EPT
Hello, sports fans! It's time once again for the favorite pastime of the TL stat desk - to comb through spreadsheets and see if anything noteworthy pops up. With a new game comes 5new opportunities to look at numbers and convince ourselves they have meaning.
I wanted to answer basic questions like what the race breakdown is among players are at the top, and how it compares across servers. Player rankings is not public yet, but I was able to scrap together a mostly-whole picture thanks to some recent threads. I slapped it all into a spreadsheet and wrote a nice script that tallies up the availible statistics from the top users. The roundups I have seen so far seem to focus on the leader of each division, which is a little shortsighted as the top few ranks shift around frequently (excluding some people who are #1 by several hundred points). To get a more complete look at the 'top player' population as a whole, I made the cutoff the top 8 for each platinum division.
Keep in mind that this data is now a few days old and will not reflect current standings perfectly (data from March 9th). It is also important to note that the sample consists of the first 8 divisions in Asia, the first 13 in Europe, and the first 16 in the US, despite some of these having more divisions. I don't consider the other divisions mature enough to matter yet, as a lot of them have people in the top 8 who still have ratings of 1000.
Finally, before I begin, I'd like to emphasize that this is meant to be fun to look at; how you interpret the data is up to you.
![[image loading]](/staff/Daigomi/hey_Player_percentage.png)
Not all add up to 100% due to rounding
This is a good starting point, it is the racial breakdown for each beta server. Protoss is definitely dominating the top of the ladder right now, but I suspect it is because toss was considered (and probably was) the best race to begin with, and we will see them lose ground in the coming weeks. It looks to me intuitively that the later created divisions have more non-protoss but I didn't run any hard analysis in these divisions.
It's interesting that US has a higher ratio of random players, but it is doubtful that it is telling of anything.
![[image loading]](/staff/Daigomi/hey_Win_rate.png)
Here are the real meat and potatoes. It is pretty cool how close everything is, with most things hovering right around 63%. Nothing stands out in any big way, but there are a couple of spots that are worth mentioning.
USA seems to be the home to the best tosses, as their winrate is both higher than other races' within their region and also higher than other protoss players' in other regions. It seems that iNcontroL and IdrA's choice to pledge their allegiance to Aiur has swayed the statistics for their entire country (Smuft playing 450 games with nearly 80% of them won certainly doesn't hurt). The beginning of a new era? Only those from the future can know for sure.
In contrast, Asian zergs are the ones who appear to truly know what is what. Maybe they discovered some tricks with mutas tuat we don't have overseas yet, maybe they have a strong lineage of amateur zergs who are giving up on BW to follow their dreams in a new game. Maybe the real reason MuMyung is off SKT is because he is training a team of zergs in an underground sweatshop preparing for a conflict with North Korea. I don't have enough information to say, but it might be worth investigating.
Interestingly, Asia has the biggest disparity between races. It's the only server where the difference between their best race and worst race is nearly 10 points. With a huge percent of their highly ranked players choosing toss, it might just mean that ZvT hasn't quite been fleshed out or that zergs are really good at ZvP and terrans are overly bad at TvP.
Asia also has the lowest overall percentage, which I'd like to think confirms the common wisdom that in Asia the casual players aren't that much worse than the serious amateurs. Still, it's difficult to say with the limited sample size available.
The highest winrate in Europe is random, which is pretty cool, but as there are only 5 players sporting the R icon it means that the two of them who are performing the best (ret and MinD have about 75% ratios) go a long way towards bringing up the average.
All in all, I'd wager Blizzard has a more accurate way of quantifying this, and they are pretty pleased with the results. It is very impressive to achieve such symmetry within the top echelon of a game in the early stages of beta.
![[image loading]](/staff/Daigomi/hey_Average_rating.png)
This is almost the same thing, but it shows us how the mystery point system rates the races/continents. Strangely enough, in Europe, the random players average a significantly lower point value than the other races, by over a hundred points (which seems significant but who knows with this system). Otherwise it's all rather even which I guess means the system is mostly working. We can say with confidence that the average "tournament qualifier" (as it is called in the interface) currently has 1700 points or more, isn't that neat?!
One last thing to mention is an incredible feat.
The player with the most games is Hyo of Asia with 845. This is especially impressive considering the numbers I'm working with from Asia are from about a day before EU/US. The highest on EU is Desis (788) while in the USA it is ReddoG (615).
Unsurprisingly, Asia as a whole seem to play more games. The average number of games played by a top player (those in the top 8) is a whopping 294 in Asia. With just under 3 weeks played, thats about 15 games a day which is no laughing matter. For comparison, the number in Europe is 246 and in US it's "only" 227. As HotBid put it:
HotBid: some of these asia server stats
HotBid: 555-220
HotBid: THATS why asia is best server lol
Last week Yoren found something pretty cool by looking at the win-rates of the top players. His conclusion is that the pack is getting much narrower. Time will tell if that is a continuing trend or if the beta is settling and truly hitting its stride.HotBid: 555-220
HotBid: THATS why asia is best server lol
Overall, do these numbers say anything significant? Probably not, but they're interesting to look at and fun to speculate about. Realistically there are way too many variables to track, the game is changing on a weekly basis after all, as are the popular strategies, people's perspectives, their chosen race, and maybe even the matchmaking system.
There is, of course, a much more controversial issue regarding play at the top of the ladder. The new battle.net system has made us a lot of promises, but has fallen a bit short for the aspect of the game we all deeply love: the competitive circuit.
Hot Topic: Competitive Play
Since the anouncement of SC2, Blizzard has told the world they are interested in keping the Brood War lineage alive by making its sequel the competitive RTS of the future. They went so far as to hire staff solely for the purpose of creating an internal e-sports (ahem-ESPORTS) team, which to my knowledge is an industry first.
For the past year the talk has been vague promises about how the new battle.net system will blow all anything we currently have out of the water. With the advent of the beta, we have some concrete info and a taste of what is to come. Right now there are two areas of concern.
Issue 1: No global rankings
Its highly annoying that the best comparison any one player has is merely to the people in his division, and the aggregate number of points awarded or taken by the system is also dependant on the division itself. When broken into groups, the number of "what if" variables is absolutely huge. Sure there are 8 people in the who are 'on top', but how are we to compare anyone in one division to someone in another? Sure, we know some people are ahead of people in their divisions, but what does that tell us? What about people who are in the middle of their leagues? How can we tell how they would do against each other? In a serious competitive setting we absolutely need a way to compare any one player in a direct manner to another - something other than "is top 8 of his league" and "is not top 8 of his league".
Its understandable to not want a gargantuan ladder ranking where people see themselves as ranked #14,678 in their region, but this division grouping doesn't cut it. Starcraft servers of the past have had a system of grade rankings, things like A through D which allows us to compare players of various skill levels quickly, easily, and without debate. A possible compromise would be to keep the division system, but make the points within each more straight forward. As of right now its not even possible to compare points across divisions, as the calculation takes into account relative levels of players within each grouping. Regardless, if SC2 wants to cater to the competitive circuit something of that calibur is a necessity.
Issue 2: No cross-server capabilities
Being able to compete with people around the world without hassle is one reason current BW is great. The recent twitter chat seems to say this is one feature that won't be available at launch. As GARIMTO said regarding the state of PvP, this is one step shy of a disaster. Being able to easily grab a 3v3 or 4v4 game will be an integral part of the game experience, this is one of the reasons the current scene is so tight knit across countries. With the community split between continents, not being able to do this effectively segregates the community.
For tournaments this makes all kinds of competition essentially impossible. For people to play worldwide they would effectively need to own (and have installed) 3 separate copies of the game. Tournaments as of right now are a way to bring people together, TSL had participants from 13 countries fight on a level playing field to decide who is the best outside the Korean scene. It is not only annoying for players who want to know where they stand, but for viewers who want to see their favorite players duke it out mano-a-mano to find out who comes out on top. Satchel Paige never threw a ball to Babe Ruth, and to this day there is endless speculation on who would have won on the baseball diamond. This absolutely needs to be a capability of the game at launch.
SC2 has a chance to fix one of our current problems in the BW community - that the Korean and Chinese scenes are outside the reach of anyone not living in those countries. The system in place right now is a bit of a step backwards, further dividing up communities when it is not needed.
All in all, its too early to truly judge, but the details we have are not promising.
This post was made by the Team Liquid Starcraft 2 Coverage Team. For more of TL's coverage, please visit the Team Liquid Starcraft 2 Beta Page.
