The Transfer Window Without a Contract: Reading Genshin Impact's 7.0 and 7.1 Banner Schedule Like a Market Analyst

Incheon, a late October night. I sat in my small studio, headphones still around my neck, and on my second monitor was a friend livestreaming. He doesn't play esports; he plays Genshin Impact. His hand rested on the mouse, eyes fixed on the Primogems left after his third pity threshold, and he turned to me with a question I have heard hundreds of times in eleven years on the job: "Spend now, or wait for the next version?"

I didn't answer immediately. In my line of work, that question has never had only two options. It is an investment decision, and every investment decision needs data before it needs emotion. My friend was standing at a transfer window with no contract at all — where what is traded is not a player, but the right to access a character within a finite time frame. I reopened my notebook, pulled his schedule onto the table, and began reading it the way I read any transfer window: with numbers, with clauses, and with the question of who writes the rules.

Genshin Impact runs on a rhythm any sports-media professional should memorize: each version splits into two phases of roughly 21 days each. In each phase, the publisher opens one or more banners — pools — where players spend Primogems for a chance to obtain a character. There is no arena here, no roster, no coach, no contested power ranking. I need to say this plainly from the start, because honesty with readers is something I do not negotiate. Labeling this content "esports" would be a classification error, and classification errors, in my profession, are the most expensive kind. A PvE game running on gacha has a content cycle, not a competitive-balance cycle; its "updates" are content launches, not power adjustments.

What is being discussed in the current window involves two milestones: phase two of version 7.0 and phase one of version 7.1. For 7.0 phase two, the two most-cited names are Flins and Ineffa, appearing as reruns — players could already have obtained them previously, and now they return. For 7.1 phase one, circulating information points to two new characters at once, Vesna and Vodyanitsa, tied to the Snezhnaya region. Phase two of 7.1, by description, leans toward reruns, including Skirk and Escoffier. Other names surface scattered through community discussion, such as Odette, Aino, Iansan, and Lan Yan, but they do not form a continuous schedule.

I must state the reliability of each line above. In the dataset I hold, most schedule information carries no source; only one point rests on an official publisher announcement, and the source text itself admits the exact banner schedule "remains to be confirmed." By my three-source verification standard, this is a gray zone. A gray zone does not mean worthless; it means it must be labeled. Numbers speak, but I learned to hear them after the 140 million shock — when I once posted a beautiful big number without checking the release clause, then had to delete the post at 2 a.m. while 64% of responses were negative. Since then, every fee, every term, every expiry date I read on air must pass double verification before reaching listeners. Tonight is no different.

What makes a system like this analytically valuable is not the schedule but the pricing architecture behind it. That is where I want to spend most of my time. First, the pity threshold. The common mechanism guarantees a five-star character within a maximum of 90 pulls. This is not a promise of value; it is a floor for cost. In transfer language, it resembles a release clause: the buyer roughly knows the maximum they must pay to secure the target, though few insiders ever state the figure publicly. The publisher does not sell a specific character; they sell a probability, and the 90-pull floor is what makes that probability look bearable. A player with a 40 million euro release clause is priced around that figure in every negotiation; a character with a 90-pull floor is the same, only in a different currency.

Then there is the 50/50 system. On an event banner, the first five-star hit has a 50% chance of being the featured character and a 50% chance of a standard-pool character. If you lose — drawing a standard character — the next five-star is guaranteed to be the featured one. This structure does not exist to be fair; it exists to create variance. Variance is the fuel of any luck-based system. One player may need up to 180 pulls for the character they want; another may get it on the first try. That gap is precisely what makes community stories lively and keeps money flowing through every window. If everyone knew they would pay exactly one fixed amount, there would be no story to spread. Variance creates anecdotes, and anecdotes create demand.

The single point I judge most important, and also the most misunderstood, is that pity is shared across banners of the same category. When your luck counter is not reset between a new-character banner and a rerun banner of the same type, the marginal cost of switching targets falls. Behaviorally, this reduces friction, makes it easier for players to jump from one target to another, and thus increases total spending frequency. In my profession, that is not a technical detail. It is a revenue lever designed to smooth cash flow across both debut and rerun windows. The publisher does not want you to dump money in one window and sit out the next; they want you always in play, always with a reason to pull one more time.

Alongside this is a rerun policy with no fixed schedule. Some characters are absent for over a year; others return within a few versions. This uncertainty is not publisher laziness; it is a scarcity mechanism. In professional sports, you know the whole season's schedule in advance. Here, you do not know when the one you want will return, and precisely because you do not know, you fear missing out. That fear is a currency. Fans see a click; I see 21 sleepless nights.

Then comes Chronicled Wish — a separate banner type with its own rule set, usually for older characters. I read it as a secondary revenue lane: a way to remonetize names that have gone dormant without disrupting the primary banner cadence. With it, the publisher need not push old characters onto the main banners; they have a dedicated channel to squeeze out the last drop. This is asset-management thinking: let no capital sit idle, let no asset depreciate through neglect. In football, people call it reselling a past-prime player to a smaller league instead of letting him rot on the bench.

And here I must name the only thing that can count as "meta": pull-priority power creep. There is no win rate, no pick-ban rate, no balance data. The only thing the community debates is which character is worth hoarding Primogems for. The source text offers no strength data at all; it only states timing. This is the core difference between a schedule piece and a decision-support piece. The reader gets an answer to "when," not "whether." In my profession, an article that only says which club a player joins without saying whether he fits the tactical system is an article that has not finished its job.

I have spent years comparing systems that seem unrelated. In football or esports transfer windows, you have a player, a buying club, a selling club, an agent, a time-limited contract, and a release clause. In the gacha market, you have a character, a publisher who is both seller and rule-maker, a pity threshold acting as a price floor, and a 50/50 mechanism acting as variance. The power structures differ entirely, but the decision logic of the person in the middle is strikingly similar: they are placed before a closing time window and must decide while information is incomplete. The difference is that sports fans can wait until kickoff to judge, while gacha players must decide before knowing the outcome.

Recall March 2026, when global football froze. I messaged 34 media staff at K League clubs, 16 replied, and I built a 47-page table comparing transfer values before and after the pandemic. The average fell 31.6%. 47 pages of data during a pandemic — when the world stopped, I kept scrolling the sheet. The lesson was not the decline figure but this: when the market freezes, those who prepared win and those who panicked lose. My friend tonight, hand on the mouse, sits at exactly that moment — except his market does not freeze, it simply closes every 21-day phase and reopens.

21 days without airing a single line, so that today I can tell a whole chapter. That is how I handle transfer news: I do not publish when I have only one source, even an exclusive one. With this banner-schedule dataset, I apply the same discipline. Information on Vesna and Vodyanitsa, on Skirk and Escoffier, on the Snezhnaya setting — I mark it "reported," not "verified." That distinction is not semantics; it is the line between an informant and someone complicit in rumor. In a market where 20 of 28 data points carry no source, readers deserve to know where they stand on the reliability map.

Here I want to push against the popular framing itself. The story most content tells is the schedule story: who arrives in which version, when it opens, whether to save. But the real story lies elsewhere: the power structure. In most esports ecosystems, at least three parties share power: the publisher, the tournament organizer, and the clubs. Here, the publisher is simultaneously the game operator, the gacha rule-maker, and the official information authority. There is no independent arbiter, no third-party probability verification. This concentration of power is higher than in any esports league I have analyzed. The publisher controls both supply and information flow.

The biggest blind spot of any "should you pull" article is that it assumes the schedule reflects value. The schedule only says timing; value comes from strength, kit, and team fit — things a schedule piece never touches. When content offers only "when" and never "why," it sells you excitement, not a decision. And excitement cannot be verified. This is the blind spot I see most clearly, and the one I once fell into myself — when I believed a beautiful big number about a deal only because many people repeated it.

Another point few name: pity rules and probabilities are defined and published by the publisher itself. There is no independent arbiter. This gives the system high resilience to calendar shocks — it depends on no external sporting event, no season postponed by a pandemic — yet it carries a fatal weakness before changes in regulation on probability transparency and minor protection. In my language, this is a systemic risk, not a personal one. It sits not in one banner but in the entire model. And systemic risk cannot be hedged by saving Primogems.

A wrong number can be forgiven, but reputation lost is hard to regain. That is why I do not call this content esports, even though I understand why people want to label it so. When you place a subject in the wrong category, you will apply the wrong toolkit, ask the wrong questions, and draw the wrong conclusions. A character is not a player, a banner is not a tournament, and a 21-day phase is not a knockout round.

My friend ultimately chose to wait. Not because I advised him, but because he read his own dataset and realized the best decision in an unconfirmed window is not to rush. I believe what the transfer market and the gacha market can learn from each other is not how to sell, but how to be transparent. One side learns to disclose clauses and probabilities; the other learns to verify independently before believing. Beyond that, every transfer window — contract or not — ends in the same place: the best-prepared person pays the cheapest price for their own excitement.

What I take from tonight is not a prediction of who appears in 7.1, but a question I will keep asking of every monetization model I analyze in the years ahead: who writes the rules, who verifies, and how much of the truth the person in the middle is allowed to know. When the transfer map bends along each source, I learn to read each curve — and the curve here shows a market where power and information rest in one hand, while the player must decide in the dark. That is the story worth telling, not the banner opening date.