EPTDo you macro like a pro? - Page 51
| Forum Index > SC2 General |
|
KevinIX
United States2472 Posts
| ||
|
Nemireck
Canada1875 Posts
On October 30 2011 13:01 KevinIX wrote: Is SQ the new APM? Actually, yes. It follows the exact same principle. APM does not measure how effective those actions are spent. Therefore, it would be false to say that "Players with higher APM are higher skilled players." Rather, it would be accurate to say "Higher skilled players TEND to have higher APM." Likewise, since SQ measures how low you keep your money, but not how effective those expenditures are, it would be false to make the claim "Players with higher SQ are higher skilled players." But true to say "Players who are higher skilled, TEND to have higher SQ." | ||
|
Cubbieblue66
95 Posts
| ||
|
sluggaslamoo
Australia4494 Posts
On October 30 2011 14:18 Nemireck wrote: Actually, yes. It follows the exact same principle. APM does not measure how effective those actions are spent. Therefore, it would be false to say that "Players with higher APM are higher skilled players." Rather, it would be accurate to say "Higher skilled players TEND to have higher APM." Likewise, since SQ measures how low you keep your money, but not how effective those expenditures are, it would be false to make the claim "Players with higher SQ are higher skilled players." But true to say "Players who are higher skilled, TEND to have higher SQ." What? APM and SQ measure completely different qualities of a player. | ||
|
StreetWise
United States594 Posts
On October 30 2011 14:46 Cubbieblue66 wrote: Is it worse to have an APM around 50 or a SQ around 50? If that APM is EAPM, then I would say actually pretty even. In SQ terms, 50 means you are floating what you are making, i.e. if your income is 1000 minerals, you are floating 1000 etc. If you have an income of 1000 and only float 500 minerals, then that gives you an SQ of around 70. Only using a small sampling of me and 3 friends, I am masters and the 3 friends are all battling it out in silver league, I average 76 EAPM, and 72 SQ. While my friends are in the 50s for both, so at least in my small group, there seems to be a strong parallel. | ||
|
Nemireck
Canada1875 Posts
On October 30 2011 16:00 sluggaslamoo wrote: What? APM and SQ measure completely different qualities of a player. You've completely missed the point of my post. Re-read it, and try again. | ||
|
sluggaslamoo
Australia4494 Posts
On October 30 2011 16:30 Nemireck wrote: You've completely missed the point of my post. Re-read it, and try again. I don't get it. What's the point of your post? APM is a very quick indicator of the speed of that player. SQ is a very slow indicator of the macro of that player. SQ will never be the new APM because it takes so long to deduce SQ compared to apm. You go up to someone and he says he has 300 apm, ok that's a very quick indicator of skill. If he says he has a SQ of 50, you go what?! Oh right just use this formula and you'll get it. ![[image loading]](http://i.imgur.com/dGY3J.png) APM also is a brief indicator of the potential micro/macro/multitask of that player. SQ is very specific. You can have very high SQ with very low APM, but you will much less likely have low SQ with high APM. Not discounting the article, its a great article, but your point makes no sense. | ||
|
Nemireck
Canada1875 Posts
On October 30 2011 16:33 sluggaslamoo wrote: I don't get it. What's the point of your post? APM is a very quick indicator of the speed of that player. SQ is a very slow indicator of the macro of that player. SQ will never be the new APM because it takes so long to deduce SQ compared to apm. You go up to someone and he says he has 300 apm, ok that's a very quick indicator of skill. If he says he has a SQ of 50, you go what?! Oh right just use this formula and you'll get it. APM also is a brief indicator of the potential micro/macro/multitask of that player. SQ is very specific. You can have very high SQ with very low APM, but you will much less likely have low SQ with high APM. Not discounting the article, its a great article, but your point makes no sense. What does the speed of the calculation have to do with anything? My original post is quite clear. Re-read it, and try again. | ||
|
Peleus
Australia420 Posts
| ||
|
sluggaslamoo
Australia4494 Posts
On October 30 2011 16:43 Nemireck wrote: What does the speed of the calculation have to do with anything? My original post is quite clear. Re-read it, and try again. Ok genius, SQ is the new APM, as they follow the exact same principle. I hereby dub thee, sir troll of trollingtown. | ||
|
Nemireck
Canada1875 Posts
On October 30 2011 16:49 sluggaslamoo wrote: Ok genius, SQ is the new APM, as they follow the exact same principle. I hereby dub thee, sir troll of trollingtown. Your problem is that you seem to be misinterpreting the principle that I've illustrated. No, I'm not trolling you. You just, again, have missed the point of my post, and likewise, the point being brought to the forefront by the post I'm specifically responding to. Re-read my original post, and see if you can highlight the point I'm making. | ||
|
MasterBlasterCaster
United States568 Posts
| ||
|
vetinari
Australia602 Posts
On October 30 2011 16:49 sluggaslamoo wrote: Ok genius, SQ is the new APM, as they follow the exact same principle. I hereby dub thee, sir troll of trollingtown. /sigh Once you know what the two are, you can use them as quick indicators of skill. If you have 300 APM, you are probably better than someone with 250 APM If you have 80 SQ, you are probably better than someone with 70 SQ. Obviously, both aren't precise. Just because you are fast, doesn't mean you actually have good macro, micro etc. Just because you have good macro, it doesn't mean you actually have good micro, speed, etc. However, since good macro, micro, speed, etc are all positively correlated, you can use one as an indicator of skill in general. That is the principle he is referring to, not the method of calculation. | ||
|
Nemireck
Canada1875 Posts
On October 31 2011 20:17 vetinari wrote: /sigh Once you know what the two are, you can use them as quick indicators of skill. If you have 300 APM, you are probably better than someone with 250 APM If you have 80 SQ, you are probably better than someone with 70 SQ. Obviously, both aren't precise. Just because you are fast, doesn't mean you actually have good macro, micro etc. Just because you have good macro, it doesn't mean you actually have good micro, speed, etc. However, since good macro, micro, speed, etc are all positively correlated, you can use one as an indicator of skill in general. That is the principle he is referring to, not the method of calculation. I feel I should clarify a couple of the statements you've made, because you really DO have to be careful about how you use these statistics. And the statements you make based on them. "If you have 300 APM, you are probably better than someone with 250 APM." You would think that this is an accurate statement. After all, the APM trend line tells us that better players tend to have higher APM than lesser-skilled players. Unfortunately, the statistic itself doesn't tell us the probability of anything, in general. The only level of play where the difference in APM really shows is at the top-tier level of competition, and even then, the stats I saw that illustrated APM vs win-rates was MLG Columbus (or Anaheim, whichever MLG Losira attended), and those satistics are old now. A 50 APM difference at the top tier level of play all but guarantees a win for the player with higher APM, but that may be an inaccurate statement now, 6 months later, as we actually don't have an accurate measurement of APM since Blizz changed the calculations. The only accurate statement that could be made, using specific APM measurements is: "If you have 300 APM, there is no guarantee that you are better than someone with 250 APM, but, better players tend to have higher APM." "If you have 80 SQ, you are probably better than someone with 70 SQ." This statement runs into the same problem. Two players, in any given game may have a specific SQ, and that tells us nothing about who won the game. So to make a statement using specific numbers it would have be worded something like: "If you have 80 SQ, there is no guarantee that you are better than someone with 70 SQ, but, better players tend to have higher SQ." The principle that I've illustrated is that the statistics are correlative, not causative. Which means, as you've stated, that just because you are a fast player with 300 APM, doesn't mean that you have good micro, macro, scouting, decision making, building placement, etc. Likewise, (and this is where an important clarification rests vet) since SQ is only a measurement of whether you are spending your money, and in no way tracks what your resources are being used for (just as APM is only a measurement of actions being made, without tracking exactly what those actions are being used for), it is possible that just because you have an SQ of 90+, doesn't meant that you have good macro, decision making, unit placement, building placement, etc. What these stats give us are just that... statistics that we can measure, and compare. Like shots on goal and puck possession in hockey, as we build up a number of statistics that we can actually track and calculate, we can start to quantify the overall skill of a player. Going back to hockey. The best team in the league may not have the most goals in the league, or the least goals against, or the highest puck possession, or the best power play, or the best penalty kill etc. But quite often, the best team in the league is in the top 5 across all categories. Any single statistic doesn't tell us anything about the skill of a team in hockey, but if you can quantify a number of meaningful statistics, and rate teams along each one, the teams that rank highly across the whole variety of statistics will begin to show WHY they are the best teams in the league. So right now, we have APM, and SQ that can we can measure quantitatively. So the chance that a player A with 200 APM is better than player B with 150 APM is one number (for the sake of illustration, we'll say 55%). The chance that player A with 90 SQ is better than player B with 75 SQ is another number (again, I'll just throw out 55% for the sake of illustration). But, if player A has 200 APM and 90 SQ, compared to player B with 150 APM and 75 SQ, then the chance that player A is better than player B could jump. Instead of 55%, it could be 60%, or 75%, hell, it could even be 95%. It's possible that someone with higher APM, but lower SQ is more likely to beat a player with lower APM and higher SQ. We don't know for certain since the stats haven't been done. But the more stats that we can quantify and correlate with higher skill levels, the more likely we can look at a collection of statistics and accurately predict the probability that one player will beat another player in a given match up. Some stats that I think we should track: APM SQ Units Lost (as a measure of resources) Units Killed (as a measure of resources) Rate of Income increase (who generates more income faster) Game Time (it's possible that better players play shorter, or longer games, or that better players tend to play games that end at a sweet spot in-between) Number of times supply-blocked There may be other stats that I can't think of that we can show correlate to skill. And we should ideally be tracking all of them. The more individual skills that we can quantify, the more accurately we can paint the picture of what SC2 skill actually looks like. | ||
|
gronnelg
Norway354 Posts
| ||
|
Dhalphir
Australia1305 Posts
On October 30 2011 16:49 sluggaslamoo wrote: Ok genius, SQ is the new APM, as they follow the exact same principle. I hereby dub thee, sir troll of trollingtown. His point is that you guys are retarded. SQ is the new APM because both are equally useless for determining the skill of a player. Higher skilled players tend to have high APM. High apm does not, in turn, however, mean you have a high skill. And the same goes for SQ. This is the worst thread and the worst concept anyone has ever come up with in the history of the world. Either SQ does correlate closely to ladder rank - in which case its pointless because we already have ladder ranking to measure skill, or it does NOT correlate to ladder rank in which case it is arbitrary and bullshit. The only thing that SQ achieves is to let someone walk into the TL Strategy forum and say "ok guys I have a SQ of 582582672862862 but I'm only in Gold because I get cheesed a lot, so I'm basically Masters". Its a way for shitty players to be able to justify to themselves and to others why they think they are not as shit as they are. | ||
|
whatthefat
United States918 Posts
On November 02 2011 08:40 Dhalphir wrote: His point is that you guys are retarded. SQ is the new APM because both are equally useless for determining the skill of a player. Higher skilled players tend to have high APM. High apm does not, in turn, however, mean you have a high skill. And the same goes for SQ. This is the worst thread and the worst concept anyone has ever come up with in the history of the world. Either SQ does correlate closely to ladder rank - in which case its pointless because we already have ladder ranking to measure skill, or it does NOT correlate to ladder rank in which case it is arbitrary and bullshit. The only thing that SQ achieves is to let someone walk into the TL Strategy forum and say "ok guys I have a SQ of 582582672862862 but I'm only in Gold because I get cheesed a lot, so I'm basically Masters". Its a way for shitty players to be able to justify to themselves and to others why they think they are not as shit as they are. Sorry you feel that way. But a lot of people have personally thanked me for developing this, as it gives them a way to track their own progress. Having a quantitative metric has helped them to improve their macro during training. The idea that any measure that correlates with ladder rank is useless because we already have ladder rank is obviously a fallacy. That's like saying there's no point in analyzing a runner's speed off the blocks, or efficiency of their stride, because we have their times. Knowing a runner's speed off the blocks isn't just so noob runners can say "I'm faster off the blocks than Usain Bolt!" It's a useful metric to know, so runners can assess whether they need to improve in that area. Overall skill at starcraft (which ladder rank is only a proxy for) is similarly due to a wide range of factors. Players in the same league may differ widely in these abilities, e.g., some will be better at macro, some will be better at micro, etc. For anybody who is serious about improving their play, it is helpful to have easy ways of assessing these factors. Such methods don't currently exist, at all. I'm sorry you consider a small step towards developing such methods the worst concept anyone has ever come up with in the history of the world. | ||
|
Dhalphir
Australia1305 Posts
I just want people to say "yes, I am in Gold. I am a Gold level player, I am not convinced that I'm a Diamond level player in disguise, I'm just plain Gold level and I suck, and I want to get better. There is no such thing as "I play at a Diamond level but I'm stuck in Silver because I get cheesed a lot". Nope, you're just plain ol' Silver. And this SQ number gives them another number to hide behind like the idiots they are. And it makes me angry. | ||
The_Templar
your Country52798 Posts
On October 30 2011 14:46 Cubbieblue66 wrote: Is it worse to have an APM around 50 or a SQ around 50? I say sq, because my SQ is usually double my apm. | ||
|
vojnik
Macedonia923 Posts
 | ||
| ||
