For Honor? Problems in the Rhetoric of Skill in Games

Ubisoft’s For Honor released on Valentine’s Day to some chuckles at the irony of a brutal if hyperbolic medieval decapitation-fest on a day popularized by elementary school kids in the US trading Star Wars and Minions cards and heart candies. As with a great many video games today that feature a large online component, critical and professional review sites delayed their formal reviews of For Honor until sometimes as much as a week after the release date in order to test the game while running on servers live with real players, but those reviewers and gamers also shared early impressions based on closed and open beta sessions that ran in early 2017.

Regardless of novice or professional, or from Ubisoft itself, For Honor is utterly absurd. In its premise, For Honor pits Apollyon, an approximation of an angel of death in the Greek and Hebrew, or essentially Ares, as a black-armored fiend driven to find the finest warriors through forced conflict. Apollyon instigates a number of cataclysmic events that drive Japanese samurai, Scandinavian Vikings, and European knights into battle.

Despite the absurd premise, player impressions and professional reviews abound that describe the exceptional and unique game mechanics in For Honor. The game draws together ludic elements from the complex sword-dueling of the Bushido Blade games or Chivalry, the battlefield hack-and-slash of the Dynasty Warriors games, intricate combos in fighting game series like Street Fighter or Mortal Kombat, and the squad-based first- and third-person-shooter tactics found in the Battlefield series and Rainbow Six games. Players find a deceptively deep and complex combat system, one that involves the samurai, Viking, or knight managing a roster of light attacks, heavy attacks, unblockable and zone attacks, charged attacks, combos, blocks, dodges, feints, parries, guard breaks, and throws. For Honor‘s ludic novelty relies on managing all these movies within three stances: left, right, and high.

Gorgeous Problems

Everyone seems to agree on one thing: For Honor is gorgeous and it’s exciting to don mythologized warriors from medieval history. There is also general agreement that Ubisoft didn’t address a series of problems in closed and open betas. Among those problems: the always online digital rights management (DRM) system, use of peer-to-peer networks (P2P), and the ambiguous matchmaking ratio (MMR).

The first references the broad dissatisfaction that a game with offline components requires a constant internet connection, something especially poignant at the contentious Xbox One announcement at E3 in 2013. Modern concerns about consistent internet availability and privacy and data collection fueled the immediate backlash and Microsoft’s quick policy reversal.

The second problem refers to the general consensus that heavily online multiplayer games like For Honor should be supported by dedicated servers for consistent online experiences. P2P server setups depend on individual user internet connections to piggy-back on one another, which works effectively enough for two people dueling but can stagger and become completely unstable for the eight players that might be involved in a For Honor match.

Third, players and reviewers have found and continue to experience frustration with an ineffective MMR, which frequently pairs players of radically different proficiency or skill.

Since launch, new problems have been identified that can mar the otherwise excellent potential in For Honor. A problem that often frustrates gamers and that reviewers identify in For Honor is microtransactions and the related phenomenon of developers releasing games only to dole out finished or upcoming additional content for more money. The in-game currency for unlocking aesthetic and gameplay gear for characters is “steel,” and For Honor offers steel packs for direct purchase for up to $100, which can be stacked. Like with many games today, players may spend several times the basic purchase price of the game. At the same time, For Honor‘s microtransactions are wholly optional and hardly the most egregious problem.

Difficulty Metrics and Skill Rhetoric in Gaming

Beyond the issues described above (and new problems, including the continued way that PvP modes don’t repopulate between games and drop all players or the AFK nonsense in multiplayer vs AI bot Dominion), the more important and more fundamental problem in For Honor and other games centers on how it determines “skill” and the way it contributes to the incessant and inaccurate rhetoric about relative gamer skill and game difficulty. Hayden Dingman’s review for PC World hits upon the problem. He writes, “Not only does [the game’s gear system] seem entirely unnecessary—the game would certainly be better if it were based on raw skill and had nothing to do with numbers.” Dignman’s review is predicated on the idea that gear that can enhance damage, for instance, muddies an experience that would be better if it relied on its purest form—player’s skill. For a collection of obvious but seemingly obscure reasons, arguments like Dingman’s that skill is an objective and measurable characteristic are inaccurate and untenable.

One element that negates an objective skill metric is getting stabbed in the back. Or knocked off a cliff by an unseen foe. Some might argue that player skill in For Honor includes the ability to see players and AI bots off screen or to manage everything on the HUD in ways that would allow you to spot backstabbers, but we’d also run into the ironic larger rhetoric about how the best gamers wouldn’t need things like the HUD. Many, many forums and reddit threads and various spots online, and no doubt representative gamers, lament the ways that players and AI “bots” use tactics that are undeniably dishonorable.

Before a broader discussion of skill in gaming, it’s worth understanding how Ubisoft and For Honor employ an in-game skill metric. In the single player and coop story mode, the player determines within conventional difficulty settings for video games—easy, normal, hard, realistic—what level of AI bot the player wants to face. A crucial difference exists in multiplayer—the game determines, through an unspecified and unclear MMR, human opponent and AI bot levels based on “player skill.” Ubisoft explains the player-versus-player (PvP) MMR in the following way:

For the skill parameter, we expand iteratively from players of your skill level from Strict to Medium and finally Extended. We expand to All Skills for any final matches before we timeout the matchmaking.

Ubisoft does not explain “skill parameter,” but gamers are broadly familiar with the concept from other MMO and online multiplayer games. A skill parameter intends to match up similarly skilled players for close-matched contests. Reading player experiences in the forums, anyone can see that the skill MMR in For Honor doesn’t currently work. We can also see from player experiences that the skill MMR in the game for cooperative multiplayer versus AI bots doesn’t currently work. While the problems facing For Honor are multitude, exacerbated by it’s wonderful potential, it’s this latter form of multiplayer—cooperative multiplayer vs. AI—that is the most interesting and fundamentally frustrating problem.

It seems quite likely that For Honor determines the MMR metric through the game’s “Prestige” ranking, yet Prestige is a poor, inaccurate proxy for skill. A player earns Prestige for any of the current roster of 12 warrior characters after completing a multiplayer event. Prestige is determined by XP, or experience points, an incredibly common function shared with countless games today. Players earn Prestige regardless of how skillfully they play, another commonality, including by just cohabitating a game with another player with “Champion Status,” which boosts the XP for all players in the team. In short, Prestige is a far more accurate representation of time spent playing For Honor than it is a representation of how skillful or effective a player is with a particular character. (It may also be the case that For Honor uses kill/death ratios or similar metrics to contribute to its MMR, but these are similarly problematic, as described by backstabbing, above.)

A misunderstanding of the “10,000 hour rule,” codified by psychologist Anders Ericsson and popularized by Malcolm Gladwell in Outliers, may contribute to the absurd equivocation of Prestige and skill. We intuitively understand that in order to become masterful (at anything), one must work hard and practice a great deal. The “10,000 hour rule” prompted Gladwell to inductively quantify and specify patterns in masterful humans to show links to their spending 10,000 hours of work and practice, especially while young. The phenomenon is hardly that clean or simple, though, and each human can offer anecdotal evidence to support the complexity of time and skill: hapless guitar skill after dozens of lessons, cooking that’s never improved despite immense effort, a body type that disallows competitive basketball play despite practicing and playing for years.

There are no doubt For Honor players who started the game with a “Prestige 0” Viking and who are insignificantly if at all improved as a “Prestige 2” Viking, a rank that represents dozens of hours playing the character. In For Honor, the character’s gear likely improves (e.g., costing less stamina to use), something Dignman laments, but For Honor‘s MMR then pits that higher-level gear against higher-skill AI bots. The issue, in other words: Top-tier clubs and exceptional shoes and slacks won’t make the casual golfer competitive with the club or PGA pro and the phenomenon is even clearer when swapping cars with an F1 driver for a race home.

Remember, too, For Honor features a current roster of 12 characters (with new characters to be added in upcoming DLC), each with a unique move set and play style. Imagine the player who is exceptional as that Viking character, has Prestige 5 and gear at 108 (while the average for players a week and a half after release was likely below Prestige 1 and gear in the 20s), and wants to try a new character, a knight with a markedly different move set and play style. The player would much rather learn while playing, and plays casually anyway, so enters a multiplayer mode for 4-player versus 4-AI bots, like Dominion, Brawl, Elimination, or Skirmish. Based on the context described above, and detailed by player experiences in the forums, that Viking player will face the efficient and dirty Level 2 AI bots, and will be slaughtered.

For Honor only notes AI bot level in quick “killed by” screens, but players can determine AI bots at higher levels (i.e., 2 and 3) by how they behave. Higher-level AI bots are much more adept at blocking, for instance, they deploy attacks while changing stances, dodge more, activate “revenge” mode far more often, use unblockable attacks more, throw more and with far greater chance of throwing the player into a hazard, and they execute players and interrupt player executions of other AI bots.

An example of a situation with a higher-level bot might be instructive. The Conqueror is a “turtle” or “tank” knight character, a bulky shield-and-flail toting character with strong defense and powerful attacks. At Level 2 and 3, Conqueror AI bots block almost everything, including guard breaks and throws, and frequently knock the player down or into hazards. The tactic I generally use when facing a Level 2 or 3 Conqueror, regardless of my Prestige 10 Kensei samurai, is to run and to do something that I find absurd—I try to find a way to draw an ally in to stab the Conqueror in the back or to backstab the Conqueror myself. In For Honor, then, the player will get stabbed in the back and the game will compel the player to stab other players and AI bots in the back. Or, as Arif writes, there is no honor in For Honor.

Git Gud

Voice these experiences and concerns, as many have in the forums and online, and the ludicrous response trope follows the general difficulty rhetoric: “git gud.” Dr Josh Call studies difficulty in video games and promotes a more nuanced understanding of difficulty than what’s embodied in “git gud.” Based on what already currently exists in the forums and online, a response to the Conqueror example posed above would likely include snark about dodging and parrying the Conqueror’s attacks. Parrying, however, is arguably one of the hardest things to do in For Honor, as it is essentially a layered system of QTEs (quick time events). To parry on default console gamepads, the player must hold the left trigger to lock on to the opponent Conqueror, time the right stick to match one of the three directions the Conqueror might use to attack (the first QTE), and if the direction is matched, time a right trigger press in sync with the moment of attack (the second QTE).

I can parry, just not with consistency or efficacy. Many in the forums comment that these higher-level bots in For Honor are better than most human players, and “better” is a remark on consistency and efficacy. As Call would likely argue, the parrying and any other tactic in the game is seen as “difficult” relative not just to a player’s pattern-recognition, understanding of a character’s move set, and time spent in practicing, but also to player dexterity, muscle memory, innate skill, and probably to player age and other more variable factors. The jumping, spinning snipers infamous in Call of Duty and other FPS games for awesome killstreaks without aimbots or the “hardcore” gamers that best Dark Souls and Nioh and similar games might spew “git gud” about skill and difficulty, but all evidence and our shared experience argues that such aptitude may have nothing to do with simply practicing to get better.

Beyond “git gud,” some of the responses to supremely tough AI bots include ideas that these bots are practice for playing against the more idiosyncratic (and sometimes equally proficient) human opponents in PvP multiplayer. Importantly, from the general hostility in online PvP play to the interest in just having fun to even that practice for PvP, there a host of reasons players may want to play against AI and not other humans in For Honor and other games. Unfortunately, Ubisoft and For Honor privilege PvP over AI play. For instance, one of the primary ways to earn XP and “steel” currency (for gear) is through Contract Orders and Daily Orders; in the former, players are offered two AI orders for every four PvP orders, while in the latter players can complete one of each (until mid-April’s 1.05 patch). Disincentivized to play against AI by a lack of control over the difficulty of the AI bots you face, derision of your skill, and diminished XP and steel, Ubisoft and For Honor then provoke you with AI bots who stab you in the back, decapitate you, and then taunt you—taunt you with taunts that you cannot use or equip.

Inazo Nitobe writes in the highly influential Bushido about the fundamental principles of the samurai and the chivalrous code of ethics for samurai and European knights. Among these principles, Nitobe argued benevolence, honor, duty, fealty, and more contributed to the Japanese concept of youyu, or a thoughtful mind and stance about conflict. He relays the story of Kenshin and Shingen. Despite being at war with Shingen, upon hearing of Hojo’s refusal to ship rice to Shingen’s province, Kenshin is lauded for shipping Shingen rice and writing, “I do not fight with salt, but with the sword.” Nitobe smartly connects the concept in bushido to Nietzsche for Western readers; Nietzsche wrote, “You are to be proud of your enemy; then the success of your enemy is your success also.” Ubisoft might learn something about the source material and take the game’s namesake more seriously. It can’t make humans play with honor (even though some do anyway, and despite attempts like the XBL’s reputation system), but an easy way to increase the honor in the game would be to allow players to be more like Kenshin, to determine the AI bot level they want to play. Making that update to the game would trump an inane argument linking time played to proficiency and would also be the easiest fix among the host of problems in For Honor, an interesting game that is absolutely brilliant when played honorably.

What do you think? Leave a comment.