|
On May 22 2013 05:46 Grumbels wrote:Show nested quote +On May 22 2013 05:40 whacks wrote:
Awesome results. Does confirm my gut feelings that I'm focusing too much on hitting my injects, and not enough on the other things I'm missing on.
It's funny to see so many people interpret this study completely wrongly. The study isn't intended to say that silvers are as good as Masters. Obviously, the people on Masters are much much better, and they can macro much better as well. The entire point of the analysis, is that hitting your injects is not a major differentiating factor. All the other things people mentioned, like taking more bases, building macro-hatches, spreading creep, not being supply blocked, keeping your resources low ... these are equally/more differentiating factors, and these are the things people may want to focus equally/more on. The type of games in master vs silver is so different that comparing inject rates is meaningless, so you can't tell whether injecting is or isn't important.
Did you read their full results? They have data broken down by division, all the way from Silver to Masters. If you want to compare Diamond to Masters, or Platinum to Diamond, that's all there in their results as well.
A while back, someone collected similar data for workers-produced, time-supply-blocked, and surplus-resources-banked. Their results showed vast differences between the different divisions, implying that the above metrics are a major differentiating factor between the different divisions & skill levels. The empirical approach does work.
|
On May 22 2013 06:03 plogamer wrote:Show nested quote +On May 22 2013 05:47 TheRabidDeer wrote:
I firmly believe that this study has data wrong. There is no way I hit injects THIS MUCH better than other players.
71.48% injects over 48 games
19.96 minutes average game time
The average masters player (the league I am in) is BELOW 60%. SC2 is that awesome. There will be players who excel through a focus on macro, like you, who injects more. Then there will be those who excel with a focus on tech-timings, or unit-control, or multi-tasking. Someone who excels in all of the above, i imagine would be in grandmaster league or pro level.
I have always known I had great mechanics (it is the only reason I did well in BW, I knew shit in that game and scouted horrifically) but I just cant believe that my mechanics are on par or better than a pro. I mean, that would mean I would have a chance at being GM if I were to allow time for my wrists to heal and then work on unit control.
Regardless... what it means (for me) is that I should probably try to exploit ~15 minute timings since I will have (on average) at least 12 more usable larvae than most people.
|
The point about number of injecting queens not being controlled for is a good point. Wonder if that would make a big difference especially early to mid game?
Also zerg's APM spikes are completely overblown in the difference that they make. Look at those spikes as a % of actions in a whole game, its a small effect at best. Zerg is an APM intensive race. Terran might be more APM intensive but this is reflected in that the top terrans tend to have higher APM than the top zergs. For example Yoda has a higher APM than any player I have ever seen.
|
I swear, every time someone provides some data point, instead of sparking interesting debate, half the responses are like your data is wrong, my intuition is perfect and therefore i reject your findings.
|
On May 22 2013 06:06 plogamer wrote:Show nested quote +On May 22 2013 05:59 TheRabidDeer wrote:On May 22 2013 05:56 jpditri wrote:On May 22 2013 05:47 TheRabidDeer wrote:
I firmly believe that this study has data wrong. There is no way I hit injects THIS MUCH better than other players.
71.48% injects over 48 games
19.96 minutes average game time
The average masters player (the league I am in) is BELOW 60%. I would consider that their conclusion would indicate you could spend less time focusing on your injects and get better results focusing on other aspects of the game. I do think their conclusion is accurate-- not only are later tier units more larvae-efficient, but better players are more efficient with their units, and the fewer units you lose the fewer larva you need to replace them. My micro does suck, but I have a hard time with micro because my handspeed is low because of wrist issues. However, the less than 1 second required to inject does not impact my micro. I also refuse to believe that pro's would have such a low inject rate when they have such high APM. This stat does not have GM league due to sample size. It is more representative of the overall ladder however. I don't think pros have such low inject rate. BUT, I recall watching plenty of Stephano games where he wasn't keeping near-perfect injects - and he still won (against a foreign protoss pro, but still pro level). He won through fantastic engagements and map control.
Injects are basically most important until you reach your desired drone saturation. After this point engagement control and map control, like you say, take precedence.
The room for further macroing will hit a ceiling very fast in SC2 so I don't find the results very surprising. There's not much you can do past the 12-ish minute mark aside from controlling what is now your presumably maxed army while looking for a good engagement.
Had larva inject supplied only 2 larva, had SC2 been a slower paced game, or had SC2 had a higher max cap than 200 I am fairly certain there would be a greater discrepancy between skill levels in the late game. The way it is now: there's no potential for taking further bases, there's no potential for building more units (you max out). So what is left to do? You control your army. Naturally the rate of injects will drop off towards the late game.
Although I would think progamer matches where there are frequent trades and constant action will tend to show different results. It's very easy to get into a lull in the late game. And I'd imagine lower skill levels get in this lull more frequently.
Although I'm really just guessing. If you watch Stephano's recent WCS ZvTs on for example Star Station, I very much doubt his inject rates late game are higher than a silver leaguer's. He basically spent minute 15-30 in the game maxed out on 3 bases while desperately trying to secure a fourth and while at all costs avoiding an unfavorable engagement.
This is what you get when you design a game with a capped ceiling. The relevance of certain mechanics fade with game length.
Thank you dsjoerg!
|
I think one reason is that at higher level player you are keep under constant pression with drops and harras, so it's harder to hit perfect injects.
For experiment may be interesting to see the inject % in a match between a master and a silver..i think in this case it will be very close to 90%
|
On May 22 2013 06:14 BreakfastBurrito wrote:Show nested quote +On May 22 2013 05:47 TheRabidDeer wrote:
I firmly believe that this study has data wrong. There is no way I hit injects THIS MUCH better than other players.
71.48% injects over 48 games
19.96 minutes average game time
The average masters player (the league I am in) is BELOW 60%. If you read in the article the standard deviation for injects is quite large, you are about 11% over the average, or 1 standard deviation from the mean... not that ridiculous Besides, i'm sure your % drops just as much as the other players as game length increases... You can't just discredit a study because it doesn't feel right, lol. If you find some actual flaw in the data, then that would be something.
I am not discrediting the study, but the data that the study used. Either via the tracking or other methods. One thing that I found when looking through my games was the issue of a queen dying. If a queen injects a hatch, then dies, it is considered to be not injecting. I had one game that I threw out because my queen injected, then died due to an early pool zerg and I was unable to afford a queen until 10 minutes which meant 5 minutes of "no injects".
Also, the SD seems to be quite high.... 11% SD means there exist people (even in masters) that have 11-17% inject rates at 30 minutes or 100% perfect injects at 20 minutes.
EDIT: If the data is right, it puts me in the top 200 zergs in the world (of masters zergs) when it comes to injecting.
|
On May 22 2013 04:13 darkscream wrote:Show nested quote +On May 22 2013 04:02 Embir wrote:
Terran is the most difficult race mechanically
Zerg ..., it is just the easiest to master, it is the most noobie friendly race by far this mentality is disgusting and a cancer on the entire starcraft 2 scene. Guess which race has the highest APM on average? No, it's not terran.. we can cherrypick stats all day long.
you do know that zerg apm spikes whenever you press S+ ZZZZZZZ or DDDDDDD etc right? There's a reason why only koreans can play terran successfully and white people play zerg l0l. Clearly it's because zerg has a higher mechanical skill cap...oh wait.
|
These data are indeed remarkable, who would have thought the percentages would be so close?
There is one important thing missing in these data though,(not sure if mentioned before, didnt read the whole thread only read the article) something wich could explain why the percentages are so close.
I would asume that master players on average have a higher number of bases/hatcheries then silver players, wich could effect the conclusion and the data.
It is much easier for a silver player to keep injecting his one base timely, then it is for a master player who has 5 bases to keep up with,
It is possible to take this into account in the study, for example compare the data where master and silver players both have the same amount of bases/hatcheries, i think you will then find a bigger difference in the percentage of inject active times and injecting properly might be more important then this study shows ,it would be realy interesting to see the results of such a study though it probably would not be that easy to do such a study.
Still i think it could be an important aspect wich might explain the small differences shown in this study.
There is one more thing wich might skew the data a bit,if i remember correctly from watching streams and tournaments manny master players use the first inject of the 2nd queen to make a creep tumor instead of injecting their expansion(because they then dont have the monney to make annything from the larva or something) this happens early ingame and sets back their inject active time by a lot, a silver player in this situation might simply inject (or not have an expension at all..) and not make the tumour (and then also not be able to make annything from the larva because he has no monney)
This could also explain partially why the figures are so close.
|
You can actually play pretty decently with no injects
|
Very interesting! Thanks for sharing this research.
|
On May 22 2013 06:41 rysecake wrote:Show nested quote +On May 22 2013 04:13 darkscream wrote:On May 22 2013 04:02 Embir wrote:
Terran is the most difficult race mechanically
Zerg ..., it is just the easiest to master, it is the most noobie friendly race by far this mentality is disgusting and a cancer on the entire starcraft 2 scene. Guess which race has the highest APM on average? No, it's not terran.. we can cherrypick stats all day long. you do know that zerg apm spikes whenever you press S+ ZZZZZZZ or DDDDDDD etc right? There's a reason why only koreans can play terran successfully and white people play zerg l0l. Clearly it's because zerg has a higher mechanical skill cap...oh wait.
As I said already this is a very small % of total APM and the difference this makes is very small. I dunno why this ALWAYS come up when discussing APM.Zerg unit spam makes a tiny difference only, its an APM intensive race.
|
This is some pretty interesting data but I think there are several other conclusion which can also be drawn from this. As been mentioned before, in higher level games its a lot more common that drops and other harass interfere with your macro and queens are often used for defence. Sniping queens to prevent injects is on top of that fairly standard.
A fully-saturated base gathers 861 minerals/minute and 242 gas/minute = 1103 resources/minute. A queen can generate enough energy to inject larva once every 44.44 seconds, which creates 4 larva. So that's 11.11 seconds for a larva, or 5.4 larva per minute. Together with the 4 larva per minute from the hatch itself, that’s 9.4 larva per minute, which is not quite the 11 larva per minute we'd need.
But if we have any macro hatches, or any hatches that are mined out, or any bases less than fully saturated, or we're making units more expensive than 100 resources per larva, then we don't need perfect injects.
While commenting on this I would first like to say that I can understand the need when it comes to simplifying gas and mineral cost and bundle it up by calling it resources. However, the assumption regarding larva generation is a bit flawed.
The 4 larvae generated per minute you get from the hatchery itself only occurs when you're actually spending those 4 larvae. If you dont inject your hatchery and dont spend your larvae making units it will never have more than 3 larvae sitting there waiting for you.
The higher level of play the more likely it is that a larva is used as soon as it spawns. In lower levels its not uncommon to forget building units which will then limit your larva production.
Because of this you will actually need to inject more to make up for the loss in larvae production at lower levels.
Master players still display a higher percentage of inject uptime for most games with a fairly standard duration. Including earlier hatcheries the total larva production gained is probably quite big. All this despite the fact that they're most likely losing more queens and inject timings from harass. Blocking your ramp with a queen against hellions simply appear to be worth the missed inject(s). Then again, I dont find that very surprising.
As was noted under "Directions for Future Exploration", queen energy would also be very interesting, knowing when to skip injecting and spreading creep, or simply when to move out with queen support to transfuse is most likely quite important.
Last but not least I feel that the rather large value of standard deviation within the leagues deserve extra mentioning. The fact that roughly 30% of the master zerg games show a lower inject percentage than the average number in the silver league does fist of all not mean that 30% of the master zerg players have worse injects than than average silver player, something which can be wrongfully interpreted when reading the following:
That means[1] that about 30% of Masters Zerg games have an inject % that’s worse than the average Silver inject %. They did worse than the average Silver game, yet somehow they are in Masters with sub-Silver injects.
Instead this most likely indicates that the games themselves and the tactics used by both the zerg player and the opponent plays a huge role when it comes to both the importance of injects and the ability to inject properly.
Edit: Great work though, I realised I sounded a bit negative and I'm sorry about that, it certainly wasnt my intention.
|
I would say this is to be expected. In lower league games there isn't much going on so hitting all your injects is fairly easy (plus it's not like the only thing most people tell a low league zerg is to just hit his injects), while in higher leagues it's a lot harder to hit every inject given that there is so much more multitasking involved. Also I find using a site that doesn't even track all games in conjuction with implying that diamond/masters players are actually high level is kind of a weak source. On top of that your inject numbers is based on hatches, and I may have been watching and playing the wrong rts here, but afaik you don't get a queen for every single hatch you ever build during a game, now do you? So since lower league players build less hatches it's obvious that they'll do fine in terms of injects.
Well as always you can't trust a study you didn't fake yourself, but this one pretty much says nothing. If you have a spell that gives you more production it will always be very very important to use that spell perfectly, and nothing said in this thread has any influence on that fact.
|
Dsjoerg, Blizzard said they'd add more data to replay files. Do you know what exactly they're gonna add (or have added, I'm not in the loop. Have they already implemented this patch?)?
If replays contained supply data it would be interesting to plot inject rate against the average supply during a game (assuming this is possible to calculate; draw/collect supply every x time interval and average it).
I would bet that that out of two 25 minute games, the one with the lower average game supply would generally have the higher inject rates.
|
When people talk about the importance of larvae injects, 99% of the time they are speaking about it in the context of macro. So when you isolate the stats that ONLY speak about the frequency of injects (nothing to do with actual production of units), it really doesn`t tell me anything in terms of macro or importance of injects. So I really can`t say these stats tell me anything of value when it comes to the importance of injects.
|
I am in no way condoning the research and work that you put in this article, the data alone is pretty interesting, but the analysis and 'conclusions' are just way off and can be very misleading.
As others have mentioned, inject percentage comparison is meaningless as master players need to put up with far more things throughout the game, you need to give numbers a lot more perspective, cross-relating harrass, multitasking and actions with injects will give you a far better scope of what to look for in these statistics.
If we looked at professional car racing and had a percentage of accidents per competitor and then compare it to the percentage of accidents from an amateur or casual driver we could see that they'll maybe have a similar percentage, but this in no way means that the amateur is near the level of a pro driver, it may be because a pro will have tougher competition, trickier tracks, and many more things to factor in. This is why a correct approach to the numbers is far more important than just having a bunch of data and immediately assume that your comparison is correct.
Because Starcraft is a game that has an enormous amount of variables to account for, you'll have a ton of variation in replays and scenarios when the winning factor is more than just a 'key mechanic', how much of this silver level percentage would drop if he was facing a masters opponent? Probably a lot. Therefore, even if we're talking about the same 'metric' (inject percentage) it's just bad practice to even suggest that those percentages can be compared and have a valid conclusion.
The problem with these articles is that many people don't even stop to analyse and just take it for granted because there's a lot of data and a nice spreadsheet and then try to account the data as 'proof' that a race is easier or harder than the others, which just invites more ignorant discussion.
TLDR nice idea, great effort, cool statistics, pretty blind analysis and conclusions.
|
Its because once you get into late game, zerg switches from being cost ineffective to incredibly cost effective. It only takes about 8 larva to roll over someone with a wave of ultras.
|
I think this is good research but the results of the data are analyzed in very narrow focused way.
Its not a helpful find to know that as the game goes on i have to worry less about injects, but what is really interesting is the implications of the graph they supplied.
with Starcraft being a game of gaining advantages we're all always looking for ways to get ahead during all stages of the game. This graph shows us that macro mechanics find their best advantages where the difference between leagues is the greatest, so about the 7-20 min range.
A good way to think of it in my mind is that advantages are earned the following ways
early game: crisp openings and proper development of economy(for mid game)n
mid game: Smooth and efficient macro. Powering up an army while developing upgrades for late game (~8-20mins roughly)
late game: Proper army composition, control and upgrades
With timings and cheese working to exploit holes in development in army or tech before late game
|
On May 22 2013 07:40 Shousan wrote:
I am in no way condoning the research and work that you put in this article, the data alone is pretty interesting, but the analysis and 'conclusions' are just way off and can be very misleading.
As others have mentioned, inject percentage comparison is meaningless as master players need to put up with far more things throughout the game, you need to give numbers a lot more perspective, cross-relating harrass, multitasking and actions with injects will give you a far better scope of what to look for in these statistics.
If we looked at professional car racing and had a percentage of accidents per competitor and then compare it to the percentage of accidents from an amateur or casual driver we could see that they'll maybe have a similar percentage, but this in no way means that the amateur is near the level of a pro driver, it may be because a pro will have tougher competition, trickier tracks, and many more things to factor in. This is why a correct approach to the numbers is far more important than just having a bunch of data and immediately assume that your comparison is correct.
Because Starcraft is a game that has an enormous amount of variables to account for, you'll have a ton of variation in replays and scenarios when the winning factor is more than just a 'key mechanic', how much of this silver level percentage would drop if he was facing a masters opponent? Probably a lot. Therefore, even if we're talking about the same 'metric' (inject percentage) it's just bad practice to even suggest that those percentages can be compared and have a valid conclusion.
The problem with these articles is that many people don't even stop to analyse and just take it for granted because there's a lot of data and a nice spreadsheet and then try to account the data as 'proof' that a race is easier or harder than the others, which just invites more ignorant discussion.
TLDR nice idea, great effort, cool statistics, pretty blind analysis and conclusions.
Their theory and analysis is solid. Rather it's posts like yours, filled to the brim with pseudo-statements, that provide the blind analysis.
Let's see what their analysis is:
1. larva to resource ratio decreases as the game progresses and you tend towards using higher tech units (ultras, broodlords etc cost both more resources and supply).
2. Mined out locations keep supplying larva. Less risk of "larva blockage" late game.
I think they are solid explanations. Only thing I'd like to see added is that you spend a considerable time maxed out during the late game (partly covered by larva to resource ratio and larva blockage explanation). As they state in their analysis: When you resupply and/or remax lategame you tend to resupply on high tech, high resource, high supply demanding units and less so on cheep larva demanding ones.
I can't spot what's so blind about their analysis? Apart from the fact that you and most others similar posters in the thread probably didn't even bother reading it. The truly blind ones are those who don't bother addressing specific points but rather use vague language to cover up their contribution to an "ignorant discussion".
|
|
|
|
|
|
EPT
