We Graded 152,602 Crypto Calls and Published Every Receipt. One in Five Was Posted After the Coin Had Already Doubled
Short answer Caller Receipts is an open file of 152,602 graded crypto calls from 16,771 X accounts, with a tweet id and a contract on every row. 9.1% of calls reached 5x, 74.2% never doubled, and one call in five went up after the coin had already doubled.
Every day thousands of X accounts post a contract address and say "this one". PumpPill reads those posts, matches every contract, and grades each call against what the coin did next. Today we publish the whole file: one row per call, with the post id, the contract, the market cap at the call and the peak after it. Anyone can check our work.
Measured October 9, 2026. The file is calls.csv (152,602 rows, CC BY 4.0), with a README, metrics.json and SHA256SUMS. The same grading powers the Influencer Track Record Checker.
Two numbers we track
Caller Hit Rate is the share of a caller's scored calls whose coin reached 5x the market cap at the call. Pre-Post Run-up is the share of calls posted when the coin was already at 2x or more of the cap we first saw it at. The first tells you whether a caller delivers. The second tells you whether the call was early or an exit.
The headline numbers
| Count | Share | |
|---|---|---|
| Graded calls | 152,602 | 100% |
| Reached 2x after the call | 39,430 | 25.8% |
| Reached 5x after the call | 13,880 | 9.1% |
| Never reached 2x | 113,172 | 74.2% |
| Under $1,000 of liquidity at our last read | 139,014 | 91.1% |
Among the 2,604 accounts with 5 or more scored calls, 933 have never had a single call reach 5x. That is 35.8%. The median caller's single best call peaked at 8.0x, so almost everyone has one screenshot to show you.
One in five calls went up after the coin had already doubled
Since September 26 we keep a market-cap reading for every token we watch, which lets us read the cap at the minute of the post for 43,599 Solana calls, 28.9% of them. This chart is where the coin already was when the call went up, measured against the cap we first saw it at.
| Cap at the post, against first sighting | Calls | Share |
|---|---|---|
| Under 1x | 20,393 | 46.8% |
| 1x to 2x | 13,574 | 31.1% |
| 2x to 5x | 7,572 | 17.4% |
| 5x and up | 2,060 | 4.7% |
22.1% of calls were posted after the coin had already doubled. 4.7% were posted after it had already done 5x.
This matters because of how a call is usually graded, by us and by everyone else: entry is the first time the coin was seen, peak is the highest cap after. A call posted at the top inherits the run. Among the calls posted after a 2x, the standard grade counts 49.4% as a 5x. Grade the same calls from the cap at the post instead, and 6.4% reached 5x. For every call with a reading, a buy at the post reached 2x 20.9% of the time, 5x 6.0% of the time, and the median call went to 1.01x. That is the number a follower actually gets.
The file carries both: peak_mult (the standard grade) and mult_from_post (what a buy at the post could have reached) on every row that has a reading.
Bigger accounts are not better callers
| Followers | Callers | Calls | Reached 5x | Reached 2x | Callers never 5x |
|---|---|---|---|---|---|
| Under 10K | 2,458 | 122,815 | 8.9% | 25.5% | 35.6% |
| 10K to 50K | 117 | 8,819 | 7.7% | 21.1% | 35.9% |
| 50K and up | 29 | 624 | 14.6% | 32.2% | 51.7% |
The 50K-and-up tier is 29 accounts, and over half of them have never had a 5x; the 14.6% is carried by a few. Follower count tells you how many people will buy after a post, not whether the post was early.
How callers are spread
| Caller Hit Rate | Callers | Share |
|---|---|---|
| 0% | 47 | 5.7% |
| Under 5% | 127 | 15.4% |
| 5% to 10% | 307 | 37.3% |
| 10% to 25% | 255 | 30.9% |
| 25% and up | 88 | 10.7% |
Of the 824 accounts with 20 or more scored calls, 88 reach 5x on a quarter or more of what they post. The rest are the average coin with a louder voice.
By month
| Month | Calls | Reached 5x | Reached 2x | Under $1,000 liquidity now |
|---|---|---|---|---|
| 2026-05 | 15,283 | 7.2% | 26.6% | 97.7% |
| 2026-06 | 16,600 | 8.2% | 25.8% | 96.5% |
| 2026-07 | 21,702 | 6.3% | 19.8% | 98.9% |
| 2026-08 | 31,458 | 5.5% | 16.8% | 99.0% |
| 2026-09 | 31,506 | 8.9% | 24.1% | 87.0% |
| 2026-10 | 35,561 | 15.4% | 38.7% | 77.9% |
October is nine days old and its coins are still trading, so its row will come down. By three months out, 96.5% to 99.0% of called coins sit under $1,000 of liquidity, close to the 98.6% Solidus Labs found for pump.fun launches as a whole.
By chain
| Chain | Accounts | Calls | Reached 5x | Reached 2x |
|---|---|---|---|---|
| Solana | 16,237 | 150,715 | 9.2% | 26.0% |
| Robinhood Chain | 794 | 1,887 | 3.1% | 16.2% |
Robinhood Chain calls are scored only once a day old and when we kept watching the token for an hour, so 1,887 of 6,083 posts are in; the rest are unknown, not losses.
The highest and lowest hit rates, 20 or more scored calls
Every handle links to its own page, where each call is listed with the post.
| Handle | Calls | 5x hits | Hit rate |
|---|---|---|---|
| sol_alph | 22 | 15 | 68.2% |
| solprofithunt | 23 | 15 | 65.2% |
| realsoltrader | 51 | 30 | 58.8% |
| solanamast9 | 24 | 14 | 58.3% |
| solearlycatch24 | 27 | 15 | 55.6% |
| sol_master01 | 58 | 31 | 53.4% |
| jbstrategy | 32 | 17 | 53.1% |
| solcryptotrader | 23 | 12 | 52.2% |
| leothetrader2 | 29 | 15 | 51.7% |
| klope_1 | 62 | 30 | 48.4% |
Lowest, all with zero calls that reached 5x:
| Handle | Calls | Reached 2x |
|---|---|---|
| kentangceria_ | 21 | 0.0% |
| baak_roberto | 22 | 0.0% |
| damiencrowns | 25 | 4.0% |
| cashing_outshi | 25 | 4.0% |
| alphamemepump | 28 | 7.1% |
| lamar0985056592 | 24 | 8.3% |
| loganpierces | 24 | 8.3% |
| simonnietoo | 34 | 8.8% |
| peanutdegen | 64 | 9.4% |
| lmzjkn | 69 | 10.1% |
How a call is graded
- A call is the first post by an account that carries a token's full contract address. A ticker alone is not a call, because one ticker can mean five coins. Repeat posts of the same contract by the same account are one call.
- Solana: entry is the market cap when PumpPill first logged the token, peak is the highest cap recorded after. Entries under $5,000 and multiples over 500x are not scored (the entry could not be read); an unscored call is unknown, never a loss.
- Robinhood Chain: entry is the nearest snapshot within 30 minutes before or 60 minutes after the post, peak is the highest snapshot after, scored once the call is a day old and the token was watched for an hour.
- A scored call under 2x is a loss and stays in the file. Deleted posts stay, graded at the time. Our own accounts are excluded.
- An account is named in a table at 20 scored calls and counted in an aggregate at 5. Follower counts are the highest we recorded.
- 492 calls are dated before May 2026: old posts found by later searches, graded against data from when we first saw the coin.
Limits
- The entry is the cap at our first scan, usually minutes after the first post we saw, not necessarily this account's post.
mc_at_postfixes that only where we have a reading, 28.9% of Solana calls so far; that share grows every month. - Peak is the highest cap we recorded. A spike we missed is a missed win, which flatters nobody.
- A low hit rate is a statistical observation, not an accusation. Any account can ask for a re-check of a call at [email protected].
Use it
The file is CC BY 4.0: take it, cite it, build on it. It is rebuilt on the first of each month at /media/caller-receipts/latest/. To check one account, type the handle at pumppill.org/shillers.
The lesson is the method. The product is where you apply it.
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.