think of SC2 now was far far far more advanced than even the way we thought of Sc1 when Boxer was the dominant player. Anyone who played both SC1 and SC2 will tell you that the skill discrepancies between the highest end players, the mid end players, and even the lower end players was a lot bigger in SC1 than it is right now in SC2, but I don't think that has anything to do with whether one is the better game or not at all. I don't even see the point in discussing it, as there's no real meaningful conclusions that you can make from that.
CL nicely put.
I love how these threads keep popping up.
|
BW elitist here. Even I don't see the point of this thread. Who cares if there is or isn't a greater relative difference between the ELO of these two games? You've got to be shitting me if this is a reason why you think BW is better.
|
I find it really funny that the elephant was originally used (IIRC) to "show" how terrible a game SC2 was compared to BW, but the TL staff managed to commandeer it into being a far cuter, far nicer "Elly the ESPORTS elephant", and it now shows on the website every time the server goes down xD
I think the whole elephant thing is getting old: now it is just used as a signpost to say "this is going to devolve into a BW vs SC2 shitfest".
As so many others have said, I don't get what your objectives are: if you're trying to convince us all that BW is superior, you're going the wrong way about it. If you're trying to stir up a clusterfuck of a thread, you spent way too long with those statistics.
|
relax, there doesnt need to be any agenda or hate/prejudice targeted toward either game in discussion. all that is necessary is to talk about the subject matter - the statistical data of Elo scores and the causal relations that are behind what we can observe, and why they are there. No need to get fired up about the somewhat provocative title 
|
Oh well I guess I'd better stop playing SC2 then!
Thanks for this BroodWarHD!!
|
Hong Kong9153 Posts
On November 29 2011 03:04 BroodWarHD wrote:relax, there doesnt need to be any agenda or hate/prejudice targeted toward either game in discussion. all that is necessary is to talk about the subject matter - the statistical data of Elo scores and the causal relations that are behind what we can observe, and why they are there. No need to get fired up about the somewhat provocative title
It's a cop-out to make a blog like this with a title like this and then tell people not to get "fired up" about it, ending your post with a smiley.
|
Yeah i guess i get to make blogs "proof of the death of BW" and have it be perfectly acceptable? nah i don't.
|
On November 28 2011 09:37 BroodWarHD wrote:
gondolin, isnt that argument only relevant in very low sample sizes? given the 100s of games both league players have played, it should no longer be relevant, would it?
I am sorry for resurecting this thread, but I forgot to look at the answers.
So to reply: the answer is no. It surprised me a lot when I discovered this (I even wrote a post to PoP a few years ago), but as I said, the XvY matchups have much lower ELO than the global ELO rank, while they have thousands of games on Starcraft. It appears that the dilatation process is very slow (and if I remember correctly, TL use k=19 or something like that, not 32), so losing against a 2000 player will hurt you a lot more than losing to a 2200 because there is only 2200 players in the OSL/MSL.
|
I tried to calculate and
even with K 19, it takes
eloafter= elobefore+10*0.7x-9*0.3x
eloafter= elobefore+16*0.7x-16*0.3x
assuming equal opponent elo, it takes x=
47
70
So i underestimated game number in first try so its hard to get high elo i understand.
|
|
|
|
|
|
EPT
