The Pity Architecture: Dissecting the Revenue Machine Behind a Banner Schedule

In the empty host room, long after a tournament feed had gone dark, I still kept the habit of reopening old data sheets. Not to find a winner. I look for structure. That night, next to the usual stats file, another file slipped onto my second monitor: a banner schedule for an open-world role-playing game, with a few lines describing its pull mechanics. Ninety pulls for a five-star character. A fifty-fifty split between the featured character and a standard one. Twenty-one days per phase. I read it slowly, the way I once read a support player's vision-score sheet at the twenty-seventh minute. And I realized I was looking at a revenue machine assembled as carefully as any tactical system. I have to say one thing first, because it belongs to the discipline of the trade. The document I read is not esports coverage. It is a schedule explainer, written for players deciding whether to spend or to save. Of twenty-eight information points, twenty carry no source, exactly one cites an official publisher announcement, and three are the writer's own opinion. Several character names and version numbers could not be cross-checked against known public data. I noted this, then set it aside. What kept me sitting there was not the schedule. It was the architecture behind it. Each version of this game splits into two phases, roughly twenty-one days each, each with its own banner. Some phases carry new characters; others carry reruns. The source says the next version opens with two new characters sharing phase one, then moves to reruns in phase two. That is the entire schedule. The rest is mechanics. A five-star character is guaranteed within ninety pulls. On a featured banner, the first five-star has a fifty-percent chance of being the promoted character and a fifty-percent chance of being a standard one. If it lands on a standard character, the next five-star is certain to be the promoted one. The community calls that the guarantee. One detail is rarely mentioned: pity is shared across banners of the same type, meaning accumulated pulls carry over between them. Finally there is the Chronicled Wish, a separate banner type with its own rules, usually for older characters, running alongside the main system. An esports viewer will find this familiar. This is a season. Twenty-one days per phase echoes the rhythm of a round. But one difference I must keep clear: there are no teams, no players, no brackets, no match rules in this document. The names that appear are game characters. If I assigned them the role of a roster, I would be betraying my own principle. This is a sales system, not a competition system. Yet its structure is clean enough to dissect. The ninety-pull soft floor does two things at once. It creates a sense of safety, because a player knows that however unlucky they are, there is a stopping point. And it sets a maximum imaginable spend, a figure players can rehearse in their heads before pressing the button. But that very ceiling opens the door to variance. Most players never reach ninety. They reach it at different times, with different amounts, and that uncertainty, not a fixed price, is the real revenue source. The fifty-fifty mechanic is subtler than it looks. With a fixed rate, players could calculate the exact expected cost per character. A fixed rate turns pulling into arithmetic, and arithmetic produces no emotion. But when the first five-star is a coin flip, later compensated by a certain outcome on the next beat, what gets created is not a cost but a memory. Players remember the time they nearly missed. Players remember the time the guarantee saved them. That memory costs more than the average figure, and it brings people back. I once sat for a long time over a player's stat sheet just to understand why a seemingly meaningless action had value. That reading method applies here. The guarantee does not exist to make players happy. It exists to turn a spending behavior into a story with a beginning, a climax, and an ending. When a system knows how to tell a story, people stop buying a character. They buy a passage of memory. Then comes shared pity. In the older system, moving from one banner to another meant discarding accumulated progress. Switching cost was high, so players had to commit. When pity is shared, switching cost nearly vanishes. Players can jump from a debut banner to a rerun without losing anything. It sounds like a concession to players. Seen from the cash-flow side, it is a revenue-smoothing mechanism: it keeps spending flowing evenly through both debut and rerun windows instead of crashing in waves. And then rerun policy. The source records one fact: there is no fixed rerun schedule. Some characters vanish for more than a year. Others return within a few versions. This uncertainty is not an operational flaw. It is scarcity by design. When players do not know when the next chance comes, they must decide in the present. Delay becomes expensive. And the fear of missing out becomes an invisible fee that players pay voluntarily. Here the structure peaks: phase one places two new characters at once, phase two retreats to reruns. Players do not merely face the question of whether to pull, but the harder one: pull for whom, and save for whom. Two new characters in parallel means two pressures stacked on one resource. This is a design that forces players to rank their own priorities, and while ranking, many will spend more than they originally intended. The Chronicled Wish serves as a secondary lane. Older characters, instead of waiting for a slot on the main banner, are routed into their own system. This frees the main schedule for new characters while opening another revenue door for characters thought to be dormant. Two goals, one mechanism. To a structural observer, this is a move worth recording: the publisher does not only sell the new; it also finds a way to resell the old without diluting the calendar. Placed beside esports, these are two fundamentally different revenue worlds. Esports lives on sponsorships, broadcast rights, jersey revenue shares, and partly prize pools. Its cash flow depends on third parties: sponsors, platforms, audiences. The gacha model lives on direct, recurring spending contained entirely inside the game, collected by the publisher itself. No season needed, no roster needed, no sponsorship contract needed. A closed loop: new content creates desire, desire creates spending, spending funds the next content. That difference carries a difference in risk. A tournament can collapse because it loses a sponsor, because rights values fall, because audiences turn away. A gacha machine depends on no third party, so it weathers calendar shocks far better. What it cannot weather is law. A single large market tightening probability disclosure or restricting minors' spending can force the entire architecture back to the drawing board. Meta does not die; it transforms into another poem. In esports, what transforms is the playstyle and the roster. Here, what transforms is the list of characters worth pulling. But the motive underneath is the same: a community constantly taught that the next version will be better than the current one. That is the part I find convincing. The rest is where I have to stop and check. The source tells only when. It does not tell whether it is worth it. It names characters but offers not a line on power, kit, or real combat value. The entire implication of saving for the next version rests on a data vacuum. Readers walk away with a schedule, not a judgment. This is a content type I call traffic-filter content: enough to keep people on the page, not enough to help them decide. People think they are reading the match; it turns out the match is reading them. With gacha, the thing reading players is the banner schedule itself. Every time a player opens a schedule explainer, they leave a trace: they are weighing, hesitating, looking for a reason to spend or to hold. A schedule does not merely announce. It observes. Then there is the question of power. The publisher writes the pity rules, publishes those rules, and is the sole beneficiary of them. No independent arbiter verifies that the disclosed rates are accurate. No third party checks that a character truly returns on time. This is the same concentrated structure I have long seen in esports, where a publisher both runs the tournament and owns the game. But in esports there are at least teams, players, and public opinion as counterweights. Here, the only counterweights are the player community and regulators in a handful of markets. And of course, I cannot ignore the weight of evidence. Twenty of twenty-eight information points lack a source. Exactly one cites an official announcement. Several characters and version numbers cannot be verified. This does not mean the content is false. It means it is unconfirmed. To someone who reads data for a living, the distance between unconfirmed and false is the distance between reportage and rumor. I do not fill that gap with belief. Across years of tracking transfers and patches, I learned one habit: always separate the question of what pain a team went through from the question of what the team did while in pain. The same applies here. That players feel spending pressure is one thing. How much they choose to spend is another. Blending the two is the fastest way to write either a hymn for victims or a verdict for the defeated. One detail is fairly optimistic: the piece itself admits the exact banner schedule is still awaiting confirmation. Content willing to say I am not sure is more honest than the average of its genre. But honesty about uncertainty does not make content certain. It only makes it easier to read, and easier to read often means easier to believe than it deserves. Seen whole, this is good training in distinguishing two kinds of text. The first tells you what is about to happen. The second tells you what is worth waiting for. The first is service. The second is analysis. The piece I read belongs to the first, even though it wears the tone of the second. I do not predict the future; I only listen to the past whispering. And the past of the game industry tells me something very old: every monetization model has a life cycle. There was an era when selling discs was king. An era when subscriptions were king. An era when gacha is king. Each model is born, thrives, then is replaced or bent by law into a new shape. Gacha today is an almost perfect machine, a closed loop dependent on no third party. But that very perfection places it closest to two pressures. The first is the wave of regulation. Probability disclosure, protection of young players, and debates over randomized reward boxes are advancing slowly but steadily into major markets. The more a system depends on player uncertainty, the more sensitive it is to transparency demands. The second is player fatigue. After years in the same loop, people begin to learn to see the structure, and once they see it, they are less led by it. The meta we love today is the meta we cry for tomorrow. Not because that meta is bad, but because what is loved too easily is also abandoned too quickly. This holds for a champion in a competitive title, for a tactic at a World Cup, and for a monetization mechanic earning more than any World Cup. But the difference is here: when a team is left behind by the meta, we can count the people who hurt. When a monetization mechanic is left behind by law, the people who hurt number in the millions, and no one counts them. For an observer like me, the interesting question is not which character debuts next. The interesting question is: after how many seasons like this, can players still ask themselves what they actually want, or have they let the banner schedule ask on their behalf. Because a schedule always knows how to ask. And it always offers exactly one answer. What I learned across years standing in host rooms is this: every system reveals its nature through what it does not say. A banner schedule never says you should pull. It simply stays silent, and inside that silence, players write their own lines. That is the best storytelling technique I have ever seen, and I give respect to whoever designed it, while keeping a professional distance from it. The stands are empty, but the echo is full. Here, the stands are a phone screen at two in the morning. The echo is a finger resting on the pull button.