Genshin Impact 7.1: The Gacha Machine and the Pricing Math Behind the Banner Schedule

The night version 7.0's second phase closed, I sat in front of a screen with a forty-seven-person chat group. Nobody talked tactics. Nobody argued about team composition. Everyone just counted — eighty-eight pulls, eighty-nine pulls — then a collective sigh when pull number ninety returned a character outside the limited banner.

I grew up with football's units of measurement, where a moment is remembered by a goal. Here, the unit is a pull, and emotion is pre-programmed by a probability table. I started hiding behind a keyboard during the 2026 World Cup, and then I could not stop writing. And that night the screen counted to ninety made me realise I was watching something closer to a trading floor than a game.

Context

Let me say this plainly up front to avoid confusion: Genshin Impact is not an esports discipline. It has no professional circuit, no world championship, no club system, no player transfer market, and no competitive balance patches. What it has is a gacha monetisation model that runs on a version rhythm, and that rhythm is what deserves analysis.

Each version splits into two phases, roughly twenty-one days each, and each phase opens its own banner set. Version 7.0 closed with a second phase built around reruns for Flins and Ineffa. Version 7.1 opens with a first phase featuring two new characters debuting simultaneously, named in the source article as Vesna and Vodyanitsa, before the second phase returns to reruns.

Beneath that schedule sits a pricing rulebook. The pity mechanic guarantees a five-star character within ninety pulls. On an event banner, the first five-star has a fifty percent chance of being the featured limited character and a fifty percent chance of being a standard one; if it lands on the standard branch, the next five-star is guaranteed featured. The notable detail is that pity is shared across banners of the same category, meaning accumulated pulls retain value when a player switches to another banner in that group. Finally there is Chronicled Wish, a separate banner type for older characters, operating under its own rules.

I took on the job of tracking this market the way I once tracked matches. Drawing on my experience watching matches, I learned to separate signal from noise before forming any judgement. The same applies here, and the noise is far louder.

Core analysis

First, on information quality: in the source dataset I have, twenty of twenty-eight information points carry no source. Only one cites an official channel from the publisher. Three are the author's personal opinion. Several character names and version numbers in the article cannot be cross-checked against known game state, meaning the risk of auto-generated or rumour-based content is real. The original piece itself concedes the exact banner schedule is still awaiting confirmation.

That is why I am not using the source article as a basis for telling anyone to spend. But the article still has value from another angle: it exposes the design principles of the whole machine.

Pricing architecture manufactures deliberate variance. The fifty-fifty plus ninety-pull guarantee is a design optimised simultaneously for perceived accessibility and margin. Players always believe they need just one more pull. But actual spending outcomes are widely dispersed: some hit the target at pull seventy, others must run two full pity cycles. That dispersion is the revenue source, not a system flaw.

Shared pity across same-category banners lowers switching cost. This is the detail most readers skim past. If accumulated pulls were not retained, players would have an incentive to dump their entire budget into one banner and stop. When pity is shared, the marginal cost of trying one's luck on the next banner falls, and total spending frequency rises. This is not a technical detail. It is a behavioural lever.

No fixed rerun schedule is a deliberate scarcity mechanism. Some characters are absent for over a year, others return within a few versions. When players cannot predict, they cannot optimise their spending plan. They can only react. And reacting is always more expensive than preparing.

The spending-pressure peak lands on a new version's first phase. Placing two new characters in the opening phase while the later phase is reruns only is an allocation decision. Players are forced to choose between two targets inside the same twenty-one-day window, or to top up. I have no revenue data to prove it, but structurally, this arrangement maximises decision pressure.

Chronicled Wish operates as a separate revenue lane for legacy assets. When long-standing characters are moved into their own banner type, the publisher can re-monetise them without occupying primary banner slots. This is a portfolio-management problem, not a power-balance one.

The last point, and perhaps the most important structurally: the publisher is simultaneously the game operator, the gacha rule-maker, and the official information authority. No independent arbitration mechanism sits in between. In football, referees can err, but at least there is a governing body, a federation, and an appeals system separate from the clubs. Here, everything sits in one hand. That is a higher concentration of power than any esports ecosystem I have analysed.

I must state my limits clearly. This analysis rests on a description of mechanics and a dataset with a very low sourcing rate. The sample is small, the filter bubble is real, and regional context is a variable I do not control. Conclusions about spending behaviour are inferences from architecture, not verified revenue figures.

The transfer market is like a chess game, but I choose to read it with my heart rather than numbers. Here it is the opposite: I am forced to read with numbers, because the heart has no pity.

Contrarian angle

The most common reading of a banner schedule is: which character is stronger, and is it worth going all-in. I think that reading is on the wrong axis from the start. The question is not how strong the character is, but how hard the decision window is engineered to make saying no.

An article about a banner schedule does not give you information to choose correctly. It gives you a deadline to feel pressure against. The entire promotional tone, combined with a low sourcing rate, turns much of this content type into a traffic filter rather than verifiable reporting. Readers walk in looking for an answer and walk out holding a deadline.

In the other direction, there is something this model does considerably better than esports. It depends on no external cultural or sporting event to generate revenue. No match calendar being postponed stops this machine. But precisely because of that, it exposes itself to a different risk: rules on probability transparency and consumer protection. A legal change in one large market can affect the entire pricing architecture, whereas a fixture change only shuffles a few rounds.

This is where I could be wrong. If the publisher publishes a fixed rerun schedule in future, the scarcity-mechanism argument weakens sharply. If the sourcing rate of such articles improves, the traffic-filter argument loses most of its weight. I accept both possibilities.

Open takeaway

The test I set is specific: if version 7.1's first phase genuinely opens with two new characters simultaneously, and if shared pity remains unchanged, then total pulls in that twenty-one-day window will exceed the corresponding rerun window. That is a checkable prediction, not a feeling.

At twenty-two, I realised I was not only commenting on football — I was telling human stories through every phase of play, and now through every pull. What is worth pondering is not who wins the banner race, but why we still let a probability table decide how we spend.