It's interesting to see how there is still such a large difference in macro skill even between diamond and masters, and between amateur players (in masters) and pros (top 100 GM).
Do you macro like a pro? - Page 50
Forum Index > SC2 General |
Warble
137 Posts
It's interesting to see how there is still such a large difference in macro skill even between diamond and masters, and between amateur players (in masters) and pros (top 100 GM). | ||
Deleted User 61629
1664 Posts
| ||
halpimcat
215 Posts
I do have a question though: what exactly do the numbers represent in the "Average unspent resources" column? I'm guessing a higher score means you actually spent more? My mind automatically goes to the idea that a lower score is better because you kept your resources low | ||
Robertxtrem
United Kingdom3 Posts
| ||
average650
24 Posts
And, none below 66 in diamond territory I'm only in plat. This will definitely help me focus more on different issues. | ||
p1cKLes
United States342 Posts
Btw, did you see some of Idra's average unspent minerals? Holy crap.. I don't even come close. | ||
Hulkoff
Sweden51 Posts
On October 28 2011 21:07 average650 wrote: Wow turns out macro is not my problem at all.... Usually just below 80, in low grandmaster/high masters territory... And, none below 66 in diamond territory I'm only in plat. This will definitely help me focus more on different issues. Macro is always your problem. Macro is more than just keeping your resources low. Anyone can keep it low by overproducing constructionbuildings and making 10 supply depots at the same time and queuing up 5 units in every building. I've said it before and I'll say it again. SQ is a bullshit stat. Don't read too much into it. What you do with your resources is way more important. | ||
average650
24 Posts
On October 29 2011 02:11 Hulkoff wrote: Macro is always your problem. Macro is more than just keeping your resources low. Anyone can keep it low by overproducing constructionbuildings and making 10 supply depots at the same time and queuing up 5 units in every building. I've said it before and I'll say it again. SQ is a bullshit stat. Don't read too much into it. What you do with your resources is way more important. You're right that those things could be the case, except I know they aren't. I never queue up units, except maybe and extra probe every once in a while. I don't build more than a couple pylons at a time, except when I lose a bunch of pylons to harrass or something. I will sometimes build an extra production facility, but so may my opponents. These things can be improved of course, but looking at all these things, my loses are not because I'm not spending my money on useful things. On October 29 2011 02:11 Hulkoff wrote:What you do with your resources is way more important. This is too vauge. the statistic is a measure of spending the resources you have. It doesn't take into account some macro things, like producing probes or queuing, but these can be analyzed easily. I have, and I don't queue. I produce probes very well, though not perfectly. This can be improved, but looking at this statistic, (combined with other knowledge I have of my own play) this is not my main problem. I struggle with things like focusing on a specific strategy; I need better planning going into matchups. This happens because I don't play enough. I also need to focus on awareness, mostly just by being active with the units I have. I don't have the best apm so I find this difficult, but when I can work it in, I feel much more comfortable. There are probably other things i have yet to identify, but the basics of typical macro are not the problem in my matchups. This does not cover all parts of macro, but it does give a good metric to measure one part of macro. Other parts can be identified in different ways. Worker production can be watched easily. Queuing is easily seen. Over producing buildings is more difficult to quantify, but can be done with work. Really, if you look at your income, and look at how quickly the buildings can spend resources, the spending should be within 1 production buildings worth of being equal. The same can be said for queen injects. Most other aspects are easily analyzed. This part is not usually, and this statistic helps a lot. | ||
halpimcat
215 Posts
On October 29 2011 02:51 average650 wrote: You're right that those things could be the case, except I know they aren't. I never queue up units, except maybe and extra probe every once in a while. I don't build more than a couple pylons at a time, except when I lose a bunch of pylons to harrass or something. I will sometimes build an extra production facility, but so may my opponents. These things can be improved of course, but looking at all these things, my loses are not because I'm not spending my money on useful things. This is too vauge. the statistic is a measure of spending the resources you have. It doesn't take into account some macro things, like producing probes or queuing, but these can be analyzed easily. I have, and I don't queue. I produce probes very well, though not perfectly. This can be improved, but looking at this statistic, (combined with other knowledge I have of my own play) this is not my main problem. I struggle with things like focusing on a specific strategy; I need better planning going into matchups. This happens because I don't play enough. I also need to focus on awareness, mostly just by being active with the units I have. I don't have the best apm so I find this difficult, but when I can work it in, I feel much more comfortable. There are probably other things i have yet to identify, but the basics of typical macro are not the problem in my matchups. This does not cover all parts of macro, but it does give a good metric to measure one part of macro. Other parts can be identified in different ways. Worker production can be watched easily. Queuing is easily seen. Over producing buildings is more difficult to quantify, but can be done with work. Really, if you look at your income, and look at how quickly the buildings can spend resources, the spending should be within 1 production buildings worth of being equal. The same can be said for queen injects. Most other aspects are easily analyzed. This part is not usually, and this statistic helps a lot. After playing around with the autoSQ toy i figured out how the measurement works in practice. Looking at my replays, my SQ was unusually high whenever i found my workers created count unusually low. This just means that I was able to spend my resources better because I never had any resources to begin with. If you're plat level (or anywhere below GM) and your SQ is high, it might mean that your worker production, is low to begin with. It becomes much more difficult managing your economy responsibly when you have 80 workers compared to 60. Or it could mean you don't really play macro games, and instead opt for a more timing-oriented play, in which case SQ is easier to manage. I really doubt macro is "not your problem" when you're in plat so just my 2 cents. | ||
average650
24 Posts
On October 29 2011 06:55 halpimcat wrote: After playing around with the autoSQ toy i figured out how the measurement works in practice. Looking at my replays, my SQ was unusually high whenever i found my workers created count unusually low. This just means that I was able to spend my resources better because I never had any resources to begin with. If you're plat level (or anywhere below GM) and your SQ is high, it might mean that your worker production, is low to begin with. It becomes much more difficult managing your economy responsibly when you have 80 workers compared to 60. Or it could mean you don't really play macro games, and instead opt for a more timing-oriented play, in which case SQ is easier to manage. I really doubt macro is "not your problem" when you're in plat so just my 2 cents. This should account for that because it expects that when you have less income that you spend your money better. But even say you're right, I don't think I'm not making enough probes. I don't play super long macro games, because someone usually makes a fatal mistake before 5 bases are up. Most of my games I get 3 bases though. Sometimes shorter, sometimes I go longer, but usually 3. I feel confident that I tend to make enough probes. In fact, in a game against a diamond opponent the other day, he offered the advice that I had made too many probes, since in that particular game I didn't get a 3rd base up, I had too many probes for 2 bases. Now, he could be wrong, but this is a good indicator that I'm doing a decent job in that department. You also underestimate how poor my army control can be, haha. | ||
whatthefat
United States918 Posts
It's true that there are ways to 'game' SQ, e.g., by queuing units, or staying on a low income and building too many production facilities. Clearly though, these aren't going to help anyone win! In general, the measure does compensate for income, and it sounds like you are spending very well for your league, i.e., your macro is probably pretty good. Now, if you were to face a grandmaster player it would inevitably drop due to the increased harassment, but you are nonetheless macroing much better than the average platinum. Regarding worker production, you could check where you lie on the worker vs. game duration graph for different leagues. | ||
average650
24 Posts
On October 29 2011 10:49 whatthefat wrote: @average650 It's true that there are ways to 'game' SQ, e.g., by queuing units, or staying on a low income and building too many production facilities. Clearly though, these aren't going to help anyone win! In general, the measure does compensate for income, and it sounds like you are spending very well for your league, i.e., your macro is probably pretty good. Now, if you were to face a grandmaster player it would inevitably drop due to the increased harassment, but you are nonetheless macroing much better than the average platinum. Regarding worker production, you could check where you lie on the worker vs. game duration graph for different leagues. This I would agree with. My probe Production over the first 20 minutes (I chose this because it was easy to get a large sample over this timescale) Is about 67 with a standard error of ~3.5. The error is worse than it really should be because I simply took games that were about 20 minutes and treated them all as 20 minutes. Your plot of this data is rather course so it's hard to compare to my statistics to your data, but it seems as though diamond or masters fit this range, though there is little difference between the leagues at this time in the game. Could you post your raw data so I could have a look at it? | ||
Snickersnee
United States241 Posts
| ||
Isaac
United States810 Posts
| ||
halpimcat
215 Posts
On October 29 2011 16:14 Isaac wrote: How can I make my SQ better on games that i am maxed? (strategically, dont tell me to build ravens and supply depots). I just got an 80 sq. (diamond league) and I was maxed killing his bases for a long time. I think people are putting too much emphasis on SQ. The number's not that important (or maybe that's just what I keep telling myself), what matters is that you are constantly making workers and constantly spending your money; SQ doesn't seem to measure your actual income rate, just the rate at which you spend it. Anywho, you should theoretically still be spending boatloads of money when you're maxed, and not on useless upgrades (medivac energy, HSM when not making ravens). You should constantly be losing units throughout your maxed period, and/or be making extra expansions and productions facilities so you can max instantly if you were ever to lose that army. Example: in a quiet macro tvp without any engagements, you should have a maxed out army by the 17-19 minute mark. After 3-5 more minutes, you should have some 10-12 barracks ready to pump out MMG after your major engagement. | ||
Nemireck
Canada1875 Posts
On October 29 2011 17:46 halpimcat wrote: I think people are putting too much emphasis on SQ. The number's not that important (or maybe that's just what I keep telling myself), what matters is that you are constantly making workers and constantly spending your money; SQ doesn't seem to measure your actual income rate, just the rate at which you spend it. Anywho, you should theoretically still be spending boatloads of money when you're maxed, and not on useless upgrades (medivac energy, HSM when not making ravens). You should constantly be losing units throughout your maxed period, and/or be making extra expansions and productions facilities so you can max instantly if you were ever to lose that army. Example: in a quiet macro tvp without any engagements, you should have a maxed out army by the 17-19 minute mark. After 3-5 more minutes, you should have some 10-12 barracks ready to pump out MMG after your major engagement. Correct. SQ is merely a correlative statistic. People with higher skill TEND to also have a higher SQ. But it is possible to have a high SQ, and still lose a lot because all of the other skills required to play are found wanting. Things like spending on the proper units, controlling those units, making enough workers and saturating your bases properly, expanding to enough bases to make your workers worthwhile, defending your opponents aggression, scouting, responding properly to the scout, and etc. etc. etc. aren't factored into the SQ measurement at all. SQ is ONLY a measurement of how low you keep your money based on your income. | ||
whatthefat
United States918 Posts
On October 29 2011 17:46 halpimcat wrote: I think people are putting too much emphasis on SQ. The number's not that important (or maybe that's just what I keep telling myself), what matters is that you are constantly making workers and constantly spending your money; SQ doesn't seem to measure your actual income rate, just the rate at which you spend it. Anywho, you should theoretically still be spending boatloads of money when you're maxed, and not on useless upgrades (medivac energy, HSM when not making ravens). You should constantly be losing units throughout your maxed period, and/or be making extra expansions and productions facilities so you can max instantly if you were ever to lose that army. Example: in a quiet macro tvp without any engagements, you should have a maxed out army by the 17-19 minute mark. After 3-5 more minutes, you should have some 10-12 barracks ready to pump out MMG after your major engagement. It uses both - see the formula. It is measuring how well you are spending relative to all other players at the same level of income. It doesn't take any other factors into account, so it needs to be averaged across many games to be truly meaningful. | ||
Primadog
United States4411 Posts
Hope to see more statcraft from you in the future. Let's make e-sabermetrics happen. | ||
MaV_gGSC
Canada1345 Posts
| ||
romnomz
United States54 Posts
| ||
| ||