If you take a look at the numbers of players in each league, you’ll see there’s a huge gap between Master and Diamond (2.6% of players versus 10.9% in Diamond), and a slightly smaller one between Diamond and Platinum (14.2%). The next time you see such a huge difference is between Silver and Bronze (21.1% of players in Silver and 34.2% in Bronze). What do these numbers tell us? There is a huge gap in the skill, and consequently the understanding of fundamentals of the game, between the various leagues on Battle.net. What it means is that there is an elite group in the higher levels that have a significantly better understanding of how Protoss, Terran, and Zerg interact that allows them to play the game at an entirely different level. I have some experience in this phenomenon. I’m a high level Diamond player. There is a world of difference in how I play as opposed to how a Bronze level Terran player functions. For instance, there’s a fundamental shift in tactics. I don’t think I’ve ever seen a Bronze player attempt a Siege/Marine/Medivac build against Zerg. I’ve seen them imitate such a build, but never come up with it. Why? Because the demands of that skill level don’t necessitate evolving your technique. Sure, you look at a higher level and say “oh, that’s what I need to do” and you thereby exclude yourself from that caliber of “poor” player because you’ve shifted your tactics to match the higher level thinking of a Diamond or Masters level player. Contrary to what most new players think, actions per minute is only a small part of that evolution to a better player. Enter sGm gaming. These guys are out to both make it in the competitive world (they recently received a major sponsorship from a United Kingdom company) and in training. They’re offering coaching and lessons for StarCraft 2 in all the races. Not exactly the usual plan from a competitive gaming team. In the spirit of getting to know how they wanted to approach the teaching of StarCraft 2 I agreed to take a few lessons with them. Of course, I’m more of an advanced student than they will normally see but I suppose that just leaves more room for improvement. Let’s get it right out there that these guys aren’t college professors. I’m certainly not going to tell you that they’re the best teachers I’ve ever seen. But they are certainly a step above the guides that have set up shop. A while back I recommended the Shokz Guide which is still an excellent guide for beginners, and beginners only. Shokz has since stopped any meaningful updates of the guide. That doesn’t mean I wouldn’t say it’s no longer useful. I did a couple of sessions with each of the “specialists” in the sGm organization. sGmStrifeX is a Terran player offering his wares as a tutor of the beleaguered humans. His understanding of the game from the Terran perspective and the fundamentals of the mechanics he teaches are sound. If you’re a Terran player you’d be in good hands with StrifeX. I also did a stint with a Zerg player sGmHyu. Hyu is a good Zerg player. I’d dare say he’d be competitive (which he is) professionally. But he very much needs work as a teacher. There is only so much a student can learn in a session. Hyu needs to understand that, and with more practice he will. His concepts of the race are also poorly articulated. It’s one thing if he understands them but entirely different to pass those on to the rest of us. Lastly, I played with sGmOverbear in a few Protoss sessions. Of course, Terran and Protoss are more similar than Terran and Zerg (and I also played Protoss in StarCraft 1) so these sessions were a bit easier. He effortlessly guided me into my first 4-gate. He did a great job of transferring my knowledge of Terran into Protoss success. Because of his ability I greatly improved my Protoss play. These guys are the real deal. If you’re looking to improve your game beyond the ability of a guide (or know you want to be better than a guide could allow) this is the way to go. I’ve had plenty of people approach me about offering StarCraft training and these guys have been by far the most persistent (half of them don’t even bother letting me train with them a bit let alone sticking with it for months). Look them up. You can contact them on our forum through Private Message or e-mail (hyubushu[at]yahoo.com) as they are members. You can also send me an e-mail (jtietz[at]starcraft2forum.org) or Private Message and I will put you in contact. Source: LiquidPedia
While your suggestion that there is a fundamental knowledge gap between one league and another (say, Platinum and Bronze) is not without merit, you are working on the assumption that all of these players have played StarCraft the same amount of time in a given time period. This, according to normal distribution in statistical theory, cannot be the case. Furthermore, we cannot conclude from these statistics alone that there is a knowledge gap, as there are variables that have not been controlled for, namely 1) time spent with the game on a weekly basis, 2) total amount of time spent with the game, 3) interest for mastering the game, 4) cognitive abilities, 5) physical abilities (high apm) etc. Obviously, these are variables that we, as gamers, do not have access to. Concretely, those who play less often will be at a clear disadvantage compared to players who dedicate more time to mastering the game. No one disputes this. However, the fact that a Bronze player will usually lose to a Platinum player is not necesarily indicative of a lack of ''understanding of the fundamentals of the game'' on the part of the Bronze player nor does it mean that a ''knowledge gap'' proved to be the deciding factor in Bronze players' defeats to Platinum players. There are many more variables in play that may exercise a more important role in the observed trend (Bronze players losing to Platinum players). I thought that as a journalist you might be interested in a more scientific critique of your statistical interpretation. Otherwise, I thought this article was excellent.
Those are valid points but your variables are different to me (as far as league placement at least). Fundamentals of StarCraft are what lead to league placement changes between Bronze and Platinum/Diamond, not time spent with the game. There are numerous indicators, including the age-old lesson that "you continue building workers" and you're almost guaranteed to be placed in at least Silver. I don't think the other variables you mentioned such as cognitive abilities or APM have anything to do with MMR ratings until you're in Diamond league at least.
Given your reply, I'm not entirely convinced you've understood my criticism. Poor wordsmith that I am, I must have not expressed myself adequately. I was attempting to suggest that players' win/loss ratio (what ultimately determines players' league placement) is influenced by a wide array of variables, not simply 'knowledge of the fundamentals of StarCraft'. I also stated that you cannot, based on the statistics available, suggest that 'knowledge of the game' is the determining factor in your win/loss ratio, as there are a wide range of variables that were not controlled for. That is a basic statistical principle, one that is frequently misunderstood. Regardless, it was a relatively minor criticism. I simply thought that you might appreciate a more pointed reaction to your article. On another note, I find it interesting that you should so easily dismiss 'cognitive abilities' as an important variable, as cognitive abilities allow us to react to a given stimulus (problem, situation, etc) with the strategies that are at our disposal --a notion that can very easily be applied to StarCraft. For example, the article intriguingly underlined what appears to be a creative discrepancy that is the hallmark of great players. For a great example of 'creativity' in StarCraft, we need look no further than 'Pimpest Plays', a truly awesome omnibus of moments of unadulterated brilliance. But are great players creative because they 'know StarCraft' as you seem to suggest or because they have relatively precocious cognitive abilities that allow them to use the strategies at their disposal in creative ways? I suspect the latter to be true, as knowledge of StarCraft does not necessarily imply creative abilities. And that leads me to wonder whether some players may be less 'creative' (obviously), and whether that lack of creativity is detrimental to their win/loss ratio (probably).
lol I love reading your posts Zealot. But know that your win/loss ratio does not determine your league placement. On the far ends of the spectrum it tends to be like that but disregarding bronze and GM everyone is roughly at 50%. It's why we have match making, to have it this way. As for the topic, creativity, among other things, is definitely an aspect which separates players from each other, but I would be inclined to agree with Jon in that knowledge of game basics (worker production, production facility production and spending income, scouting, etc.) constitutes a bigger part of a player's gaming skill than cognitive skills. I base this off of myself: I don't use build orders, I don't know timings and can barely execute basic tactics, but because my mechanics (mentioned above) are solid I'm in diamond and I'm positive I could get master's if I tried, a league whose members I would consider good players.
You're right, of course: I was speaking of the talent spectrum's two extremes. You're also correct to point out that the formula to determine your ranking is more involved than just your win/loss ratio. In essence, it's your game performance versus the game performance of those under you, those at your level, and those at higher levels. It's basically a delineated Z score. (Some statiscian had fun inventing this one, to be sure). I also believe that Blizzard takes into account some other variables, such as APM, game length, units built, resources gathered, etc... Does anyone know what variables, other than win/loss ratio, that Blizzard takes into consideration when ranking you? I'd be curious to know what they are. As for our discussion regarding creativity versus knowledge of fundamentals, I would like to point out that all the examples that you mention (worker production, spending income, scouting, etc.) are concepts that lend themselves to all of Blizzard's RTS titles, from WarCraft II to StarCraft II. It could also be applied to many other relatively generic RTS series, such as Command and Conquer. More startingly, I could apply these notions to an RTS with a truly different feel, Sins of a Solar Empire. I draw your attention to this not to be polemical, but in the hopes that we might engage in a fruitful conversation. Here's my question: what do we mean by 'knowledge of StarCraft's fundamentals' when we are obvioulsy referring to general notions that can be readily applied to many other RTS titles? And given this, in what way, if any, are 'StarCraft's fundamentals' readily distinguishable from any other RTS? Are the 'fundamentals' to which you and Jon refer simply not basic principles of Economics being aggressively tested in a wide variety of RTS' temporal and geographic contexts? Finally, how would you use 'knowledge of fundamentals' to explain why a player might be brilliant at Command and Conquer yet only average at StarCraft II?
My point is that fundamentals don't make the difference between "average" and "brilliant"" players. I'm just saying that specifically in StarCraft 2 you have these mechanics--which I think Kuvasz described better than I did--that determine your level more than anything. People who play casually simply don't understand these techniques and so despite other factors--cognitive reasoning, intelligence, APM, and weekly time spent--they will hit a wall until these concepts are mastered. You can see a direct correlation (proof isn't required when the trend is obvious) between Bronze players who lack the ability to produce sufficient economy and don't understand the term "saturated" to Silver players who usually understand this economic ability but have a hard time supplementing with a good production, timing, or build scheme. It takes a combination of these techniques to reach a level of Platinum or Diamond at which point the factors you suggested above come into play. Simply as a player who understands these mechanics but plays almost 0 games per week on average I can maintain my Diamond status.
But aren't all of the mechanics that Kuvasz mentionned applicable to a wide range of RTS games, as I pointed out above? That implies that if I'm good at StarCraft II, I should also be good at Sins of a Solar Empire and Supreme Commander II. I maintain that among other things, time spent with the game is a crucial variable, as you will otherwise be unfamiliar with the game's various idiosyncracies and deviations from the standard RTS template. Any novice statiscian will point out that a correlation does not necessarily imply cause to effect. For example, did you know that people have more headaches when flies breed at an accelerated rate? Surprising, but the two variables are nonetheless correlated. Of course, they're not correlated directly, because one does not cause the other. Thus, until you can control for the other variables in an experimental setting, you cannot determine cause to effect. That has been my point all along. I recognise the importance of the so-called 'fundamentals', but I think that there are other variables to take into consideration that are just as important. Whether I'm right or wrong is irrelevant, because you cannot, given the data in your possession, assert that the skill gap is caused by 'knowledge of fundamentals', as you haven't controlled for any other variables. It doesn't matter how 'obvious' it may seem, you must determine cause to effect before making an emphatic claim. (Do flies breeding cause headache pandemics?) You might, due to familiarity with the game, feel very strongly that your assumption is correct (and it may be), but you cannot state that the data empirally shows a causal link between the two. This is what I was I attempting to convey in my clumsy manner. All this to say that you should have said something to the effect that: ''One possible interpretation of the ranking system is that lower-tier players lack knowledge of the fundamental mechanics underlying StarCraft.'' I only brought this to your attention as you're a jounalist; the vast majority of journalists are simply not trained to interpret statistics, and I thought you might find it interesting to see why you could not, statiscally speaking, interpret the data in the manner that you did. If you still don't understand my criticism, that's fine: 90% of your readers will never know the difference anyway.
For the first part: Blizzard has stated that the matchmaking system does not take into account anything other than who wins in a particular match. Be it a 6-pool, a 4-gate, or a 50 minute game with mass BCs, at the end of the day it's still just a regular win for one and loss for the other. This is so playstyles are left to develop however they want (as long as they're balanced), and this is also why we have people in master's who have 6-pooled their way there. Annoying? Perhaps. Should the system take into account the details you listed? No, because then the game becomes an e-peen contest of who made more workers and who spammed their APM higher. It doesn't matter how you win. This is also why win trading/portrait farming is possible on the ladder and is to this day an offence that may result in suspension. As for the second part, you are right in that the basics I listed and Jon alluded to are applicable to RTS games in general, but understand the awkwardness of the article were Jon to say RTS fundamentals instead of StarCraft. Not only could it be taken more of an insult (I would rather not go into details on this or the instability of some players' self esteem - feel free to skim through some posts on the official battle.net forums), but it would've caused distraction - is the article about StarCraft or RTS games in general? I would also like to point out that I interpret his sentence slightly differently. The underlined word clearly limits the focus to StarCraft and not RTS in general, as these races do not appear in any other title. Know that the interaction Jon refers to here includes knowing your opponent's timings (when to send out a scout on which map to catch a tech path, when to expect 3-gate robo push, 4-gate, 2-rax bunker rush, 7-pool, etc.) as well as your own timings (at which percentage of completion of the gateway should you send your probe for a cyber-wall, when to start making extra supplies [espcially with respect to game progression], etc.), things that even a pro C&C player has no clue about because these are idiosyncrasies.
I think the underlying argument here is that you're saying my assumption that there is a correlation between these "RTS" abilities and the league level of a player may not exist independent of other variables. Without a scientific study on the matter where I can eliminate all the other variables (likely not possible in a game such as StarCraft 2) I can't prove this decisively for you. But I can promise that the correlation is there. There's picking something apart and then there's ignoring the reality of the situation. No, I didn't cite a scientific study here. Yes, there is a correlation. Undoubtedly. Irrespective of studies done, could you not with reasonable certainty--and I emphasize reasonable--suggest that there is a correlation between driving impaired and an increase in accidents? I would say yes. You can say "well, you didn't take into account other variables such as age, car model, road conditions" etc. but you can with some certainty say that the impairment of the driver caused an increase in such accidents. I'm pointing out a 100% correlation between a very specific variable and the MMR ranking of players. This is solely from observation and nothing else. But if you check any of the replays of a Bronze level player my above mechanics will undoubtedly be missing from their playstyle. And you can take that to the bank Unless you use a debit card cause then they'll charge you $5. And I understand that you're saying since I didn't prove it empirically that I can't state it as fact. I think you're right there, but that's where journalistic license takes over. At what point do I have to weigh clarity of opinion versus statement of fact? I'm not obligated to be academically correct. I'm obligated to be clear in the meaning of my words. So when a trend is experiential and apparent, there's no reason to muddy the waters by saying it is possible to be caused by other variables (which it's not).
