Onimusha: Way of the Sword, 36 Boss Fights and an "Esports" Label That Does Not Fit

Incheon, Tuesday morning. While clearing up the newsroom tagging sheet, I found a line sitting in the same "esports" column as my LCK match reports: "How long does Onimusha: Way of the Sword take to beat?" The person who filed it was not being careless. "Capcom 2026," the community heat, the flood of streamer guides — all of it pushed a consumer buying explainer into a drawer it does not belong in. I read the piece four times, then took a notepad and checked every number. What came out was more interesting than the game itself: a marketing document wearing the clothes of measurement, and a label stuck to the wrong product. Two things I have run into far too often, both in esports newsrooms and in the way the media handles women's sport.

Onimusha: Way of the Sword is a single-player action title from Capcom, one of three major releases the publisher has scheduled for 2026 alongside Resident Evil Requiem and Pragmata. It is the return of a franchise that went quiet after the PlayStation 2 era. The name has been rebuilt, the setting pushed to Eastern Kyoto in the Edo period, and the messaging chosen very carefully.

That messaging fits into one sentence: the Onimusha campaign is significantly longer, with more hours of content than Resident Evil Requiem and Pragmata combined.

This positioning is not new for Japanese publishers. Runtime is the easiest unit of value to sell to a single-player audience: it turns money spent into hours received, a division anyone can do in their head. But the original piece never supplies the runtime of either comparison title. A comparison without a denominator cannot be verified, and it cannot be falsified either. It can only be believed, or not.

In Korea, where I work, there is another layer. Three major titles from one publisher in one year split a single wallet. Premium single-player revenue concentrates almost entirely in the launch window: players pay once and, within weeks, the money flow nearly stops. There are no seasonal skins, no battle pass, no continuously operated store anywhere in the source data. Which means the publisher has to sell now, and "longer than both of those combined" is the line built to win the first purchase in a player's wallet.

Now to the actual numbers.

The original piece offers a fairly tight set: players on Action difficulty need 30–40 hours to finish; players unused to action games, or chasing full completion, need 50 hours or more; the main story runs about 30 hours; the game contains 36 boss encounters; the optional content layer runs 10–15 hours; and to earn the PlayStation platinum trophy, a player must finish the game at least twice.

The first error sits between two sentences. The main story "about 30 hours" and Action difficulty "30–40 hours" do not match. The gap is small, but in a piece that sells runtime as its headline, two different figures a few paragraphs apart tell you about the method behind them: nobody measured, everyone estimated.

The most revealing number is the 36 bosses. Divided by roughly 30 hours, the density lands at about one boss every 50 minutes. For a single-player action game, that is an unusually heavy cadence, and it says a great deal about the product's structure. High boss density cuts both ways: it delivers the feel of constant climax, but it is also where players notice repetition fastest. A game can hold that rhythm if each boss brings its own move set, its own arena and its own rules. The source only notes that some fights are "arguably tougher" than others, without saying whether that difficulty comes from new design or simply a longer health bar. At that point, 36 stops being a number of variety. It can be a number of padding.

The 10–15 hour optional layer works the same way. It is not purely collectible. Side quests and Ace Archer challenges return iron and leather, the materials used to upgrade Musashi's sword and clothing, and the source describes them as essential. When core upgrade materials sit behind optional content, players who rush the main story walk into late bosses under-geared. That is a familiar design tension in Japanese action games, and it is the most practically useful thing the source lets slip: to have an easy endgame, you have to work in the middle.

The rest of the optional layer differentiates clearly. Curious Cases carry story content about myths and legends around Eastern Kyoto in the Edo period; treasure hunts are collectible work; chance encounters are colour; Lion Dogs tie to sword skins. Two trophies are named, "No Job Too Small" and "For the Love of Dog," which shows the developers still enjoy naming achievements with a straight-faced joke.

Then there is the two-playthrough requirement. This is the heaviest structural point in the whole dataset. It turns 30–40 hours into a minimum of 60–80, and it turns that 50-hour completionist figure from a claim into reasonable arithmetic: 10–15 hours of optional content, plus a second cycle. Mathematically the source is self-consistent here. Experientially, it means something very different from a game with 50 hours of new story. The second cycle is built for speed and trophies, most likely carrying progression forward, not for new narrative.

One more variable the source places on the table and then abandons: difficulty as a runtime multiplier. The same product takes a practised action player 30–40 hours and an unpractised one 50 or more. That spread is roughly 67%. When the range of playtime is decided by player skill more than by design, runtime stops being a property of the product and becomes a property of the buyer. Which makes every cross-title runtime comparison methodologically fragile unless it states its difficulty setting and its reference player.

For years of covering competition, I have grown used to every number having a source: win rates, creep scores, game times. Esports built a verification culture because prize money and qualification slots depend on it. Single-player games carry no such pressure. There, a playtime figure is handed from one article to the next, losing a little more of the original measurer's fingerprints each time, until it becomes something everybody repeats and nobody traces.

And what the source leaves out matters too: platform availability, price, age rating. For a piece that exists to help readers decide whether to buy, missing all three is a bigger hole than any hour count. I have learned to listen to what the pitch whispers when nobody is filming — and here the silence says the piece was written for a search engine, for a "how long to beat" query, not for a buyer.

This is where I have to say plainly what that Tuesday tagging sheet got wrong.

Across the entire Onimusha dataset there is no team, no roster, no coaching staff, no tournament, no balance cycle, no prize pool. No qualifiers, no slots, no transfers. What exists is a platform-issued trophy and achievement framework and a single-character upgrade screen. Tagging that product as esports and feeding it into an esports analytics pipeline is contaminating your own data.

I am not annoyed by a technical error. I am annoyed because that label is money, headcount and attention. A slot on the esports desk spent on a runtime explainer is a slot not spent on actual competition. An unannounced door often opens onto the biggest stadium — but here it runs the other way: attention flows automatically toward whatever already makes noise, not toward what actually has competitive structure. The women's circuits I follow have teams, rosters, schedules and prize pools enough to earn that label, and they are still waiting.

The second counter-intuitive point is how we measure value. The industry is using runtime as the yardstick, and "longer than both combined" is the peak of that habit. But a 30-hour game that repeats can be worth less than a 12-hour game that does not. The honest measure of a single-player product is not hours; it is the number of distinct decisions the player is forced to make. Thirty-six bosses sounds magnificent. Thirty-six bosses handled thirty-six different ways is the thing worth discussing.

The third point is timing. Three major titles from one publisher in one year are not three independent opportunities; they are one endurance test for a wallet. When a publisher has to compare its own products to its own products, that signals internal competitive pressure rather than confidence. The "longest" positioning is therefore defensive more than promotional.

And the comparative claim itself remains unverifiable. No runtime for Resident Evil Requiem. No runtime for Pragmata. No method, no sample, no measurer.

Three signals are worth tracking over the coming months: crowdsourced completion times, if they diverge more than 20% from the 30–40 hour mark; player feedback on the pacing of 36 boss fights; and sales performance across all three Capcom titles in the same launch window. In sport, the most important match sometimes takes place behind the dressing-room door — and for a single-player game, that match happens in player reviews, weeks after the store page opens. Whoever measures first speaks first. Whoever only repeats someone else's number is just holding the microphone.