The Curious Case of soO's Macro Mechanics

There has been a lot of discussion in the last few days about macro mechanics and their role in SC2. A lot of statements were made without context or argument. In this article I will argue why Macro Mechanics are fundamental to SC2 and discuss the impact they have on gameplay.

Blizzard's Statement

Here is the original statement. I’ve broken it down into points and counterpoints:



Queen energy shows how many times you’ve missed an inject. 25 means you missed 1, 50 means you missed 2 and so on. The production difference is also keenly felt in the mid and late game. We have seen many a zerg die to pressure simply because they did not have enough larva—lack of queens or poor injects liable.



The opponent isn’t supposed to know what you’re doing. That’s the entire point of fog of war and limited information. Both concepts are fundamental to the RTS genre. However if they scout they could see the energy on your cc/nexus/queen, the number of larva you have banked, the building you are chronoboosting or the number of mules mining. The same argument applies for almost every other facet of the game. Your opponent doesn't know how well your economy is doing—until he scouts your mining bases. Your opponent doesn't know how well your upgrades are doing—until he clicks a unit to check. Yet that is exactly how it should be.



Every subtle action a pro performs does not need to have some visceral or visual effect. Most subtle things pros do can’t be seen or understood by the average viewer. It’s the same in every competition: Sports, Poker, Chess, LoL, CS, Dota2, FGC, Smash. Subtleties add to the richness of a sport for competent viewers without harming the experience of novices—since, as they noted, novices don't notice. That isn't an argument for removal.



Cutting something because it’s too difficult isn’t an argument. It is important to find a balance between difficulty, purpose and effect. If an action is too difficult for its purpose, then a tweak might be necessary. If an action is too easy for its effect, then a tweak might be necessary. Removing an action simply because it is difficult, without analyzing its relationship with its purpose and its effect, is extremely shortsighted. These macromechanics also do not serve as a barrier to entry because players can play without any knowledge of inject, mules or chronoboost. Learning them, however, makes one a better player. Removing it because it is difficult will have no effect on making the game easier to play for a wider group of people.

Community Complaints

Beyond that, Blizzard gave no additional reasons as to why they are considering cutting or simplifying macro mechanics. However, these are the main points other people have brought up in the ensuing discussion:



Being “not fun” is not an argument. Making SCVs or depots/pylons/overlords is not fun. Getting cannon rushed is not fun. Getting surprised by mutas as protoss is not fun. Getting DT rushed is not fun. The fact that it's "not fun" to play against these strategies is no reason to cut out these units or builds. In fact, the assertion that these macro mechanics are not fun is flawed. How many times have we been excited by Maru, on his last breath, mule a base with abandon to give his economy an adrenaline injection? How many times have we been awed by soO gathering just enough units in time to repel INnoVation's advances? How many times have we been held on the edge of our seats as PartinG's crucial chronoboost allowed his +1 to finish before his forge died? All aspects of the game can be fun or not fun depending on the circumstances, and using it as a reason for removal—again, without investigating its purpose/effect—is folly.



The second argument is that mechanics—and therefore macro-mechanics—play too large a part in a player's victory or defeat in a game that's supposed to be about strategy. This is only true if there is a large disparity between player skill. Here is a broad list of most of the greatest SC2 players of all time and the current era (in no specific order): HerO, herO, Dear, Classic, Stephano, Bomber, sOs, Leenock, Soulkey, PartinG, DRG, MKP, Maru, soO, Rain, INnoVation, Zest, Nestea, MMA, Polt, MC, Taeja, Life, Mvp, Dream, Dark. In that entire list, only four players dominated parts of their field mechanically: Mvp (2011), DRG, MKP, INnoVation. Yet all of them were strategic or tactical innovators on a large scale.

At the pro level of play there is a base level of mechanics that you need in order to compete. After that it is up to the player to determine where he dedicates his APM and attention. These minute decisions are almost imperceptible but they often dictate the outcome of a match. The reason is because mechanics without strategy is nothing. There have only been two short periods where a purely mechanical player was at the top: MKP’s brief period of dominance in 2012 and INnoVation's early half of 2013. Beyond those two exclusions, mechanical skill has very rarely been the deciding factor as to who were the best players of a given era.

Even in the second tier of pro gamers, mechanics is often beaten by strategy or tactics. Take for instance Cure. Cure is arguably the 3rd most mechanically sound player after INnoVation and Maru. He might even be better than them as he makes more units faster than anyone and almost never gets supplied block. He played even against Hydra playing KR to EU after already playing 20 hours without sleep. He barely beat Snute to qualify for IEM Katowice only with his macro mechanics alone.

He is leagues ahead of other second tier terrans like SuperNova or FanTaSy. Yet both SuperNova and FanTaSy, for most of HotS, had far better results than he did because both of them were better at strategy and tactical execution.

Cutting macro mechanics to make the game more rewarding towards the strategic/tactical side is in my opinion a flawed one as the beauty of SC2 comes from the balancing act of mechanics and strategy and tactics.



The last argument is that it creates an artificial skill floor and it is just mindless APM spam. Both are wrong for various reasons. First of all, both terran and protoss macro mechanics require decision making. Terran has a choice between using the mule or saving for a scan. Many games have been decided by the simple act of muling at the wrong time. Later on in the game they can go for an entire mule economy and create a large army or use the scans to lock down a protoss (though only Taeja has ever done the second). Protoss have to use chrono boost strategically based on the situation. Chronoboosting probes leads to a better economy but slower tech. Chronoboosting warpgate early gives a timing for pressure. Chronoboosting upgrades ensures you have more efficient units. It only becomes negligible once the protoss gets to 3-4 bases.

Zerg: A Special Case

As for zerg, the answer is much more complex. First you need to understand my definition of macro. Macro is the management of your bases, building, production and economy and how that all feeds into your strategy and tactics. So if you cut macro mechanics for terran or protoss the effect would be small as both races must still take out a worker and build a depot/pylon and build production. Protoss will still time their gateway explosions while unlocking supply and banking resources for a large warp-in. Terran will still need to decide when to make production, when to cut units, when to put on add-ons, etc.

Macro for zerg works on a completely different scale. First, all units come from the hatchery. Second, you do not need to make a building to unlock supply; you make an overlord from a hatchery so you never have to do the extra actions of selecting a worker, pressing the hotkeys, selecting a location, and shift-queuing back. Finally, there is no end point to creep spread and inject. Those two mechanics can be done throughout the game ad infinitum with decreasing importance. To cut or put inject larvae on autocast is to simplify zerg macro beyond comparison. The inject larvae forces zerg to imitate the macro requirements of the other races (where they have to make production and pylons/depots). Additionally there is no end state for zerg. Eventually, a protoss/terran gets the maximum amount of production feasible. Zerg’s macro mechanics of inject and creep spread must be done for however long the game lasts, whether it’s 5 minutes or 2 hours. But in return zerg gets a larger strategical advantage the more larvae they bank and the more creep they spread.



Most importantly, all the macro mechanics create a baseline of multitasking in the game. This means that a player must make decisions and prioritize them. A player must judge and evaluate the benefits of microing a fight compared to injecting larvae or making a depot or chrono boosting upgrades. All of a sudden the macromanagement of your base and economics is now also a tactical decision that impacts your overall strategy. Inject is not mindless APM because it forces a player to make minute decisions that can have reverberating effects on how the rest of the game can play out.

It cannot be exaggerated how large an impact autocast inject will have on the viability of harassing zerg as terran and protoss. There is a rhythm to how every SC2 player plays. They tab to their base, make drones, check supply count, make supply, check army, check production, think about when/how they want to attack, do that move, repeat and recycle. That is why counter attacks are so strong. They do more than just kill workers/bits of army. They are mental attacks meant to take your opponent out of their comfort zone. If you take out inject, all of a sudden that 40 second rhythm that protoss/terran can attack is gone and zerg will have a much easier time defending light harassment/full on attacks.

The concept of "indirect damage" is often mentioned in XvZ matchups, and the reason why it's such a big deal against zerg is because of the race's macro mechanics. A terran or protoss can do light harassment knowing that if they aggravate or distract the zerg, they could force the zerg to miss an inject or build units they don't want (which they can't cancel). In tangible terms, this means the terran or protoss will have 4 fewer "units" to deal with in the immediate future. In the early game this could be drones; in the latter stages, it could be reinforcements or crucial spellcasters. The punctuality of injects in a given game can be the difference between hitting the perfect timing and falling into a trap. Removing injects would effectively alter the tempo of XvZ by easing the tug-of-war between the zerg managing his economy and the terran or protoss trying to shatter his rhythm.

Another limited resource that is rarely acknowledged is attention. There are many different aspects to attention: highlighted or "active" units/buildings, screen positioning, minimap awareness, game prompt awareness, and screen awareness. With injects, that is one more unit that zerg has to select, one more skill a zerg has to activate, one more location that the zerg has to visit, a few more split seconds that he has to count his larva, and a few more split seconds that he has to reorientate his internal timer. Multiply that with the number of bases and you can feel how much attention a zerg must pay to his hatcheries. Take that out and there is suddenly an abundance of attention freed for other actions. The common argument for this is that it will allow the zerg to make more interesting actions such as counterattacks. However, it also reduces the incentive for harassment against zerg because a.) he now has more than enough attention available to deal with everything; and b.) indirect damage is nerfed. By hypothetically enabling the zerg to do more interesting actions, removing injects will eliminate interesting actions and interactions from protoss and terran. It is a solution that "solves" one problem while creating two others.



It is important to note that attention as a limited resource also applies to terran and protoss. They have their own minutiae to deal with, and rhythm and tempo also applies to matchups not involving zerg. Yet it is the clearest example of how removing macro mechanics inadvertently diminishes indirect damage and how making something easier for one race will make other things harder for the other two. The impact of changing macro mechanics on attention as a limited resource should also be discussed—how players manage their attention and how they try to dictate how their opponent manages theirs—because it is subtly one of the most interesting things about Starcraft strategy.

Finally, even at the top level there are almost no zergs who can truly inject perfectly. After the third hatch goes down, inject efficiency starts to decrease with every additional hatchery. Autocast will artificially increase the skill of every zerg player so that they can make tech switches, reload instantly or play mass ling/bling/muta without fear that their larvae will be unable to sustain their strategy. It is possible for two zergs to play exactly the same game with the outcome determined solely by their larvae count during battle.


The best example of this was GSL Season 3 2014. More specifically Flash vs Solar and Flash vs soO. From 7/29/14 to 8/31/14, Flash had destroyed Proleague. He destroyed the IEM Toronto Qualifiers, he destroyed GSL Ro32 and he won IEM Toronto. He had a 79.03% winrate and was 23-2 in series wins. He was playing the best SC2 of his life. It could even be argued that he may have been the best player at the time. It is with this context then that we look first at his game vs Solar.

In this game Flash made a mockery of Solar. Solar got an early lead as he easily thwarted Flash’s early hellion marine medivac pressure and was able to easily get to the 4 base ling/bling/muta. Yet even with the lead, Solar was crushed without a fighting chance. For context, Solar was easily the second best zerg at the time. He had just gotten Ro8 the season before and would eventually be in the Ro8 again. He got second place at IEM Shenzhen only losing to Taeja.

Now take a look at the best zerg of the era, soO.

This was played right after Flash had won IEM Toronto. In this game, Flash killed 29 drones early on. He then pushed and got murdered. He pushed again and got murdered. He pushed again and got murdered. soO did what Solar could not. He dominated Flash in mechanics. He perfectly hit every inject because he understood that the answer to Flash’s BioThorBat style was mass ling/bling—the most larvae dependent style possible—with some mutas for support. And he was far more decisive. The second Flash stepped onto creep, he immediately attacked because he understood two things. First the strength of Flash’s army could not be realized unless it was spread out. Second, he could commit to these seeming hail Marys because he had the mechanics to inject every fight and knew a flood of ling/bling would arrive on time before the fight even ended.

What was even more telling was their 2nd game:

On Merry Go Round, soO loses his 4th before it ever starts mining, while Flash secures his. Usually this would be a winning move considering Flash’s execution and better upgraded army. Instead soO fights for 10 minutes and nearly routs every army that comes at him with mass ling/bling/muta. To do this he had to hit nearly every larvae inject off of his 4 hatches while simultaneously trying to spread creep, scout, counter attack and delay Flash. The only other player that could have even played a game like this was Life at peak condition. From his rise in 2014 to this very day, soO has had the best injects of any zerg bar none. While his mechanics aren't as easily appreciable because inject doesn't have a visual check like creep spread—which he doesn't do as much as Scarlett, for example—, it doesn't matter. He focuses all his attention on hitting injects and making decisive attacks, and the result is there for all to see. This is the reason why he is one of only two zergs to fight off creep and consistently win against large terran armies (the other being Life).

Recommendations

I’m against cutting macro mechanics. They may not be visible to the viewer, but they add a level of depth and complexity to macromanagement and overall strategy. The effects they have on gameplay are visible for everyone even though the root may not be obvious. And I’m not sure why the idea is to simplify the macro mechanics. Instead, add more strategic depth to the macro mechanics. Terran as it is now is really good. They have to think about when to mule/when to scan. In the end game, it allows them to sac SCVs to make larger armies or to use scans to win the game positionally. Protoss chrono boost loses relevance as the game goes on so it’s fine as it is. Perhaps if Blizzard wanted to add depth to it and make it more useful in the late game that could work too. For zerg, inject is not just an add-on mechanic, but core to everything a zerg does and essential to how modern XvZ is played. It balances out the mechanical requirements as well as strategical/macro needs. If anything, another skill could be added to the queen so that there is more decision making involved.



At the same time inject does have some problems. Because of the way a zerg's production is staggered, the outcome of attacks can often be extremely polarized. Attack when he has just made a round of drones and he will have no way to defend. Attack just as larvae spawn and he will definitely have enough to overwhelm you. Most players are reluctant to venture on creep unless they have an army large enough to hold their position, a safe way to retreat (medivacs or recall), or the intention to all-in.

Ideally, a midpoint should be found that eases zerg's staggered production without removing what makes it unique. How larva is spent in the early game has too significant an impact because of spawn timings, while banking larvae in the mid game while waiting for the perfect counter discourages most pressure builds. If zerg's inject is to receive a balance rework, it should have the purpose of reducing early game situations where no larva is available in exchange for reduced incentive to stockpile larvae in the midgame.

At the same time inject may be too important for zerg. As mentioned above, soO's ability to inject even under duress could be pinpointed as the source of his wins against Flash. One aspect should not be credited with victory so easily. One aspect should not influence the outcome of games so heavily. However, removing it outright is not a solution as discussed above. Instead, tweaking its effects while giving the zerg some other mechanic that will allow players like soO a chance to shine should be our goal. Rather than the complete removal of all macro mechanics, a revamp of zerg's could be the answer.

Conclusion

I feel that SC2 as a game should be embracing subtle complexities, not shying away from it. What makes SC2 great in my mind is that there is no real victory condition. In every other esport there are objectives. In LoL/Dota2, there are clear road signs for what you should be doing next. Farm your items. Take the towers, take rosh/baron/dragon, take the rax/inhibitor. Take the throne/nexus. In CS you take 16 rounds to win the game, with only 4 possible outcomes. In FGC/Smash you KO or send the other guy flying off stage. In SC2, the victory conditions are destroy all opponent buildings. Yet that almost never happens except in Polt games. Every player is given tools on how they want to win. Then they decide how they want to win whether it be by deathball, economic starvation, constant multipronged attacks, parade pushes, doom drops, base trades, all-ins, counter build-orders or micro. SC2 is a game that forces players to find the path of victory for himself and through that struggle, through that journey, we come to see our own individuality in our games.

To take away a core tool, to simplify it so that everyone else can do it not only hurts the strategy, but it also hurts the very essence of what makes SC2 great. It takes away another aspect from which players can differentiate themselves from their peers.

In a way this reminds me of the transition from WoL to HotS with the addition of Medivac Boost. While it was a needed tool for terrans to retreat from large battles, it eventually hurt two of the most stylistic players to have ever played the game: MMA and GuMiho.

Because for them, victory was attained through multitasking, coordinated chaos, a 5-7 pronged attack from all sides against a zerg or a protoss. Any terran could win, but only GuMiho or MMA won by creating 5 simultaneous drops. But once medivac boost was implemented, it was easier to do and far less risky. Every terran could emulate what they did without the strategy, without the risk, without the hard work. It took years for GuMiho to find a way to express his kind of madness and win his games. And while MMA was successful, his games in HotS never had the incredible flair and chaos his WoL games had.

I understand the need to try to simplify the game to make it more user friendly, but this change to inject will make everyone have the injects of soO without the mechanics, without the strategy, without the hard work.

If this went through as is, then everyone can be special like soO.

And because everyone is special, no one is.