How to judge a crypto caller: timestamp, entry, peak after
Short answer A call is a claim about a moment. Without the time, the market cap at that time and a peak measured after it, a screenshot proves nothing. How to check one.
You want to know whether a caller is any good. Most callers make the judgement impossible, not by lying about a token, but by leaving out the one detail that turns a claim into evidence: when the call went out and what the token was worth at that second. Without it, a screenshot is a picture of a chart. This lesson gives you the three fields a call needs, the arithmetic trick that hides behind most track records, and where to check ours.
A call is a claim about a moment
"Token X did a hundred times" is not a statement about Token X. It is a statement about a starting point and an ending point. Move either one and the number changes.
The starting point that matters is the market cap at the moment the call was published. Call it the entry. If someone posts about a token after it has already run, the entry is inflated and the multiple they later claim was never available to anyone reading them.
Three things have to be present before a call can be judged at all.
- A timestamp. The moment of publication, precise enough to identify a price.
- An entry recorded at that moment. Not a round number, not "around here". The market cap when the words went out.
- A peak measured only after that moment. A high that happened before the call is not the call's high. It is someone else's.
Miss one of the three and the claim cannot be checked. A claim that cannot be checked is not evidence, and a caller who leaves the fields out is keeping their options open.
What counts as a winner
A tall candle is not a win. Most tokens on Solana can print a tall candle for a few minutes. The useful definition has two parts: a named signal fired before attention arrived, and the token's supply structure was survivable once it did. A launch with a clean distribution behaves differently from one where two wallets hold the float and the pool is the only exit. Same chart at minute three, different outcome by minute thirty.
So when we log a call we log the lane that fired and the structure read attached to the supply, along with the market cap and the time. That pairing separates a tracked entry from a hindsight screenshot.
Timestamp discipline
Ask three questions of any roundup. Flagged at what market cap? At what time? On what signal? A list missing any of the three is entertainment. Our Every call table on /tables prints all three for the last thirty days on each chain, with the misses kept in, and the Room calls table takes the entry as the first reading at or after the post, never before it.
Survivorship, shown with arithmetic
Survivorship bias is judging a set by the members that survived. A fund that closes its bad portfolios looks excellent. In crypto it works one deleted post at a time.
Our own Solana log, as recorded, runs to 40,415 calls between 27 January and 3 October 2026. Of those, 8,416 reached double the market cap where we logged them, which is 20.8%. Meanwhile 21,583, or 53.4%, never got ten percent above it.
Now watch one editing decision. Drop the 21,583 that went nowhere out of the denominator. The same 8,416 doubles now sit on a base less than half the size, and the apparent rate climbs toward one in two. Nothing happened in the market. Only the denominator changed.
That is the whole trick. A caller who mentions only the tokens that worked is reporting a denominator they chose after the fact, and because the losing half is the boring half, nobody screenshots it.
Peaks have to be proven too
A timestamp fixes the entry. The peak needs its own discipline. A single price reading on a thin pool can be false. One painted pool, or one wrong trading pair, can make a token read at many times its real value for a few minutes, and a screenshot of that reading looks like a win.
We learned this on our own records. On 2 October we re-checked stored wins against repeated readings taken after each call. Of fifty-seven that failed, the median corrected multiple was 0.7x, and only four remained at ten times or more. We deleted the posts built on the false readings. A peak counts when more than one reading after the call sits near it. Ask a caller the same thing: how many readings, from where, and when.
The clock matters as much as the price
Timing decides whether a multiple was reachable. On Solana, the forty-seven calls since 26 September that reached ten times their entry hit their peak reading a median of 4.1 hours after the call; 91% of them peaked inside a day. On Robinhood Chain the runners are slow: of two hundred calls whose own snapshots reached ten times, the peak came a median of 52 hours after we logged them, and under a third inside a day.
So a Solana call shared a day late is a call shared at the top. A caller who posts the chart after the move has shown you history. The timestamp on the original post is the only way to know which one you are looking at.
Compare like with like
The last trap is comparing a caller in one kind of token to a caller in another. On Robinhood Chain between 5 August and 2 October 2026, as recorded, meme tokens reached double 4.6% of the time (n 19,587), stock-paired tokens 12.1% (n 1,775) and utility tokens 24.6% (n 1,215). The odds differ before anyone opens their mouth. A caller working in utility tokens is not several times as skilled as one working in memes because their doubles are more frequent. Look at the size of the bucket too; in a population of a few hundred, a percentage moves with each week of data.
What this does not tell you
A full set of receipts tells you what happened after each call. It does not tell you what will happen after the next one.
Entry here is our first sighting, often minutes after the poster's. A real fill is later and worse.
A peak is a reading, not a fill. Nobody sells the top, and a small pool cannot absorb a sale the size of its own liquidity.
Our log holds what our scrapers and callers surfaced. A token nobody mentioned is not in the denominator, so even our base rates describe a selected set.
What to do next
Before you follow any caller, pick one of their calls and find the three fields: the time, the market cap at that time, and a peak measured after it. If one is missing, you cannot judge the call, and neither can they. Then open Every call on /tables and read a month of ours, misses included, and the base rates with their dates on /research.
The lesson is the method. The product is where you apply it.
Members get every call 30 minutes before the public channel. See Premium
Paste any Solana or Robinhood Chain contract into the PumpPill trade bot: you get the risk read first, then buy or sell in one tap, with limit orders and auto-sells.