Sitting down at my computer, overwhelmed by the massive data set I somehow was able to collect, I had to think about what I should look at first. I have a lot of things I want to look at, but I the answer was obvious: Peacock.
The Skullgirls Bogeyman
Going into this I knew Peacock was by far the most hated character in Skullgirls, and I fully expected this to be reflected in my survey questions about player frustration. She is a character that is complained about frequently, and requests for nerfs are constant. I believe this can be blamed on three broad aspects: First, she is considered the best character in the game by many (if not most) high-level players. Second, she is a "zoner", which means her gameplay is based around keeping the opponent out and safely chipping away at them; zoners are often frustrating to play against across fighting games as a genre. Finally, aside from having very strong zoning capabilities she also has a powerful offensive game, making her generally well rounded. To be transparent, I tend to agree with these sentiments, particularly the last one. Many people enjoy Peacock’s gameplay and feel zoners need to be present in the game. Having a diverse set of strategies to pick from creates diverse gameplay and gives players room to express themselves through character choice. My goal is to figure out whether Peacock is as problematic as she seems, or if anecdotes are painting an inaccurate portrait.
Peacock's zoning pattern
Character Frustration
Something you should understand about the Skullgirls community is that players complain a lot. Anecdotally, I have heard complaints about every single character in the game. The game is very fast, demanding, and punishing. The high stress situations the game creates can cause players to become frustrated. Because of the baseline frustration the game has for some people, and because complaints are loud, it’s hard to parse whether a problem actually exists.
One way to show evidence that Peacock is problematic is if frustration in Peacock is substantially different from the players’ overall frustration with the other characters. For each character, survey participants asked how much they agree with the statement “(X Character) is frustrating", with strongly disagree, disagree, neither agree nor disagree, agree, and strongly agree as the three choices. To create a baseline, I averaged each person’s frustration across all 15 characters, a plot of which you can see below. The x-axis is the level of frustration and the y-axis is the number of survey respondents.
As you can see the average is focused around neither agree nor disagree, at a mean of 3.23. To me this suggests that players are generally not too frustrated with the game as a whole, contrary to what the discussion boards may lead you to believe. Contrast this with the next plot we’re going to look at, frustration with Peacock.
This is Peacock’s plot, with a mean of 4.20. Yes, over half of the 300 participants responded with Strongly Agree. Of all the characters, Peacock rates highest, which is what I thought was going to be the case. This is obviously very different from the sum plot, but in a vacuum it’s hard to say this is totally out of the ordinary. Let’s look at the next most frustrating character, Fukua.
Fukua has the second highest frustration rating at a mean of 3.53, which is 0.66 less than Peacock. She’s still more frustrating than not, but it’s certainly less severe. Interestingly, there is some disagreement about Fukua’s placement in the tier list. Some consider her fairly high and some rate her as the worst character in the game.
Is Peacock frustrating because she’s a zoner? She isn’t the only dedicated Zoner in Skullgirls, there’s another named Robo Fortune that has the same general zoning game plan with a different slant. Next we’ll look at frustration with Robo Fortune to see if we can draw some comparisons.
Here we can see a very even looking distribution compared to Peacock with a mean of 3.06. The difference between this plot and Peacock's could be explained by the way the two characters differ. Peacock uses objects as projectiles, which take time to move across the screen. Robo in contrast primarily uses laser beams to zone, which instantly (or near instantly) travel the length of the screen linearly. A slow enough projectile allows you to approach the opponent because even if the opponent hits you, they’ll still get hit by the projectile. This is a brutal feeling, as in most other other cases when you land a hit on the opponent you are advantaged or at the very least even. Peacock can run forward under protection of her projectiles, but she can also quickly teleport in front or behind the opponent allowing her to go straight from zoning to offensive play. This can even lead to full screen mixups that are difficult to react to. Peacock has a move that drops a projectile from the sky onto the opponent, and different versions of the move can track the opponent depending on where they are on the screen. This allows her to create safe pressure by making a risky attack, letting the item drop and cover her, then attacking with something else and/or forcing the opponent into an unreactable 50/50.
Robo Fortune zoning pattern
Robo Fortune does have a set of projectiles like Peacock’s, however they have more restrictions. They all require her to stockpile “heads”, which are a resource that she needs to earn by using another specific move. None of the attacks come from behind her and her only forward moving projectile moves linearly and is too fast for her to outrun, so they can’t interrupt combos in quite the same way as Peacock’s. She has a mine that she can set on the ground that won't disappear if she is hit while it is being deployed, but if it’s been planted, hitting Robo will destroy the mine; it can still trade, but it won’t interrupt a combo. Finally she has a projectile that launches a bunch of missiles into the air that then come down on the opponent, which she needs to spend one, two, or three heads to use. Spending one head causes missiles to fall where the opponent’s position was when the attack started, two heads launch missiles that target where the opponent is when the missiles begin falling from the sky, and three heads will launch missiles that will track the opponent fully. These can’t be interrupted by hitting Robo Fortune, however the first two strengths can be dodged and only the third one is guaranteed to break combos. Spending three heads is not insignificant, as heads have a variety of important applications, so it has less prominence than the combo breaking projectiles peacock can use freely. Peacock and Robo Fortune have significant differences, however they both squarely fall into the zoner archetype, suggesting that frustration with peacock isn’t due to her broad archetype alone.
Player Experience and Frustration
Peacock players should know their character better than anyone else, so it’s prudent to check how they feel about fighting against their own character. I compared Peacock frustration based on whether the player has experience using peacock. 46 participants list Peacock as one of the three characters they have the most experience with, while the remaining 254 do not. I ran a Welsh two-sample t-test and generated a plot from it which you can see below. A t-test can indicate if there are differences between two groups. In this case, the groups are players who have a lot of experience with Peacock and players who don’t.
The 254 participants that did not report experience with peacock showed a significantly* higher rating of frustration (M = 4.28) than the 46 that do have experience with peacock (M = 3.76). This result was significant, t(57.01) = 2.70, p = .009. It's not surprising that familiarity with a character would reduce players’ frustration with that character (this will likely be a future topic). This difference could make nerfs a tough pill to swallow for Peacock players.
Oftentimes fighting game characters that are frustrating for new players become less frustrating as they gain experience. Let's look at how frustration changes based on how often someone plays.
We can plainly see there isn’t much of a difference between frequency of play. I was actually expecting there to be something more here, but this suggests that experience doesn’t make Peacock less frustrating. “You just need to get better at the game, you just have to learn how to counter it” is often something experienced players tell inexperienced players that are frustrated with some gameplay element. It’s actually very good advice, but that doesn’t seem to be a valid response here.
Qualitative Measures
There are some trends I found while looking through the character feedback for Peacock, and many of them mirrored the opinions I've commonly seen discussed online. A lot of frustration was very broad, both her zoning and her aggressive play were mentioned frequently. There’s a common opinion that the character lacks weaknesses, and that she gets a high amount of reward for a very low risk play style. Specific examples include her teleport mix ups, safe offense, and high meter gain. Her defensive capabilities were mentioned, with her crouching light kick anti air and her combo breaking projectiles as stand outs. A number of people complained that her synergy with assists is too high, that they cover her zoning blind spots too well. There were a few matchup complaints, specifically for Parasoul and big body characters.
There seems to be a general frustration with nerfs not doing enough to curb her gameplay, and sense that peacock players have been able to easily get any meaningful nerfs reverted. For instance a few people have complained that the recent damage nerf to item drop, an attack that gives Peacock a large part of her offensive capability, was rendered meaningless once other combo routes were found. She is often jokingly referred to as evil, a war criminal, etc. The developers will be given access to these responses in full so they can be better informed than I can do here.
Peacock on trial for war crimes (again)
Closing thoughts
There’s tons of analysis that could be made between Peacock frustration and many of the variables I have collected, but I’m trying to keep things reigned in. That being said, from this initial look I think Peacock is a problematic character. There just isn’t any other character that is comparable to the amount of frustration she’s creating for players. Even Peacock players find Peacock frustrating. She isn’t a character that gets less frustrating with experience either which is worrying, and she seems to be frustrating to players in a consistent way. I don’t think you want a single character to be this much more frustrating than the rest of the cast; at some point you go beyond empowering a player with a strong game plan that feels good to use and you start negatively impacting everyone else’s experience. We set out to find whether the complaints were minority opinions, and the data shows they are not. I look at this character and I see a problem that needs to be remedied.
As for specific solutions, you could really take it a lot of ways. I’m not in on the design goals of peacock or Skullgirls in general beyond vagaries and guesswork. My understanding is that the game is built to be oppressive, momentum based (a good decision can turn the tide), and that every character is meant to feel overpowered in some way. My understanding of what makes Peacock unique in the cast is that she zones with projectiles that deal high amounts of chip damage, which forces the opponent to make risky plays or else die by 1000 cuts. I’m going to list out some changes that I think could make Peacock less frustrating while preserving her core identity. These are all nerfs, and it should go without saying that they would need to be thoroughly tested before committing to any of them. Fighting games are deeply complex systems, and even small changes can have unintended consequences.
Peacock’s projectiles disappear harmlessly if she is hit with an attack that is not a projectile. This would keep her ability to have projectile wars with other zoning opponents, but if you are able to weave through her projectiles and hit her you are guaranteed the advantage you feel you've "earned". It would also make it riskier for her to go from her zoning pattern to her aggressive play, because the projectiles couldn’t be relied on for safety in as many situations. A less severe version of this nerf would also add assists to the class of attacks that don’t cause projectiles to break, as certain armored assists are very good at bypassing projectiles altogether.
Peacock’s projectiles disappear harmlessly when she teleports. This would weaken her ability to escape as easily when backed into the corner, as she wouldn't be able to make her teleports safe as easily. It would also make using teleports to transition from zoning to offensive play more difficult, as she would have to rely on assists to cover her.
Nerf her normal damage. Normals are attacks that make up the bulk of a combo, and in peacock’s case many of these are relegated to close range combat. Decreasing base damage across the board for attacks that are not used in zoning would significantly decrease the reward she gets from her offense. This would keep the way that she plays offense intact, but it would incentivize her to zone instead.
Nerf item drop (again). Peacock’s item drop is brought up a lot, even though it’s been the target of scaling nerfs already. These nerfs have been said to not significantly impact her damage output. The scaling could be increased again but I’m inclined to think this would be treating the symptom rather than the disease.
Do you have a better solution? Do you think there’s something I missed? Are there any characters that seem like they aren’t frustrating enough? What should my next article be about?
Thanks again to the SG community for supplying such wonderful data. I hope this will be helpful for balance discussion, or interesting at the very least.
Comments