Blizzard announced their proposed changes to the multiplayer mode in Dreamhack last weekend. I took some time to digest the content, and I think it says a lot what Blizzard want to do. There are several things I disagree with on the problem identification stage.
I will start off by singing praise to Blizzard’s continuous effort to make improvement. Well done. It is safe to say few of us expect that many changes even though Blizzard had been teasing us here and there. Perhaps the biggest surprise is the magnitude of some of the changes, as it feels like Legacy of the Void 2.0. Blizzard’s willingness to take risky but potentially rewarding steps at this stage of the game is commendable.
I must stress that these changes are not final, so let’s not get ahead of ourselves. With that in mind, I think it is more meaningful to look at the direction that Blizzard are looking at and what they intend to achieve instead of how these changes may affect the balance and match up. It may seem silly to bring up the thinking process behind the proposed change (see below), but it helps to keep to discussion focused.
Goal definition –> Problem identification –> Solution proposal
Blizzard have certain design goals for Starcraft, and they have been pretty clear and consistent about them. I believe most, if not all, players agree with these broad goals in general. For example, the players should be given options at their disposal to deal with different situations, and the unit design should help to differentiate the skill level of the players. Specific observations are identified to be problematic because of how poorly they aligned with the fundamental goals, and the solutions proposed should be based on these goals.
This discussion highlights why I think some of the proposed solutions may not solve the identified problem, and how they may not fit well with the general design goals. Thus, I won’t go through every change mentioned.
Unit utility and role
Increasing unit utility is one clear theme in the proposed changes. At the basic level, Blizzard want every unit to be utilised, and it has been a clear goal since Wings of Liberty. Many of the proposed changes are based on the observation they are not used as often as Blizzard want, because either other units do the intended roles better or the roles they have now do not fit well with the goals.
The problems identified are spot on in many cases, but I strongly believe that the “usage rate” of the units does not have to be close to “equal” as Blizzard may have implied in their explanation. Blizzard were clear that certain units were meant to fit a specific role. For example, the healing ability of Reaper added in Heart of the Swarm was meant to extend the scouting utility duration, and the reduced damage further pushed it to be a scouting unit. It is unnecessary to add a “late game ability” upgrade for Reaper to ensure it still remains an option beyond the early game. However, when you look at some of the changes proposed, it appears that this is the direction that Blizzard want to push.
Let’s use Dark Templar as an example. I will start off with saying, it is a well designed unit. Dark Templar is an expensive unit that requires opponent to make investment to counter it, and its effectiveness varies according to the players’ ability (watch Zest’s games). It is still a good late game harassment option, as it forces opponent to make specific investment and tactical choices. The ability to morph into Archon also gives it the utility beyond just being a harassment unit, so its utility should not be identified as a problem. However, based on the quote below, Blizzard suggest that they want to improve its utility in the late game but not in the mid game.
We also want to point out that we clearly don’t want to buff Dark Templars at the stage which they first enter the game, so we’re introducing an upgrade that will come into play after a Dark Templar Shrine has been completed for some time.
The proposed change does not appear to be based on a well identified problem, because I will argue that Dark Templar’s late game utility is not a problem. If the current utility state of Dark Templar needs to be improved in the late game, one can make a stronger case that Reaper should receive a similar, if not stronger, treatment. I am not discussing about the balance of these two specific units, but I just find the utility justification for Dark Templar unconvincing.
As Blizzard have stated, they are, and I quote, “highly experimental” with this change. I think it is a good thing to push the boundary a bit when you are at the exploratory stage. However, the goal to have fewer “bench” units, and then identify Dark Templar as a unit that should be used more in late game does not appear to be a good premise for “highly experimental” changes.
Other proposed changes appear to have the same issue in problem identification, for example, Viking and Raven.
Utility is not only about how often the units is used, but also when it is used. Despite Liberator has no issued of being underused, it is considered to fill too many roles. The fact that it outperforms Thor, which is designed to be the anti-air aoe unit, in anti-air, there is something wrong. Moreover, Liberator is attention-efficient in harassment, and it also provides strong area control for the main army. When you put these points together, you start to question what exactly is the role of Liberator to begin with.
Polt’s recent TvP games demonstrate how Terran players adjust their mid game plan to maximise Liberators’ effectiveness. His main convergent point is 5-1-2, and the two Starports focus on Liberator production. I know he made vikings to counter Tempest in the vod below, but the key thing I want to highlight is the tactical choice to favor Liberator over Medivac. His go-to build moves out with three Medivacs, and he just keeps making Liberator after that. Previously, in Heart of the Swarm, Polt is a player who is extremely good with multi-pronged attack in the mid game, and that heavily relies on Medivacs. However, he now changes his mid game plan to be more head-on, and only make additional Medivacs to have two dropships of units attacking elsewhere in order to find the right opportunity to set up his Liberators for his main army.
Therefore, I think the problem identification for Liberator is spot on, and the solution proposed is reasonable. This makes Liberator a unit that is specialised in high damage single targeting area control unit. Although there is still role overlap with Widow Mine, the aoe effect is strong enough of a justification to pick Widow Mine over Liberator in certain situations.
Since I bring up area control and aoe, I cannot not mention about Siege Tank. Although Tank, Liberator and Widow Mine are good in controlling an area in their own ways, they are all reasonably mobile. For the purpose of area control, it seems intuitive that the lack of mobility is a reasonable and fair weakness, but that is not the case even for an iconic unit like Siege Tank. I questioned the introduction of Tankivac when Legacy of the Void was released, and I said “Tank is just too mobile and its role of providing positional advantage is deeply affected.” Tankivac has been a huge topic for the first six months of Legacy of the Void, and Blizzard had considered removing Tankivac previously. Given that Blizzard continue to advocate that the mobility of Tankivac is worth discussing, it is clear that they have high confidence in their problem identification. In fact, they put forward the same solution, which is to remove the Tankivac pick up and increase the damage of Tank, so it is likely to be implemented eventually. The only way I see it doesn’t get through is that Blizzard have figured out a better solution for the problem.
Another unit that has role overlap problem is Adept, and I have also pointed this out at the beginning of Legacy of the Void that Adept is taking over Zealot. Zealot’s role is to provide the meat shield for the main army, close gap against range units, and harass in late game warp-in. Adept has replaced Zealot in every way in all match ups in the main army, and you can even argue it is the core of the Protoss army composition now.
The Psionic Transfer ability allows Adept to close gap, harass mineral with “run-by” and even scout. Adept’s potential is much higher than Zealot in all-ins and normal engagement at any stage of the game. The only thing that Zealot seems to do better is late game harassment with Warp Prism warp-in, because Zealot does not cost gas and they require less attention. I am not sure buffing Zealot alone is enough, and I think Adept should receive the same treatment as Liberator by having one of its roles weakened.
Tech path options
Like I have said earlier, Blizzard have always been opened that players should be given tech path options, and they will go a long way to make it happen. It can be something that is as direct as adding a unit to a tech tree (e.g., Oracle in Heart of the Swarm because Stargate is underused in Wings of Liberty), or something that involves a lot of trial and error with different unit interaction in the balance process like mech.
I have voiced my concern about the intention to ensure mech works, and my stand remains the same. Thus, I disagree with Blizzard on their persistence with mech at the goal definition and problem identification stage. However, I can agree with their argument that the Legacy of the Void economy system is not favorable to mech, and I think this observation is a good starting point to make adjustments. In line with the economy system argument for mech, the Siege Tank changes should allow mech to spread out more to secure more bases, because you can lock down the same area with fewer Tanks (see the difference in the image below).
Although I believe this Tank damage buff will make mech stronger and allow mech to expand easier, I disagree with the Tankivac’s effect on mech. I actually repeated several times in previous posts that I don’t buy Blizzard’s long standing argument that Tankivac is holding mech back. Blizzard elaborate on this argument further this time and suggest that Medivac is not the right choice for mech (see quote below).
Yet, constructing Medivacs (a unit which heals only biological units) is often not the right choice for Mech players, and so these players are not able to take full advantage of the Siege Tank’s strengths.
Of course, Medivac benefits bio more than mech, but the unit also has an important role in mech. The value of Medivac for mech is not just about healing, but it is also about dropping Hellbats. Hellbats provide the essential “meat shield” for the Tanks to do their job at the back, and effectively make it difficult for the bio to engage well. The vod below between Bomber and GuMiho shows that.
Another notable point is the redesign of Cyclone, which is a unit that appears to have a much smaller influence on the metagame than other new units. Interestingly, we are seeing more Cyclone openings in TvP recently, and it shows that players are still exploring the new units. If you look at INnoVation’s recent TvP build against Patience (see vod below), it is clear that the Cyclone has a major role in his build, because he went for a gas first fast tech opening and delay his Command Centre just to get a Cyclone (even though he was careless with it).
What caught my attention about the changes of Cyclone is the suggestion to use it with Hellions for mech (see quote below), and this makes me wonder how hard Blizzard want to push mech. The Hellion and Cyclone early game combo was experimented in TvZ when Legacy of the Void was just released, but it didn’t work out because of the Roach and Ravager metagame. Blizzard give me the feeling that they attempt to make mech viable in a broader sense, and not just in the Terran mirror match up. Mech is already becoming more and more common in TvT recently, and hence, Blizzard’s effort may indirectly kill bio.
This way, the Cyclone can combo well with the Hellion early-on in order to defend additional bases or to keep the opponent contained while playing a more aggressive style.
I am surprised that Blizzard go as far as making changes to Tempest in order to give mech a chance in TvP, as it is perhaps the last match up you will see mech. I am less interested in the actual change itself, but the intention behind it is extremely adventurous. The fact that they use Stalkers in the Cyclone before and after gif image tells you they are serious about it.
Zerg mid game
The Roach and Ravager combo has become a staple for Zerg’s mid game across multiple match ups, and it has been identified as a problem. There is little doubt about the observation, and I am not too sure about the proposed changes. The reasoning is sound: nerf Ravager and buff Hydralisk. However, I am thinking whether the competing unit(s) is Hydralisk, as I think Zergling, Baneling and Mutalisk composition is the “real” competitor.
The Baneling change may seem minor and appear to align well with the Roach and Ravager observation, but it may harm the viability of bio along with other changes.
A potential reason with the Roach and Ravager problem observation is the strength of Zerg’s tier three units. Roach and Ravager are tier one, and require less investment than Hydralisk and Mutalisk. This allows Zerg to tech to tier three quicker, and it is rewarding. Previously, one of the weaknesses with sticking to Roach and Ravager in mid game is the lack of anti-air, but the recent Queen anti-air buff may have further promoted this play style.
I think most of the proposed solutions are reasonable apart from those I have highlighted above. The utility and role overlap issues for several units should be further scrutinised. Last but not least, I am not a fan of going that far to make mech work, and I am worried that bio may suffer. Eventually, it will be going against the goal to ensure both bio and mech to be viable.