I generally don’t watch others play games, and I don’t really even play many video games anymore. But for some inexplicable reason, on a procrastination-heavy Thursday afternoon, YouTube recommended me a 6-hour speedrun of a game that I have never played before. Perhaps even more confusingly, I clicked on the video and thoroughly enjoyed all 388 minutes of it.

I had the pleasure of stumbling upon a 100% Twilight Princess run by Zelda speedrunner bewildebeast, who holds the world record for this and many other Zelda speedruns (although this video did not depict the world-record run). This video, filmed after his run, featured extensive commentary by bewildebeast explaining his route. Having played two games from the Zelda Franchise, Skyward Sword and Breath of the Wild, but not the critically acclaimed Twilight Princess, my initial, split-second decision to click on this speedrun was motivated by a general desire to learn further about the game’s story, progression, and atmosphere. In the first few minutes, when bewildebeast mentioned while skipping the opening cutscene that Twilight Princess is one of the few Zelda Games to allow you to do so, I learned that this speedrun would provide little of that. But what bewildebeast’s commentary offered—a portrait of how speedrunners “optimize” game mechanics in ways, both unintended and intended, by the original developers—is something infinitely more interesting.

First, I would like to establish that I do not consider myself an avid watcher of speedruns, (a fact that further adds to the mystery of YouTube’s recommendation). However, prior to watching bewildebeast’s commentary, I was not without a removed appreciation for speedrunning, primarily motivated by video essayist and game critic Hbomberguy, who did a video on the topic (I also highly recommend watching his video). After all, one cannot help but appreciate the precise skills involved in executing a run, in addition to the creative community efforts that go into shaving off time.

Despite having respect for speedrunning, I never especially enjoyed watching speedruns of games I have played. Take another game in the Zelda franchise, Breath of the Wild, for example. Breath of the Wild is a particularly popular and fast speedrun because after players exit the tutorial area, they can go straight to the boss fight, woefully unequipped. And so, when I watched a Breath of the Wild speedrun for the first time, I was excited to see how players would skillfully navigate the puzzles and traverse the vast landscape to reach the boss fight whereupon they would unleash a perfect set of sword combos. Instead, the top speedrunners completely skipped many of the puzzles and much of the time-consuming navigation by clipping through walls, exploiting glitches, and launching themselves across the domain of Hyrule. After viewing these 20-minute speedruns of a game I spent a hundred hours on, I was left stupefied; they seemed to be playing a completely different game than I was.

Having no previous experience watching or playing Twilight Princess, I entered bewildebeast’s video with no expectation of how the game is intended to be played. Sure, seeing bewildebeast maneuver Link solely in forward rolls hinted at the peculiarity of the gameplay, but I had no idea of the larger progression of the game. Only when bewildebeast mentioned how his run deviated from “casual” play did I realize how unconventional his progression was. For example, the Master Sword, which is the most powerful sword in Twilight Princess, enables players to defeat enemies much faster compared to early-game weapons, to portal around the map, and to transform between wolf Link and person Link, providing further ability and flexibility. These are all valuable qualities for speedrunners. However, the Master Sword is designed to be available to players only after they defeat three out of nine dungeons. In just the first half hour of the playthrough, bewildebeast is able to access the area in which the Master Sword resides—having not completed any dungeons—by positioning an enemy close to a ledge, launching a charge attack, and jumping onto an otherwise totally inaccessible area that just barely enables access to the Master Sword area. After obtaining the Master Sword, bewildebeast is now able to complete early areas of the game that never accounted for these expansive powers.

If there were no commentary, I would have very few cues from the raw game footage to notice these major sequence breaks. Apart from the funky textures of the out-of-bounds ledge, the acquisition of the skyward sword was seamless. Bewildebeast simply worked within the coded physics of combat and movement to traverse a hard-to-reach but technically accessible area of the map. The developers never intended this sequence, but bewildebeast, with this very act, demonstrated how speedrunners could operate in a liminal space between this original intent and the coded infrastructure, fashioning a novel way of approaching gaming.

Bewildebeast’s commentary focused on explaining these major sequence breaks, such as the Master Sword glitch, and more minor, but constant, optimizations, which aggregated over the course of the speedrun. One such trick that is frequently utilized is called a Long Jump Attack (LJA) in which the player throws Link’s boomerang over a void, pulls out his sword once the boomerang is over the void, and executes a jump attack, inexplicably allowing the player to jump farther than a normal jump. This trick is employed frequently to reach various ledges, but also to avoid character triggers that would force dialogue, a decision that saves mere seconds during each encounter. The LJA is not a sequencing-breaking glitch in the same sense that the early Master Sword is, but they are both side effects of coalescing combat mechanics and game physics that operate within these marginal spaces.

I never expected to become so hypnotized watching a speedrun of a game I have never played, which was enjoyable in large part thanks to bewildebeast’s stellar commentary that manages to be accessible to those who are unfamiliar with the game or speedrunning, while also clarifying the more intricate aspects of his run. Bewildebeast recorded his commentary not during but after conducting his run. One may expect his commentary to function more as retrospective, critiquing elements to be improved next time; but instead, the video acts more as a walk-through, enhancing the amount of insightful, calm intelligence available to the audience during the six-hour video. Apart from the content of the comments, bewildebeast’s dialogue assumes an ASMR-like quality, his clear voice taking an understated tone, producing a soothing effect on the audience, which is another important quality for a six-hour endeavor. And yet the video is never boring, helped by the nuggets of dry humor bewildebeast sprinkles throughout.

Even if the Zelda franchise or speedrunning are completely foreign concepts to you, bewildebeast’s 100% speedrun of Twilight Princess offers an immaculate primer to an unconventional approach to media in which masters of the material go far beyond what was ever intended by the creators, deriving their own objectives and journey. One need not experience all six hours of bewildebeast’s Twilight Princess gameplay to fully appreciate the liminality and constant optimization of speedrunning; however, I implore other completionists like myself to slowly chip away at the video, treating watching the run as its own kind of complete journey, which fortunately does not have to take a contiguous six hours.