how it works
the cascade thesis · the glossary · what this is not
the thesis
hype moves through platforms in a predictable order: telegram first, then reddit, then tiktok. each stage is a different kind of person finding out.
telegram is where a token gets discovered — small groups, early callers, people who found it before it had a chart worth looking at.
reddit is the middle. bigger audience, slower, more skeptical.
tiktok is the end of the line. by the time something is trending on tiktok, the people who found it on telegram are already deciding whether to sell to you.
hype travels telegram → reddit → tiktok. by tiktok, you're the exit liquidity.
that's the whole thesis. doodah exists to show you where a token is on that path, not to tell you what to do about it.
glossary
everything on the board, defined exactly.
velocity
cascade-weighted mentions per hour. telegram counts 3x, reddit 1x, tiktok 0.5x, because a telegram mention today is worth more than a tiktok mention today — it happened earlier in the cascade. a bigger number just means more weighted chatter right now. it is not a price signal.
accel
this hour's mention rate compared to the last 6 hours' average. it's the only field that answers "is this speeding up or slowing down," which is the actual question, not "how big is it."
- ×1.2 and above → accelerating
- ×0.8 to ×1.2 → steady
- below ×0.8 → cooling
SUB-FLOOR
the token has enough combined mentions to be on our radar, but hasn't individually cleared any single platform's noise floor yet. this is the "we're watching, nothing confirmed" zone. most tokens live here and never leave.
confidence
how many platforms are independently showing this token above their own noise floor, right now. not a quality score, not a prediction — corroboration.
- UNCONFIRMED — on the board, confirmed nowhere
- LOW — one platform confirms it
- MODERATE — two platforms confirm it
- HIGH — all three confirm it
this is a live read, not a permanent record — if the mentions stop, confidence drops back down on its own.
lifecycle
where a token sits on the cascade path:
- seed — first signs, unconfirmed anywhere
- discovery — confirmed on one platform, moving at a normal pace
- acceleration — confirmed on one platform and breaking out
- expansion — confirmed on two platforms, tiktok not among them yet
- retail — tiktok chatter has started appearing, ahead of tiktok's own noise floor
- distribution — the cascade reached tiktok. this stage never expires.
- dead — momentum has been completely flat for 30+ minutes. overlays seed through expansion; a token can die before it ever takes off.
STALE
the dead-overlay above, spelled out: velocity at zero and accel below ×1.0 for six straight scoring cycles. a token that looked promising two hours ago and hasn't moved since is stale, not early — the label updates to say so instead of quietly staying wrong.
DERIVATIVE
the ticker looks like a copycat of something already on our radar in the last 7 days — either it contains an older ticker's name, or it's one character off from it. this label never hides or suppresses the token. it still scores, still alerts, it just also tells you a namesake exists so you can check you're looking at the right contract.
rug gate
automated on-chain checks, run against every token: is the LP unlocked, does one wallet hold an outsized share, is the mint still live, is liquidity too thin to trade. any flag raised here is shown, never hidden.
what the rug gate can't catch: a team quietly selling through many small wallets, a contract with logic our automated scan doesn't flag, or anything that just requires someone reading the contract by hand. a clean rug gate means the obvious, mechanical scam patterns aren't present. it does not mean the token is safe, and it says nothing at all about whether it's worth anything.
what doodah is not
doodah does not tell you to buy or sell anything.
it does not predict prices. nothing here is a price target, and no field on this board should be read as one.
the patterns it surfaces (cascade stage, acceleration, distribution) are pattern-matching against what's actually been observed for that token — not a model, not a backtest, not a guess about what happens next. if the history isn't there yet, neither is the pattern.
most tokens on this board go to zero. that includes ones flagged clean, ones with high confidence, ones accelerating right now. attention is not value. this tool measures the first thing. it has no opinion on the second.