Cross-game meta
Every game side by side in one place — how generous each gacha really is, and how fast each roster’s ceiling climbs. One normalized view.
Gacha generosity
What it actually takes to land one named featured unit in each game, on the same axis — the same question in every row, never “any of the rate-ups” in one and a specific unit in the next. The chart compares the averages at a glance; the table below it gives every figure — the range from a typical run to an unlucky one, and the pull where the game finally guarantees the unit. Four have no guarantee in the pull at any count. One has a ceiling the game runs that is not part of our data yet, marked “not carried” rather than counted either way.
| Game | Grade | Pulls to the featured unit | Guaranteed by | Pity mechanic | |
|---|---|---|---|---|---|
| Silver And Blood | Generous | 50 avg 36–111Measures a rate-up banner — the 80-pull guarantee gives an S-tier drawn evenly from a 26-unit pool, so it is a guarantee of the RARITY and reaches the rate-up unit only about 4% of the time | No pull guarantee | Hard pity at 80 · ~50% featured rate-up | Planner → |
| Blue Protocol: Star Resonance | Generous | 65 avg 68–71 | 80 | Soft pity from 65, hard pity 80 · no featured split — the pool's SSR is the advertised reward | Planner → |
| Reverse: 1999 | Generous | 65 avg 70–123 | 140 | Hard pity at 70 · 50/50 featured | Planner → |
| Arknights | Generous | 69 avg 57–142Measures standard headhunting with ONE rate-up 6★ — our rates carry the ramp for standard banners, and a banner sharing its rate-up between two 6★ operators takes more pulls to a named one than this row shows | No pull guarantee | Soft pity from 51, hard pity 99 · ~50% featured rate-up | Planner → |
| Azur Lane | Generous | 76 avg 58–191Measures the Wishing Well — the one pool where you name the ship you are chasing; a standard build has no featured ship at all, so its UR rate answers a different question | 200 | 200-pull spark guarantee | Planner → |
| Wuthering Waves | Generous | 83 avg 75–145 | 160 | Soft pity from 70, hard pity 80 · 50/50 featured | Planner → |
| Hololive Dreams | Middle of the pack | 87 avg 69–200 | 200 | 200-pull spark guarantee | Planner → |
| Persona5: The Phantom X | Middle of the pack | 88 avg 74–145 | 160 | Soft pity from 70, hard pity 80 · 50/50 featured | Planner → |
| Genshin Impact | Middle of the pack | 91 avg 80–156 | 180 | Soft pity from 74, hard pity 90 · 55/45 featured | Planner → |
| Duet Night Abyss | Middle of the pack | 94 avg 81–156 | 180 | Soft pity from 74, hard pity 90 · 50/50 featured | Planner → |
| Honkai: Star Rail | Middle of the pack | 94 avg 81–156 | 180 | Soft pity from 74, hard pity 90 · 50/50 featured | Planner → |
| Zenless Zone Zero | Middle of the pack | 94 avg 81–156 | 180 | Soft pity from 74, hard pity 90 · 50/50 featured | Planner → |
| Limbus Company | Middle of the pack | 106 avg 96–200 | 200 | 200-pull spark guarantee | Planner → |
| Blue Archive | Middle of the pack | 108 avg 99–200 | 200 | 200-pull spark guarantee | Planner → |
| Fate/Grand Order | Steep | 116 avg 87–287 | 330 | 330-pull spark guarantee | Planner → |
| DIGIMON UP | Steep | 127 avg 139–200 | 200 | 200-pull spark guarantee | Planner → |
| Umamusume: Pretty Derby | Steep | 133 avg 93–306Measures the character (Pretty Derby) scout — the support-card scout is a separate gacha | Not carried | No pity on the pull — every summon independent (the game's Scout Exchange Point ceiling is not in our data yet) | Planner → |
| Dragon Ball Z Dokkan Battle | Steep | 140 avg 97–322Measures the summon itself — a banner's own guaranteed-SSR step is carried in our data but not modelled here, and it only improves these odds | No pull guarantee | No pity — every summon independent | Planner → |
| CookieRun: Crumble | Steep | 167 avg 116–383 | No pull guarantee | No pity — every summon independent | Planner → |
Sorted most generous first. Grade is the average alone: ≤85 pulls = generous, ≤110 = middle of the pack, above that = steep. Whether a game guarantees the unit at all is a separate column, not folded into the grade — two games can average the same number of pulls while one caps your losses and the other never does. The figures already account for each game’s pity or spark, and are pull counts, not spend (a pull costs a different amount in every game).
Power-creep velocity
How fast each roster’s ceiling climbs — the 95th percentile of a game’s headline stat, compounded over 30 days so games that patch at different rates are comparable. It measures the headline stat only, not a unit’s kit, and a game is charted only once we have archived at least 180 days of it; shorter windows are listed below as first readings.
First readings, not trends. These games have a measurable change but a window shorter than 180 days — too short to tell a real balance trend from one well-timed release, so their figures are listed rather than ranked. They join the chart above once the archive covers enough time.
Not enough history archived yet for these — a velocity needs at least two archived builds to compare. Every build is kept permanently, so they join the chart as soon as they have a second one.
These are archived, but their data has no single headline stat that means the same thing across the roster, so there is nothing honest to compound. They stay out of this chart rather than being given a number.