EPTThe issue with starting with gas right away is that new players cannot handle it. If we start with gas, then the initial learning curve is more steep. Additionally, sticking to step 1 all the way through helps players set up a good long term mentality: that losing is good, and winning is good. The ladder will make it so that you will lose 50% of your games, regarless of your situation. This is the part of TheStaircase that players say help get them over ladder anxiety.
[G] TheStaircase - An Alternative Improvement Method - Pag…
| Forum Index > StarCraft 2 Strategy |
|
JaKaTaKSc2
United States2787 Posts
The issue with starting with gas right away is that new players cannot handle it. If we start with gas, then the initial learning curve is more steep. Additionally, sticking to step 1 all the way through helps players set up a good long term mentality: that losing is good, and winning is good. The ladder will make it so that you will lose 50% of your games, regarless of your situation. This is the part of TheStaircase that players say help get them over ladder anxiety. | ||
|
TheKleszcz
Poland11 Posts
 Losing against those tanks was good for me too - I realized I need something else to beat tanks. It seems that different skills grow at different rate and even though my macro is still lacking I decided to push the D button for Marauders rather than push A for Marines. Games are more fun now and I'm learning how to adjust sightly. I'm looking for a balance between using your system and not playing mindlessly into wrong situations. For me it's also better because I do not have to handle an artificial situation of pulling SCVs away from the Refinery once the gas starts going through the roof - it's a quirk created by the Staircase point system in the time after you've stopped doing only mineral stuff, but before you can spend gas on units, and after you had your share of time going with one unit only. After all it's only a game and a flexible approach is better than the game becoming dull and rigid. | ||
|
JaKaTaKSc2
United States2787 Posts
| ||
|
Becuula
Germany65 Posts
The goal is to be not supply blocked and constantly build workers until full saturation is reached. It's hard to believe that there is a way to measure that with saturation speed. It depends on the strat and if you are being harrased or not. I think it would be more accurate and easier to understand if there is a benchmark like "time not building workers" and "time supply blocked (while not maxed)". | ||
|
fengshaun
149 Posts
| ||
|
JaKaTaKSc2
United States2787 Posts
If you're down to help, we'd have to start compiling a list of all strategies and order them in skill/power ratio. To do that we'd want to find an objective way to analyze the amount of skill required to execute a strat, and what league/rank the strat would get a player to. It would have to be updated fairly often, as strats become more and less powerful over time (though I think skill would be fairly constant). @Becuula Constant worker production is only part of it. A player must also defend themselves from harassment. If I am playing zerg, I get get saturated bases very quickly, but I wouldn't be able to defend myself. Also, if I have many workers, but they aren't spread evenly throughout my bases I am not really macroing efficiently. The situation could arise that it is not yet safe to take a 3rd base, and our first 2 bases are already maximally saturated (24/3/3) or even efficiently saturated (16/3/3) and it would be better to use those minerals in order to secure a new base. There are probably other scenarios, but the main idea with saturation speed is it gets to the root of the mechanics. The point of producing workers is to increase income. By measuring income instead of workers we encompass all of the important aspects that create the correlation between worker count and win % instead of just worker count itself. | ||
|
Gauranga
India25 Posts
So how do I know if I'm meeting my Saturation Speed league goal or not? **********EDIT************ ok, I just found out you have to enable "TheStaircase view" in the options. Cool! Looks good now. | ||
|
Becuula
Germany65 Posts
At the moment I kind of ignore the Saturation badge because it's not easy to find out why it is only bronze. Have you some tipps how to watch the replay to find that out more quickly? If am playing TheStaircase mode and my type of units is limited, it can be hard to defend against harrasment efficiently. It's unsatisfying to not get a pass if at some point there are lings in your base just because you have no forcefields. | ||
|
TheKleszcz
Poland11 Posts
It's just like in the Bronze League Heroes match posted earlier - the guy crushed the Zerg guy with one macro Zealot wave after another mixed with sloppy micro, but needed some air in the end to overcome 12 Mutas. I guess the most important thing here to consider is that it's workable first to get a feeling that you've exhausted your options with current units. A Mothership Core time warp with Zealots catching Zerglings in some choke point or scouting more to catch the opponent with Marines while he's moving his tanks can be something easily overlooked if you go with "I should really build something else ASAP". JaKaTaK stressed this above - it's the learning that's most important. | ||
|
JaKaTaKSc2
United States2787 Posts
I think there is a solution to pretty much any problem. For your specific example, why not have a wall with a zealot in it? More importantly, why are you spending so long on the first step? Based on your response it seems like you are more focused on winning the game than you are on achieving the benchmarks (this happened to me a bunch when I was going through it for the first time). I realize that is very presumptuous for me to suggest. I only suggest it because it happened to me, and many players that I coach using TheStaircase. You might want to consider it. We will have more stats eventually!!! Our plan long term is to have Saturation and Spending as the major benchmarks and then have a bunch of minor benchmarks that effect 1 or 2 of them. Like you mentioned, if you have poor saturation it might be for a number of reasons, supply blocks, worker count, worker efficiency, poor harassment defense, etc. We feel that each major aspect in Sc2 can be boiled down to gathering and spending money. Which is why we have decided on these 2 benchmarks to start. But the long term goal is for the player to be able to click on their bronze badge saturation speed and see an analysis of their weak and strong points that affect their saturation. As far as watching replays goes it depends on your race. Protoss and Terran are fairly easy to start off. Try to constantly produce out of all of your facilities, build your supply structure ahead of time, and always distribute your workers evenly throughout your bases whenever your are past 16 workers on at least 1 of your bases. (having 24 workers on minerals in one base and 8 on another is terrible for your income.) If you are doing all of these things perfectly the answer likely lies in strategy. If you attempt to take a base too early, your saturation speed will suffer because you will not have sufficient units to defend. If you take a base too late, you will mine out one of your bases, lowering your income. I'd be more than happy to take a look at a replay where you didn't understand why one of your benchmarks are so low. In addition, we are going to be reworking spending to include gas. This should allow for more flexible strategic choices, and hopefully more separation between the leagues. (as it is now, oftentimes it will be bronze or masters and that is all). @TheKleszcz You're spot on with that last paragraph. Its easy to jump to the conclusion "the only solution is to change my unit composition" but this is so often not the case. A big part of TheStaircase is to attempt to break the player out of this narrow minded philosophy. | ||
|
VerdeCreed
United States27 Posts
In WoL I never used the zergling, I really banked on roaches as my bulk army unit. Suddenly, theStaircase has forced me to learn, and love, the powerful swarm unit. Enemy Skytoss is massing Voidrays behind a turtling wall? Expand all over the map, defend with queens early, and always threaten a run by. The speed and numbers of he little guys might just win you a base race when you have structures all over the map and they are easy to clean up on their main platform! Ling run bys in general put out buckets of pressure. I love watching a maxed, two base turtle enemy army march out across the map, (usually past my fourth and fifth base) to attack my mined out main. Meanwhile I begin gutting his main base with my hordes of lings while testing his micro by trying to snipe tanks unseiged in bad spots. Lings can be terrific problem solvers. Leave one, even unburrowed, near an enemy's third base for good intel. Mercilessly seize Xel'naga towers, and have a line of three or four lings at his ramp, mid map, and perhaps other common movement paths. Make a power showing and the enemy will usually be forced into making more units, or teaching further than normal before feeling safe to take an expo. With good micro lings can form walls, and I love trying to pop banelings with individual lings! So, at each step, taking the time to problem solve will really help mold a stronger StarCraft player. The more time you intimately get acquainted with a unit, the more creative answers you will have as you try and answer different problems that spring up as move up theStaircase. I saw you noticed that a zealot can be used to wall off lings instead of a sentry and To that dude that lost to tanks with his marine army and felt like marauders were necessary: Don't let a single defeat turn you off and stop trying. Your goal is not to beat a marine army with tanks, but learn how to maximize a marine versus a tank. Perhaps you can ignore the threat. Let him push, and then run towards his main to threaten a base race. With your good spending hopefully you are ahead in bases and even spread out your barracks. A base race should be a scary decision for your opponent to make. Perhaps when he pulls back you capitalize on a moment to pick off the slow moving tanks from the back of the army, or as I've experienced it, they forget to unseige at all and thier army leaves without them! Have you tried floating rax onto high ground and producing marines from inside his base? Push bunker rushing and bunker contains to the limit, even on a third base. What about ebay blocking? There are so many awesome, funny, and crazy things you can do without gas, but including it open up even more hilarity. Obviously you can get upgrades like stim, but what aout mass sensor tower to help you achieve perfect positioning? Or perhaps a Planetary Fortress and bunker contain? Often, mass amounts of low units outright kill someone who is greedy with expanding, often they will take two, turtle, then look to kill you, or use thier army to take their third. Make a point of denying that third, make them fight up a ramp and into a choke for it. Tier 1 units are more than adequate on most maps. Then just play the, "I'll contain you until you starve game." During this process you can learn what marines are really capable of doing. so the next time you're caught greedily expanding, and a tank and 20 marines and marauders roll into the Planetary fortress known as your third base, you'll know exactly how many marines it's going to take to hold! So, be creative and explore all your options to the fullest. At worst, they will fail and you'll know just how awful PT rushing with marines really is, but at the best you'll teach yourself to play creatively and discover the power of a creative solution to solve an old problem, like learning to use a zealot to block a ramp instead of always relying on a sentry! | ||
|
JaKaTaKSc2
United States2787 Posts
Well said. I fully endorse this comment. | ||
|
astazha
United States29 Posts
1) Give both an overview grade that is an average, as well as a breakdown for performance on 1st base, 2nd base, etc. This gives both the ballpark performance at a glance as well as the ability to tunnel in and see that the problem is the 3rd base, or whatever. Grade based on average for purposes of advancement up the ladder. 2) Create a system that works regardless of the mineral/gas mixture. Here's what I'm thinking. Grade gas separately - probably only track the 1st 6 geysers. If I build a geyser, there should be gas income from that geyser within a certain amount of time. The gas score would also have an average, and then a per geyser so the player could notice that they forgot to put workers into geysers 3 & 4 for a while, or whatever. (Also, maybe only have the 1st geyser require the 101 and then use an average after that - it seems unlikely that 101 * 4 geysers is going to be accurate, and would report as saturated even if a worker were missing.) The 4th geyser going up would create a benchmark to hit of 101 + 114 * 3 gas, or 443. It would only be possible to hit that benchmark with 12 workers in gas. Mineral income is still 640 per base, but: the benchmarks are compared only to replays from X league players who also had that many (completed) geysers at that time (when the income is achieved.) So, for example if I'm doing a 2-base 3-gas build, my mineral income hitting 1280 will be compared to the timings of players who hit a mineral income of 1280 after building their second base and 3 gas facilities, and this time will be (appropriately) slower than players who are on 2 base 2 gas and faster than players on 2 base 4 gas. 1 base income with an early gas will only be compared to similar, and so on. The goal, mineral wise, is still to saturate your minerals efficiently to 16 workers, and that doesn't change based on your gas choice - though the benchmark timing should. That system would accurately grade the player for 0 geysers, 6 geysers, or anything in between and without (much) regard to how early or late the gas is taken. This could be gamed with clever geyser timings, but presumably players are actually trying to improve and not learn to cheat the GGtracker. Also, if you built on 2 geysers right before hitting 640 minerals to try to get credit for a mineral benchmark based on 2-gas builds, then either the gas grade will be lower (because you have to build more workers to go into the geysers rather than pulling what you have) or the benchmark I'm about to describe will be impacted: Another potentially worthwhile benchmark would be whether that income is maintained after hitting the saturation point. Worker harassment, idle workers, zerg making building, or diverting workers to gas can all cause mineral (or gas) income to dip - permanently if not addressed. Maybe take the time from saturation point to the end of the game and report % of that time that the player maintained at least that income. This would encourage people to notice whether they are staying saturated after getting saturated. Some builds call for intentionally taking workers out of gas, but I don't think it's necessary to account for all possibilities. Not sure what to do with MULES. Hope that helps. Love this project. | ||
|
TheKleszcz
Poland11 Posts
 @VerdeCreed What is an ebay blocking? | ||
|
JaKaTaKSc2
United States2787 Posts
 We have to get to the point where we can change variables and check the results, then repeat. To do that right now takes a lot of time and resources. I have this habit of trying to make things perfect before they are done and then start to explain them and teach them. This approach does not work because we're constantly finding things to improve. dsjoerg has left my contact info to some people who have volunteered to do mass replay analysis. When that happens, we'll get a data dump from ggtracker and go to town on it. Until then, we're going to work on the presentation of the benchmarks we have now. This doesn't mean that we've stopped improving, just that we'll be improving while presenting. This will slow the progress of our improvement somewhat, but what's the point of a training system if only a few people have heard about it. | ||
|
dsjoerg
United States384 Posts
| ||
|
dsjoerg
United States384 Posts
| ||
|
Cook
France130 Posts
I'm looking forward to upcoming features and improvements to the calculation behind the benchmarks. Keep up the great work! About the saturation speed explained page: You can compare your speed to the average saturation speed for any matchup and league. If you are Gold or below, you should aim for the Platinum-Diamond average speed to give yourself an economic advantage. If you are above Gold league, you should be aware of the average Saturation Speeds, but of course in any particular game you may over- or under-focus on economy for strategic reasons. Does this mean that for the staircase levels, only the first base sat speed apply for player above gold league? e.g. I'm plat doing a one base all-in which damage my opponent but don't kill him right away (he loses his 2nd base but i can't kill him right now because he has a tank). I set up a second base for myself really late and win the game a few mins later. I have a diamond spending skill. A diamond first base saturation and a bronze second base saturation. Does this count as a game that's good enough to count in the 5-to-pass games or not? | ||
|
JaKaTaKSc2
United States2787 Posts
| ||
|
astazha
United States29 Posts
On to step 2. | ||
| ||
