GCW Zero - An Open Source Gaming Console Built by Gamers ...

TL;DR I thought Kerrigan was a poorly designed hero for her role, decided to do a full analysis on her to prove it, then offered a potential rework. Decided that I should share it with others. This post is long.
ReCrafting the Queen
Class 12 psionic waveform detected. The Queen of Blades has returned.
Sarah Kerrigan is one of the most iconic heroes of the StarCraft franchise, reprising her role as the lost love, powerful ally, and primary antagonist throughout the course of the series. As such, she carries a number of identities with her: Raynor’s beau, betrayed ghost, queen of the zerg, powerful psi user, and the visceral Queen of Blades. Given the existence of the other zerg heroes within the game as well as her current kit, I’ll focus on the ‘Queen of Blades’ aspect of her identity. The Zerg modifications to her body have evolved her into the ultimate killing machine: super-human strength and regeneration, razor-sharp claws, skeletal scythe-like wings, an armored carapace, and heels— Abathur really covered everything. Thus, the ‘hero fantasy’ or ‘end-game fantasy’ of Kerrigan is that of a predator: leaping from foe to foe, eviscerating her prey before leaping to the next, leaving a trail of blood and death behind her. Given these traits, the core identity of Kerrigan fits wonderfully into the general archetype of melee assassin.
Melee assassins in MOBAs are by design some of the most interesting characters available— and some of the most difficult to balance. Their role in team compositions is to threaten the traditional backline (healers, ranged auto-attack carries, and mages) as well as cut through tanks when improperly positioned. This is, however, where their problems stem from.

• As their role in a team composition is to threaten the backline, they have to have some sort of engagement skill to gap-close onto the ranged carries/supports.
• After getting onto the backline, the melee must generate more threat (often damage) than the ranged character, else there’d be no reason to play him/her.
• Due to the plethora of cc and disengage available in the game (as well as simple attack animations), the melee needs some way to remain on their target after the engagement. This can be in the form of in-fight mobility (e.g. Windfury, Sweeping Strike) or crowd control (e.g. Feral Spirits, Frozen Tempest)
• As it is trivially easy to turn focus-fire down a low-hp melee carry, they need to have some sort of damage-prevention skill to ensure their survival— or in the case of their death, that the time and focus taken to eliminate them provided their team with an advantage.

These issues are balanced on a razor’s edge, however. If any element of this set is too strong, it destroys the viability of many backline heroes, but if any element is too weak, the heroes end up with win-rates that scale inversely to skill or encroach into other design spaces. This leads us to the core issue of Kerrigan’s design: she is a tank in a melee assassin’s body. To justify this claim, I’m going to break down her current kit and give my understanding of each element before offering a solution.
Current Kit:
Trait— Assimilation: serves as her defensive ability by taking a percentage of damage dealt and returning it to her in the form of a stackable shield. This serves as a reasonably balanced defensive ability, and fits very thematically into both her core identity and her end-game fantasy. However, while thematic, any ability with the vampirism trait has to be treated with the utmost care, as it is prone to degenerate situations. As Trikslyr revealed during a Developer Insights interview, early iterations of Kerrigan had an over-tuned trait, allowing her to 1v5 enemy teams (given memories of the Wings of Liberty finale, still arguably thematic).
Ravage: This ability was taken almost directly from Starcraft 2, and serves as her gap-closer. Not only is it incredibly thematic, I believe this ability sits at the core of her identity, and allows players to really feel like “Kerrigan”.
Primal Grasp + Impaling Blades: Impaling Blades is a trivially-easy to dodge AoE stun, and Primal Grasp is a weak cone-pull; each ability on its own is incredibly lackluster, but together they form The Combo. After placing W, it’s possible to use E directly afterwards to drag the target(s) into the stun, resulting in a CC-combo that is nearly impossible to dodge. This combo is what she was prized for in professional play, and in my opinion, is what is preventing her from ever feeling like Kerrigan. Because both the stun and the pull is an AoE, this allows Kerrigan to consistently get 1-2 targets in her stun, with a best-case scenario of 4-5: this is equal to or better than a Johanna heroic ability on a 12-second cooldown. Because of the immense power of this internal combo, the rest of her kit’s power-level has to be lowered to accommodate, resulting in a melee assassin that, almost by necessity, loses to any other melee assassin. Furthermore, because this combo requires a minimum of two abilities to pull off, it leaves her playstyle 1 dimensional and team-oriented— a far cry from the feral predator of the Starcraft cinematics.
Maelstrom: This heroic ability feels thematic, channeling her Brood War psi-storm past, while also fitting well into her melee-brawler identity and her trait.
Summon Ultralisk: This is the secondary heroic ability for Kerrigan, and while it doesn’t fit her melee assassin identity particularly well, it does match her identity as leader of the zerg, and provides her kit with additional utility by allowing her to apply single target pressure. The problems plaguing Kerrigan’s Ultralisk, however, are nearly identical to the problems the Starcraft 2 ultralisk had: it’s big, slow, gets body blocked or stuck easily, has relatively low damage, and isn’t tanky enough to fulfill its role. The recent changes to Ultralisk help alleviate these problems, but I think more could still be done to make the ability satisfying to use.
Talents: As a Kerrigan player in the current meta, your job is to engage onto the enemy backline and catch high-priority targets in your combo, allowing your team to follow up. You use Q to initiate, W+E to combo, and the R to provide you with more shielding. Because this combo uses her full rotation and her cooldowns are oppressively long (due to their high team impact), she has no means of applying significant pressure to targets after her engage. This frequently leaves her maneuvering defensively, trying to live long enough to get a second combo off. This sort of ‘live long enough to cc again’ strategy is the precise playstyle of melee tanks, like Diablo or Muradin. Unfortunately for Kerrigan, her base kit doesn’t have the health pool to pull this off, so she puts every talent into abilities that either increase her health pool or get her combo out faster. According to HotsLogs^1, here are the most frequently picked talents as of 3/29/2016: ^1. (All leagues on the first line, filtered for Master-only in italics/quotes on second)
Level 1: Siphoning Impact 75.1%

Siphoning Impact 93.0%

Level 4: Clean Kill 51.8%

Clean Kill 93.0%

Level 7: Battle Momentum 45.1% / Assimilation Mastery 39.5%

Battle Momentum 75.0%

Level 10: Maelstrom 86.1%

Maelstrom 90.4%

Level 13: Lingering Essence 42.7%

Eviscerate 58.0%

Level 16: Aggressive Defense 68.5%

Essence for Essence 47.0% / Aggressive Defense 42.8%

Level 20: Omegastorm 46.8%

Bolt of the Storm 57.4% / Omegastorm 30.5%

As you can see here, every popular talent choice forgoes damage to increase her survivability, with higher-level players preferring talents that help her execute her combo (battle momentum, eviscerate, bolt), reinforcing her pseudo-tank playstyle. This trend highlights Kerrigan’s other major issue: her talents are bad. Due to the incredible strength of her base-kit, Kerrigan’s talents lack the transformative properties we’ve generally come to expect from certain talent tiers. In addition to this, she has few damage talents, particularly when compared to other melee assassins, and those she does have are weaker than her counterparts. This is most apparent when comparing her to Azeroth’s most beloved baker, Illidan Stormrage. Illidan’s level 4 talent, Friend or Foe, has almost identical strength and functionality to Kerrigan’s level 7 (Adaptation) and level 13 (Eviscerate) talents combined. Furthermore, if you compare that to the level 4 talents offered to Kerrigan (Fury of the Swarm, Clean Kill, Psionic Pulse, Envenom), it’s clear that Kerrigan’s talents are both less interesting and less impactful. While this makes sense given her base kit, it has serious implications for her role in competitive play— or lack thereof.
Dethroning the Queen
In Technical Alpha, Kerrigan was one of the most contested picks of the competitive scene, seeing play in approximately 60% of games over the course of the Alpha and achieving a win-rate as high as 90% (Technical Alpha patch 6, GosuGamers). In the last month, however, she has been picked a mere 5 times (2% of games) and has achieved a 0% win-rate. I believe this dramatic fall in popularity and viability is due in large part to her identity as a tempo-based hero and Blizzard’s (rightful) attempts to lower the impact of early leads.
As mentioned earlier, Kerrigan’s base kit includes a gap-closer and a high-impact cc, making her a top-tier pick in rotation-based pick comps. Furthermore, her combo does not rely on transformative talents (owl-build Tyrande, Boomkin Malfurion) to activate, allowing her to be at full-strength at level 1, unlike the vast majority of heroes. This gives Kerrigan teams a strong early-game advantage and allows her to act as a pincer, forcing opponents to either take bad engagements or sacrifice objectives, which in turn leads to a greater advantage. As time goes on, however, Kerrigan’s weaker talent options and the enemy’s access to strong defensive abilities (cleanse, Divine Shield, etc.) mitigate a great deal of the threat she provides. This creates a terrible game-play binary for Kerrigan: either you succeed in keeping up tempo, snowballing the game into a non-interactive victory, or you fail to keep up the tempo, becoming a liability for your team as the game progresses. During the Alpha phase of the game, it was relatively easy for Kerrigan to create and maintain pressure, but with the heavy nerfs to envenom, the removal of rewind, and the strengthening of comeback mechanics, it became difficult to create the runaway victories she was known for. The most impactful change to Kerrigan’s win-rate, however, came with the scaling-changes introduced in late 2014. The changes from additive to %-based scaling on health and abilities minimized the impact of level differences, and resulted in her near-disappearance from pro-play. The statistics from before and after the scaling changes speak for themselves:

• The Rexxar Patch she was picked 29 times (19%) and had a 66% win rate
• The Lt. Morales patch she was picked 30 times (13%) and had a 57% win rate
• The Cho’Gall patch she was picked 23 times (7%) and had a 43% win rate
• The Lunara patch, she was picked 3 times (2%) and had a 0% win rate.
• In the last month, she has been picked 5 times (2%) and has had a 0% win rate.

The Rework
Trait— Assimilation: Gain 8% of damage dealt from Basic Attacks and Abilities as Shields for 6 seconds. Shield amount gained doubled against Heroes.

• Explanation: As was mentioned earlier, the current Kerrigan trait is incredibly thematic, stealing essence from her enemies to fuel her rampage. Due to the changes to her basic abilities and their new average case, however, I think a slight nerf is warranted to prevent degenerate situations from occurring.

Q— Ravage: Leap to a target, dealing 267 damage. If the target is marked, the mark is consumed and this cooldown is instantly reset.

• Cooldown: 8 seconds, Mana: 60
• Explanation: This ability is taken from her current kit, and forms the core of the revamped Kerrigan’s playstyle. To ameliorate the loss of a 3rd damaging skill, I increased its damage slightly, but also increased its mana cost by 10 to prevent low-cost reset harassment during early laning-phase.

W— Psionic Pulse: Reveals nearby enemy heroes, marking them for 5 seconds.

• Cooldown: 12 seconds, Mana: 30
• Explanation: This ability is the limiter of the revamped Kerrigan, and brings some of her psionic identity into her kit. I felt that resets were a core part of Kerrigan’s identity, but the reset-on-kill mechanic on Ravage felt too inconsistent to be effective in PvP as well as uncharacteristically PvE. Furthermore, I wanted to keep the ability to get multiple stuns off on Kerrigan while greatly reducing her current best-case scenario. Psionic pulse allows Kerrigan to have the fantasy of leaping from target to target across the battlefield while dampening the single target assassination potential that has defined many of the newer heroes.

E— Impale: Kerrigan impales her target with her wings, dealing 180 damage and applying a bleed that deals an additional 80 damage over 4 seconds. If the target is marked, the mark is consumed, the target is stunned for .75 seconds, and the cooldown is reset.

• Cooldown: 10 seconds, Mana: 70
• Explanation: I had originally considered making this an AoE hamstring, but after reviewing the core fantasy of Kerrigan, I found her to be more of a 1v1 duellist flitting from target to target than the AoE brawler that she seems to have initially been designed to be (or Xul has succeeded in becoming). Aside from removing the AoE stun, I’ve also reduced the duration by .25 seconds, as I feel she should be providing more disruption than outright CC. This should prevent her from becoming too oppressive and allow her damage threat to prosper.

Heroic Ability 1— Maelstrom: Deals 85 damage per second to nearby enemies. Lasts for 7 seconds.

• Cooldown: 100 seconds, Mana: 100
• Explanation: Borrowed straight from her current kit, I think this ability is fantastic and needs no changes whatsoever.

Heroic Ability 2— Summon Ultralisk: Summon an Ultralisk that casts burrowed charge at the target and attacks for 104 damage. Can reactivate the Ability to retarget the Ultralisk. Lasts for 20 seconds.

• Cooldown: 100 seconds, Mana: 100
• Explanation: After spending a great deal of time contemplating this heroic ability, I can understand why the Ultralisk was chosen. I had originally planned to make this talent a Psi-Emitter, periodically spawning various zerg units to attack the marked target, but realized this would create incredible problems with intended targeting amongst lower-level players and a great deal of visual clutter. I also considered the Leviathan, Broodlord, and Guardian as replacements, all of which suffer from the same shortcomings while also creating a sieging unit, splitting her identity. Unfortunately, the Ultralisk in its current state remains lackluster, rarely connecting with its intended target. If, however, the Ultralisk was able to cast burrowed charge periodically (or, if that proves too oppressive, at least once after being summoned), it would allow the Ultralisk to connect with its target and force teams to play reactively to it, rather than killing it incidentally. To compensate for the increase in the average case efficacy of the Ultralisk, I removed the splash damage component, reducing its best-case scenario.

As stated earlier, the goal with this rework is to do away with Kerrigan’s current identities of tanky initiator or AoE melee brawler and re-imagine her as a hunter and duellist. The single-target nature of her abilities means she focuses on one target at a time, while the limitations placed on resets and forcing a decision between resetting her leap and her stun should create complex and interesting play patterns, a far cry from her current binary kit. Below are some examples of new gameplay patterns she can pursue:

• Single-target catch: Q > W > E. This allows Kerrigan to catch a single target, sacrificing future mobility for a stun and a reset on E.
• Duel: E > W > Q > Q. This expends all of her cooldowns in outputting the most damage she can deal to a single target, but sacrifices the stun and puts all 3 abilities on cooldown, leaving her vulnerable to other threats.
• The Flank: W > E > (E) > Q > (E) > Q: In her current gameplay, Kerrigan’s combo is Q > W > E > R, regardless of her position in relation to the target. With this rework, Kerrigan will be massively rewarded for being able to get into melee of her targets without having to use her Q, allowing her to get stuns off on multiple targets. Conversely, this adds another level of counterplay to Kerrigan, as marked targets out of melee cannot be jump->stunned, as the jump would consume the mark.
• The Huntress: W > Q > E > Q > Q > Q: This is the scenario that drove the rework for me. The greatest feeling I’ve ever had as Kerrigan was leaping to a low-hp minion, then another, then a third before comboing a Valla who was too shocked to react. I want to create this sort of in-fight mobility and spread-pressure as the hallmark of Kerrigan, leaping from enemy to enemy, threatening a commitment to any target she has managed to mark. As with The Flank, the pre-requisite to this fantasy is the ability to mark many targets with W without using her gap closer, adding another level of strategy and counter-play to Kerrigan.

With these and other forms of emergent gameplay, I believe Kerrigan will see a Renaissance in professional play as well. Given that she has to choose between a gap closer and her disruption, she won’t simply be another Illidan or Greymane diving headfirst into (or over) the enemy. Instead, I imagine her fulfilling a role similar to that of a striker in soccer— a flex pick that’s able to disrupt her hyper-aggressive melee assassin peers, then immediately switch to a non-cc backline threat. This will serve to provide some counterplay to the overwhelming trend of assassination-based comps, reduce the amount of cc backline heroes have to worry about, and allow some immobile damage threats to come back into the meta. While every change to a hero must be done with the utmost care, I believe these reworks to Kerrigan’s basic abilities will bring a more interesting and positive play experience to the Nexus overall.