Trading signal and call groups

The research was done by AI agents that open web pages. Some sites refuse them; where that happened we say so. The small numbers point to the source list at the end. Where a figure is our own sum or guess, it is marked “our estimate” and the basis is given.

The claim

Our collection for this scheme holds only 5 tweets¹. Only one comes from an account large enough for us to name under our own rule (about 10,000 followers or more). So this page rests mainly on how the method is promoted and sold generally, and the tweets are examples. The page judges the scheme, not the people who tweeted.

“I kept telling you about $ORBIO - 865X

$PAID - 674X

$PONS - 897X

$CATE - 793X

And now I’ve found another memecoin that looks seriously overlooked. 👀

Want in?

Drop a comment and I’ll send the details. 🔔”

@yeaknoya, 23 September 2026 (date worked out from the tweet’s ID number), 1,344 views as collected¹

“865X” means the price is said to have multiplied 865 times. A “memecoin” is a crypto coin with no business behind it, traded on attention alone. Names such as “$CATE” are tickers, the short codes coins trade under; two different coins can share one.

The other four tweets come from smaller accounts, so we describe them without quoting or naming them¹:

• One says the author bought a new coin on the Solana network at a market value of about $230,000, gives the coin’s address, and says it could rise a lot.
• One says a coin rose steeply, that the author’s Telegram group profited from it, and that anyone copying the group’s trades would have had the same result.
• One offers a free group for talking about stocks and asks readers to comment for an invitation.
• One invites readers to a crypto community through a link.

What the five tweets give is multiples, one entry level and adjectives. None states a monthly income. None gives the amount of money put in, the selling price, how many calls were made in total, how many lost, the trading fees, the price of a group, or how the author is paid¹. We read the text of the tweets only; their images, videos and any threads or replies were not viewed.

What the scheme is

A “call” or “signal” is a message saying what to buy, and sometimes when to sell. A call group is a chat on Telegram or Discord (two messaging apps) where one person or a small team posts calls and members act on them. The person posting is the “caller”. “Alpha” is trading slang for information that gives an edge, so an “alpha group” is a group that says it has some.

There are three common forms:

• Free groups. Anyone can join. Three of our five tweets mention a group or community in their text; the other two give a coin call or ask readers to comment. None shows a price¹.
• Paid or “VIP” groups. A monthly fee buys the calls, or buys them earlier than free members get them.
• Copy trading. A program places the caller’s trades in the member’s account automatically. On Telegram this is done by “bots”, programs that trade from inside the chat app.

The assets vary: stocks, large crypto coins, currencies and gold, and memecoins. Three of our five tweets are about memecoins¹.

The pitch: someone skilled finds the trade, you copy it, you get the same result. The questions are whether followers make money, and how the caller is paid.

The arithmetic

No tweet in the collection states a monthly income¹. So we test two round figures, $1,000 and $10,000 a month, from the member’s side and the owner’s.

Step 1: the member’s sum

A member’s income is:

stake x number of calls x average result per call, minus trading fees, minus the subscription.

The tweets supply none of these terms. The quoted tweet does not say how its multiples were measured or whether anyone bought and sold at those prices; it gives no coin addresses, so we could not check them¹.

This table shows what a member’s whole pot must return every month, after all costs, to pay each income (our estimate: income divided by pot).

The 2.6% rows use the best measured monthly edge we found for any callers: a minority of stock tipsters, not memecoin groups (Step 3).

Step 2: what calls have returned when measured

Crypto influencers. A study in the Review of Accounting Studies covered 35,569 tweets by 180 prominent crypto influencers over two years⁷⁹. On the day of a tweet the coin rose 1.83% on average⁷⁹. The study then measured from the second day onward, and its co-author summarises: “average cumulative returns ending 10, 30, and 90 days after the tweet are -2.24%, -6.53%, and -18.90%, respectively”⁵⁷⁹. The paper estimates that $1,000 put into a coin outside the top 100 on the tweet date and held 30 days “would incur losses of $79 (7.9%)”⁷⁹. A university write-up reports the same figures⁶.

The result is not only the 2022 crypto fall: after adjusting for the whole crypto market, “longer-run market-adjusted returns are significantly negative”⁷⁹. And 44.1% of the mentions were of the ten largest coins, so the sample is not mainly memecoins⁷⁹. The authors conclude: “One can only profit from crypto-influencers’ advice by exiting the position shortly after the initial tweet”⁷⁹.

Memecoins promoted on X. CoinWire, a crypto news site, tracked 1,567 memecoins promoted by 377 X accounts with at least 10,000 followers⁷. After one week 80% of the coins had lost 70% or more. After three months 86% were down 90%⁷. It reports that “Only 3% of memecoins promoted by influencers ever achieve a 10x rise”⁷. The quoted tweet advertises multiples of 674 to 897¹; in this sample even ten times was a 1-in-33 event.

The same study has a more favourable finding. Accounts with over 200,000 followers showed “39% negative returns after one week” and “89% negative returns after three months”. Accounts with “less than 50K followers” showed “25% positive returns after one week” and “141% positive returns after three months”⁷. These are averages, so a few very large winners can lift them while most coins still fall. The study is not peer reviewed, and promotions deleted before it looked were not captured. The quoted account, at 66,507 followers, sits between the two bands¹.

Organised pump groups. A “pump” is a coordinated buy: a group agrees to buy one coin at a set moment to push the price up. A study of 355 pumps in a finance journal says: “Puzzlingly, people join in despite negative expected returns”¹⁰. A second journal study concludes: “The evidence we document, including price run-ups before P&Ds start, suggests wealth transfers from outsiders to insiders”¹¹. (“P&D” is pump-and-dump.) Its authors wrote earlier that “only investors who buy in the first 20 seconds after a P&D begins can make a profit, and they can do so only if they do not hold their tokens for very long”¹². A study of 765 pumped coins found “an average 30% relative drop in price a year after the pump event”¹³.

Two 2026 preprints (papers not yet peer reviewed) add newer data.

• One searched a year of messages from 14,499 public Telegram channels of all kinds, tested mention spikes against price data, and flagged “47 events consistent with potential pump-and-dump activity and 73 sustained market reactions”¹⁵. It puts the trading volume of the detected pumps above $200 million¹⁵. The paper does not say what share of channels took part, so the 47 is a floor, not a rate for call groups. It also found that “pump-and-dump messages are linguistically indistinguishable from organic discussions”¹⁵. So a group cannot be judged by how its messages read.
• The other shows the extreme end. It watched 83 Telegram channels, chosen because they coordinate pumps, on two other networks (BNB Smart Chain and Ethereum, not Solana). The coins were built so that members could buy but not sell. In that trap “organizers gain in 99.3% of cases, extracting over $7 million in profit”¹⁴. The 99.3% describes that trap, not ordinary pumps or call groups in general.

One call from our collection, traced. One tweet gives a coin address and an entry level, so we looked up the minute-by-minute price history¹². This is price history, not advice about the coin. We withhold the address because the account is too small for us to name.

• The tweet was posted on the afternoon (UTC) of 6 October 2026 and says the author bought at a market value of about $230,000¹.
• In the minute of the tweet the price ran from about $0.00026 to $0.00028 per coin. It had been about $0.00023 two to three minutes earlier².
• The highest price after the tweet was $0.000407, 18 minutes later. That is 77% above $0.00023, and 47% to 58% above the price in the minute of the tweet (our estimates)².
• 24 hours later the price was $0.0000426, about 81% below $0.00023 (our estimate)².
• On 10 October it was about $0.0000044, about 98% lower, and the coin’s market value showed as about $4,300²³.
• The coin’s highest price ever, $0.000899, came about 1 hour 40 minutes before the tweet².

We do not know whether the author bought when stated, or whether and when the author sold. One call is an anecdote, not a rate. A second tweet names a coin and a past rise but no entry price, so there was nothing to test⁴.

The sum. The first hour or day after a call is slightly positive on average⁷⁹⁷². A member buying by hand pays fees and gets a worse price, and those costs are larger than that edge (Step 4). For holding for days, the measured average for crypto is negative⁷⁹¹⁰. With a negative average, no stake and no number of calls produces $1,000 a month; a larger stake makes the expected loss larger (our reasoning). The one positive after-fee result we found is an automated one-hour strategy (see “The upside”)⁷².

Step 3: the exception, stock tipsters

A working paper from the Swiss Finance Institute studied people who post stock tips on StockTwits, a social network for traders. Its abstract says: “28% of finfluencers are skilled generating 2.6% monthly abnormal returns, 16% are unskilled, and 56% have negative skill”⁸. (“Finfluencer” means financial influencer. “Abnormal return” means the return above what the market gave.) The 56% trailed the market by 2.3% a month⁸. The institute’s summary says the study covered 72 million posts from over 29,000 accounts⁹.

At 2.6% a month above the market, $1,000 a month needs about $38,500 of capital and $10,000 a month needs about $385,000, before costs (our estimate: $1,000 / 0.026)⁸.

There are three catches. First, the abstract says the tipsters with negative skill “have more followers and more influence on retail trading than skilled finfluencers”⁸. A reader who picks by audience size picks badly. Second, skill was measured in hindsight; the paper’s own tested strategy is the opposite of following: “a contrarian strategy yields 1.2% monthly out-of-sample performance”⁸. That is, betting against the crowd’s tips made money on fresh data. Third, the study covers public posts about stocks, not Telegram or Discord groups. We found no comparable study for memecoin groups.

Step 4: the costs the tweets leave out

The subscription. On the sellers’ own pages on Whop, a marketplace where such groups are sold, we saw three prices: a memecoin community at $50 a month or $600 a year¹⁸, a memecoin call group at $29.99 a week or $375 for lifetime access¹⁹, and a trading community at $150 a month²¹. A 2026 round-up lists paid groups from about $30 to $500 a month; we did not confirm those on the sellers’ pages²⁹.

$29.99 a week is about $130 a month (our estimate: 29.99 x 52 / 12). Against a small pot the fee is large:

These are our estimates (fee divided by pot): what the calls must earn each month just to pay for the group.

Trading fees. Pump.fun, the site where most new Solana memecoins start, charges 1.25% on every buy and sell of the smallest coins, falling in steps as a coin grows⁶⁹. At the size of the coin we traced the fee was 1.15% (our estimate from the fee table and the coin’s market value)⁶⁹. Trojan, a Telegram trading bot, says it “charges 1% per successful trade”²⁴. A buy and a sell together cost about 4.3% to 4.5% (our estimate: 2 x 1.15% or 1.25%, plus 2 x 1%).

A worse price. A member buys after the caller, and after faster members. Trojan’s own help page says of copy trading that it “will require a higher number, especially if it’s a heavily copied wallet”²⁴. The “number” is slippage: how much worse than the quoted price you agree to accept. Trojan suggests starting at 10% to 15% in general²⁴. A 2026 paper on Pump.fun-style markets states: “Even when a copier immediately replicates the same trades as a smart money wallet, the copier necessarily trades at a worse point on the curve”⁶⁵. That is, each purchase pushes the price up, so whoever buys second pays more.

Step 5: the owner’s sum

The owner’s income does not depend on the calls being right. There are three documented routes.

Subscriptions. Whop charges “2.7% + $0.30 per successful transaction for domestic cards”²². At $50 a month the seller keeps at best about $48.35 per subscriber (our estimate: 50 - 1.35 - 0.30). Other fees stack on this: 1.5% more for international cards, up to “6% + $0.30” with every option switched on and tax collected, and payout fees such as $2.50 per bank transfer²².

These are our estimates, at the lowest fee. They ignore refunds, cancellations, tax and the cost of finding subscribers.

A share of members’ trading fees. A free group can be paid by the trading bot or exchange its members use. This is called a referral or affiliate share.

• GMGN, a Solana trading bot, charges 1% per trade and says: “Earn 10% to 30% in commissions”²⁶. Its page title says “easily earn over $8000 monthly”; that is marketing, not evidence of typical earnings²⁶. A second page explains how to add its bots to a Telegram group so that “when group members click on the message link and trade on the GMGN website or Bot, you can earn trading commissions”²⁷.
• Trojan’s referral page lists a “Total Rev Share” of 27.5% to 47.5% of fees, paid on “both direct and indirect referrals” across five layers²⁵. Layers are people referred by people you referred. The share on a member you referred yourself is lower, and the page does not give it.
• Bitget, a crypto exchange, offers affiliates: “Start with a 40% rebate ratio, and earn up to 50%”²⁸. A rebate ratio is the part of the member’s fee paid back to the referrer.

With a 1% fee, each $1,000 a month to the owner needs this much member trading (our estimates: $1,000 / (1% x share)):

The Trojan rows are minimums, because the direct share is lower than the five-layer total. The owner earns this whether members win or lose. We did not check whether any account in our collection uses a referral link.

Telegram’s own paid channels. Since August 2024 Telegram has offered “invite links that allow joining a channel in exchange for a monthly payment in Telegram Stars”, its in-app currency³¹. Its creator terms value a Star at “an equivalent of 0.013 USD worth of rewards”, so $1,000 a month is about 76,900 Stars (our estimate, assuming subscriptions pay the same rate as paid posts)³⁰.

Step 6: how many owners reach it

One large store. The largest memecoin community we found on Whop calls itself “the #1 online crypto platform where over 110,000 members collaborate to master blockchain trading” and shows “124,417 joined”¹⁸. We treat it as a call group because its products are named “Potion Memes Access” and one reviewer writes “Huge wins called daily”¹⁸. Its products page lists four paid products, each with a price and a number beside it: monthly access at “$50.00 / month” (104), the same paid in crypto at “$50.00” (45), one-year access at “$600.00 / year” (23) and the same paid in crypto at “$600.00” (1)¹⁸. That is 173 in all, about 0.14% of those who joined (our estimate). Whop does not say what the numbers mean. If every one is a current payer, the two recurring products bring in about $6,350 a month before fees (our estimate: 104 x $50 + 23 x $600 / 12). If the crypto-paid buyers also renew, the figure is about $8,650 (our estimate). Sales elsewhere are not visible, so the real figure could be much higher or lower. The store shows a rating of 4.8 from 392 reviews¹⁸. Reviews on a seller’s page measure satisfaction, not trading results.

The whole marketplace. Whoptrends, a third-party site that estimates sales from Whop’s public pages, wrote in February 2026 that 87.8% of Whop products earn nothing and that “Only 3,841 products (2%) earn more than $1,000/month”²³. Of 21,079 products in its “Trading” group, 4,282 show any revenue: about 20% (our sum from the site’s two tables)²³. So about 80% earn nothing. Among the 4,282, the median is $333 a month and the average $4,647²³. “Median” is the middle product: half earn more, half less. The site says of its own numbers: “these are estimates, not verified financial data”²³. Its tier table does not add up exactly (the rows above $1,000 sum to 3,941, not 3,841), and “Trading” includes courses and tools as well as call groups.

What people who tried it report

No source gives the figure we wanted most: how many members of call groups end up ahead. We found no study, no platform disclosure and no dataset. Some platforms publish what signal providers earned (see “The upside”), but we found no confirmed record of what their followers earned. Every member profit claim we met came from a seller’s own feed or from a site that earns from referrals.

One hands-on test. A trading education site says it joined 7 groups (5 on Discord, 2 on Telegram, free and paid) for four weeks and followed 112 US stock signals exactly as posted³². A “win rate” is the share of calls that made money. The site reports: “54% of losing trades were never acknowledged, 9% were deleted, while 100% of winners were highlighted – creating a ‘public’ win rate that looks far better (46% in our log vs 62% after deleted losses are excluded)”³². It says several groups advertised above 70%, and that 68% of signals were posted after the move had begun³². Signals marked “urgent” lost on average 0.42 times the amount risked per trade; the rest averaged a gain of 0.07 times³². The test is small and self-reported, names no groups and gives no result in money. The site earns from broker referrals. A win rate is also not a return: it does not say how big the losses were or whether a member could have bought at the call price.

Sellers’ own figures. One Whop seller’s page advertises a “60% WIN RATE (ALL CALLS OVER 4x)” for one week²⁰. For a coin it calls Cate, one post claims a “582X CALLOUT” and a later post “1000X”²⁰. None of these figures is audited, and the page does not show what any member bought or sold.

Memecoin traders in general. These figures cover all memecoin traders, not group members, and come from news reports of public dashboards that we did not open.

• Of 13.4 million wallets on Pump.fun, “only 55,012 wallets … have realized $10,000 in profit or more”, about 0.4%, and 294 had realised over $1 million, by January 2025³³. (A wallet is a crypto account; one person can have several. “Realised” means the coins were sold, so the profit is no longer on paper.) Pump.fun’s co-founder disputed the count, saying the true number of profitable wallets is “an order of magnitude LARGER” because most gains come after a coin leaves the launch stage³³.
• An August 2024 report said “more than 60% have lost money” on Pump.fun³⁴. An August 2026 report said that over 90 days 6.25% of memecoin-trading wallets were in profit, and “Of all traders in the green, 88% made under $100”³⁵.

The coins. Solidus Labs, which sells market-monitoring software, found that 98.6% of coins created on Pump.fun from January 2024 to March 2025 fell below $1,000 of liquidity, meaning there was almost no money left to sell into³⁶. Chainalysis, a blockchain analysis firm, used a much stricter test on other blockchains and found “3.59% of all launched tokens in 2024 display patterns that may be linked to pump-and-dump schemes”³⁷. Both figures describe coins, not groups.

Leveraged trading. Many currency and gold signal groups trade CFDs, bets on price moves made with borrowed money. In a March 2018 notice the EU markets regulator said national studies show “74-89% of retail accounts typically lose money on their investments” in CFDs³⁸. That is the loss rate for the activity, not for signal followers.

What the rules allow now

For a member. Joining a group and reading someone’s trade ideas is legal for the member almost everywhere, on the sources we read. None of the three platforms bans call groups as such.

• Telegram’s terms forbid using it to “send spam or scam users” and say nothing specific about calls or signals⁴³. It reports over 27 million groups and channels blocked in 2026 for all causes, with no figure for financial groups⁴⁴.
• Discord bans “market manipulation (including “pump-and-dump” schemes)” and says it “may take action against accounts that are present in violative servers”⁴⁵. So a member can lose an account for being in the wrong server. Its rules for paid features do not say whether a paid signal server is allowed⁴⁶.
• X’s policy on financial deception does not mention calls or signal groups. The live page refused our agents, so we read a 2024 archived copy⁴⁷. News reports say X suspended over 20 crypto accounts in June 2025 without explanation⁴⁸.

A group existing on a platform does not mean anyone has checked it.

Coordinated buying. Where the asset is a listed share, agreeing to buy together to move the price is a crime. Australia’s regulator reported in December 2025 that “Four co-conspirators who used Telegram app group chats to pump up the share prices of Australian stocks before dumping them at inflated prices were convicted and sentenced”, to terms of 14 months to 2 years served in the community⁵³. The US futures regulator says of crypto pump groups that “market manipulation is against the law and many participants end up losing money”⁵⁸.

Calling and then selling. The US securities regulator (the SEC) names two practices: “Scalping–recommending a stock to drive up the share price and then selling shares of the stock at inflated prices to generate profits”, and touting, which is promoting without disclosing payment⁵⁷. The main US case on this is undecided; see the next section.

Selling advice without a licence. This is a risk for the person running a group.

• Hong Kong: a man who ran a paid Telegram group at US$200 a month giving stock recommendations “was sentenced to six weeks’ imprisonment” in November 2025 and intended to appeal. We read a republished copy of the regulator’s release and did not check the outcome of the appeal⁵².
• Australia: unlicensed advice carries “up to five years’ imprisonment for an individual”⁵⁴.
• UK: in a week of action with eight other regulators in June 2025, the UK regulator made 3 arrests and issued 50 warning alerts that it said would lead to over 650 takedown requests⁵⁵. Its enforcement director said of finfluencers: “They must act responsibly and only promote financial products where they are authorised to do so – or face the consequences”⁵⁵.
• EU: a news report cites “fines of up to €5m, or criminal sanctions” under market abuse rules; we could not find the regulator’s own page⁵⁶.
• US: in an agreed (consent) order in 2023, a court required a service that sold signals and automated trading for binary options from 2013 to 2017, and its owner, to pay $100,000 for operating without registration as an adviser on futures-type contracts⁵⁹.

Each case concerns only the companies and people named in it.

Memecoins in the US. SEC staff said in February 2025 that typical meme coins are not securities, and “Accordingly, neither meme coin purchasers nor holders are protected by the federal securities laws”⁵¹. This is a staff view, not a ruling. So the follower of memecoin calls has fewer remedies than the follower of stock calls.

Other risks to the member’s money.

• Payments in Telegram Stars: “All sales of Telegram Stars are final and Telegram does not issue refunds”⁴².
• Trading bots hold your keys or your permission to spend. A security firm reported that users of two Telegram bots lost about $1.1 million in October 2023 through a flaw in that permission⁶⁰. News reports describe two further cases in 2024, one of about $523,000 taken from more than 300 users⁶¹⁶².
• Tax: in the US, “If you have digital asset transactions, you must report them whether or not they result in a taxable gain or loss”⁶³. Dozens of calls mean dozens of reportable sales.

Groups offered by strangers. US regulators warn about one pattern. FINRA, the US brokerage regulator, describes groups advertised on social media that move people into private chats and tell them which stock to buy, until “the investors become unable to sell”⁵⁰. The FBI wrote in July 2025 that it “has seen at least a 300 percent increase in victim complaints referencing ramp-and-dump stock fraud from 2024”⁴⁹. Its 2025 annual report counts about 1,600 complaints and $160 million in reported losses under “Investment Club Scams”³⁹. The US futures regulator says of testimonials in unsolicited group chats: “It’s all fake, lies designed to steal your money”⁴⁰. These describe a pattern. They are not statements about any account in our collection.

Who makes money from it

The group owner, from fees and subscriptions. This is the best documented flow. The bot takes 1% per trade²⁴²⁶. GMGN pays 10% to 30% of that to whoever referred the trader; Trojan pays a total of 27.5% to 47.5% spread over five layers of referrers²⁵²⁶. On a $1,000 buy and sell through GMGN, the member pays about $20 in bot fees and the referrer receives $2 to $6 (our estimate). Subscriptions come on top (Step 5). Most owners still earn little: about 80% of trading products tracked on Whop show no revenue (Step 6)²³.

The platforms. The bot keeps the rest of each fee, and Whop and Telegram take a share of each payment^24262231.

The caller, by buying first. In organised pump groups the pattern is measured. A study of Telegram pumps estimated “that admins made a net profit of 199.52 BTC, equivalent to 1.1 million USD, through 348 pump and dump events” and that “The estimated return of insiders averages 18%”¹⁶. The authors of another study wrote that “some pump group organizers offer premium memberships to allow some investors to receive pump signals before others do”¹². On copy-trading exchanges, an exchange’s blog says 97% of lead traders profit personally while fewer than half of their followers do; we could not open the study behind it⁷⁷.

For stocks, the largest example is alleged, not proven. In December 2022 the SEC charged eight people: seven who, it says, promoted stocks to followers on Twitter and Discord and “regularly sold their shares without ever having disclosed their plans to dump the securities”, and an eighth it says aided them through a podcast⁴¹. The SEC put the gain at about $100 million; a criminal indictment put it at about $114 million⁴¹⁶⁴. The Department of Justice said the group was “promoted as one of the largest, free online communities in the world for individual stock traders”⁶⁴. Its case page says: “All of the defendants have pleaded not guilty and denied these charges”⁶⁶. A district judge dismissed the indictment in March 2024; a summary by a law firm acting for the defence says the judge found the alleged scheme deprived investors of accurate information, not of money or property⁶⁷. On 2 October 2025 an appeals court ruled: “we REVERSE the district court’s dismissal of the indictment”¹⁷. The case page, updated 5 August 2026, lists a trial of seven defendants for 3 May 2027 and a guilty plea by the eighth in March 2023; an older entry under “Previous Updates” records the dismissal that was then reversed⁶⁶. The case is pending and concerns only the people named in it. The group was free to join, so the case, whatever its outcome, is about a free group, not a paid one.

The caller, from paid promotion. CoinWire estimated, using a third-party earnings calculator and not payment records, that an account in its sample could earn about $399 per promotional tweet⁷. In a 2022 settled case over one celebrity Instagram post about a crypto token, the SEC’s order found the celebrity was paid $250,000 and did not disclose it; she agreed to pay $1.26 million “Without admitting or denying the SEC’s findings”⁶⁸. The case is about an advert, not a call group; it shows that US law requires paid promotion of securities to be disclosed. A news report carries a private investigator’s allegation, not a regulator’s finding, that fewer than five of roughly 160 crypto influencers who accepted one promotion deal labelled their posts as adverts⁷⁰.

Sellers of alerts. In a US Federal Trade Commission case that ended in a settlement, a paid stock-alert company agreed to pay $2.425 million⁷¹. The FTC’s complaint alleged that “consumers who purchased the site’s services lost millions of dollars in their investments”⁷¹. A settlement is not a court finding, and the case concerns only that company.

We did not establish how any account in our collection is paid.

The upside

Calls are not pure noise, and some people do make money around them.

An automated strategy that beat the market after fees. A peer-reviewed study took 28,700 crypto buy and sell signals posted on X, Reddit, StockTwits and Telegram⁷². In the hour of a buy signal it found “a statistically significant positive abnormal return of 0.003497 … and 55.83% of signals resulting in positive returns”⁷². That is 0.35%. For small coins the figure was 0.67%, and it reversed slightly in the next hour⁷². The authors then tested a simple program on 747 later buy signals for small, recently weak coins, from 1 July 2023 to 15 February 2024. It bought at the start of the signal hour, sold one hour later and paid “a 0.3% transaction fee per trade”⁷². It returned 115.23%, against 67.72% for an index of large crypto coins and 11.80% for the US stock index⁷².

The limits: it is a calculation on past data, not money traded, and it dropped the best and worst 3% of trades. Only 27.97% of its trades won, and the authors say its performance “is driven by a few highly profitable trades”⁷². The period was a rising market. The authors call it “only a baseline example” and add: “For retail investors, however, the short time frame in which these signals are predictive may limit their utility”⁷². The 0.3% fee is an exchange fee for large coins, far below the 4.3% to 4.5% round trip on Solana memecoins (Step 4).

A skilled minority of stock tipsters. 28% of the StockTwits accounts studied were skilled, at 2.6% a month above the market⁸. This is the strongest evidence that following the right person can pay, though the figure is before costs and skill was identified in hindsight (Step 3).

Smaller accounts did better. The skilled stock tipsters had fewer followers⁸. CoinWire’s memecoin promotions from accounts under 50,000 followers averaged a gain at three months, where those from accounts over 200,000 averaged a loss, with the caveats in Step 2⁷.

A community that worked, then stopped. A study in the Review of Financial Studies looked at written research posts (“due diligence”) on Reddit’s WallStreetBets stock forum⁸¹. Its abstract says: “Before the Gamestop (GME) short squeeze, recommendations are significant predictors of returns and cash-flow news. This predictability is eliminated post-GME”⁸¹. After the squeeze, the share of posts built on hype rose “dramatically”, and the loss of value was concentrated in those posts⁸¹. So reasoned research in a forum not yet full of promotion had measurable value, and lost it when the hype grew.

Callers with public profits. Kolscan, a site that ties Solana wallets to X handles, runs a live leaderboard. On its default “Daily” tab at about 12:15 UTC on 10 October 2026, the top wallet was up 116.02 SOL (about $12,715) and the fiftieth up 10.83 SOL (about $1,187)⁷³. (SOL is the currency of the Solana network. The board changes through the day; an earlier reading showed about 120 SOL at the top.) Several top wallets had more losing trades than winning ones, for example 24 wins against 45 losses⁷³. It lists winners only, and shows the caller’s wallet, not the result of people who bought afterwards. A firm that sells tracking of such wallets says of those it scored: “Only 118 wallets — roughly 1 in 4 — hold a win rate above 50%”⁷⁴.

The largest trader outcomes. 294 Pump.fun wallets had realised over $1 million by January 2025, per a news report of public data³³. They are traders, not group members, and the figure is only as good as the dashboard.

Owners at the top. Whoptrends estimates that trading products on Whop bring in $19.9 million a month in total, 31% of the platform²³. Its top-ten list includes one trading product at an estimated $419,000 a month with 331 members, and its site lists a crypto trading product with about 11,300 members in its “$100K+ /mo” band²³. These are third-party estimates from public member counts and prices, not verified income. Beside them: the median trading product with any revenue is estimated at $333 a month, and about 80% show none²³. The large store in Step 6 appears to gross about $6,350 to $8,650 a month from its listed products (our estimate, uncertain)¹⁸. We found no group owner with a published, payment-verified income.

Signal marketplaces with public records. This is the form of the scheme a reader can check before paying. MQL5, the marketplace built into the MetaTrader trading program, advertises “Over 8,000 signals for copying”⁸². The signals we saw listed cost 30 to 35 USD a month, and each listing shows growth, the share of winning trades, the age of the account and the number of subscribers⁸². The site says: “Monitored accounts are provided with a detailed statistics and trading history”, and also: “Historical statistics cannot guarantee any profitability in the future”⁸². The page did not say how accounts are verified. Listings show the survivors, most are leveraged currency or gold accounts where the 74-89% loss rate applies³⁸, and follower results are not published.

eToro, a regulated broker, lets customers copy named traders with public records; the minimum is $200⁷⁵. The five traders it featured ranged from -9.49% to +54.89% over two years, with a median of about +32%⁷⁵. The platform chose them for its own sales page. The page’s warning that “51% of retail investor accounts lose money when trading CFDs with this provider” is about CFD accounts, not people copying traders⁷⁵.

Independent evidence on such platforms is less kind. A 2018 journal study of social trading platforms found that “simply investing in those traders with the highest accumulated returns leads to high losses”, that “no strategy reveals a positive abnormal return after transaction costs”, and that “there is no wisdom-of-the-crowd effect”: more followers did not go with better returns⁸⁰. The abstract does not name the platforms or the sample size. An exchange’s blog gives the share of copy-trading followers in profit at three crypto exchanges as 66.50%, 43.65% and 57.79%, and a “Blended Average” of about 48.48% that we could not reproduce from those three; we could not open the study behind it⁷⁷.

A stock-pick newsletter at $199 a year states its picks have returned 951% since 2002 against 215% for the US stock index; that is the seller’s own unaudited average, not what a new subscriber earns⁷⁶.

A realistic good result (our estimate). No source measures what members earn, so this is inference from the evidence above.

• Following memecoin or crypto calls by hand: a small loss to roughly level before the subscription, with occasional large wins that do not repeat. The basis is a first-hour edge of 0.35% to 0.67% against Solana fees of about 4.3% to 4.5%, and negative averages after the first day⁷²⁷⁹⁶⁹.
• Running a program that trades signals within the hour on an exchange with low fees: the one tested example beat the market over seven and a half months on past data⁷². This is a software project, not a group membership.
• Copying a provider with a multi-year public record: possible gains, but the one study we found says choosing by past return loses and no strategy beat the market after costs⁸⁰.
• Running a group: about four in five trading products on Whop show no revenue, and the median one that sells is estimated at $333 a month²³.

The best case, not a realistic expectation: following one of the skilled 28% of stock tipsters with $40,000 would make about $1,000 a month above the market before costs⁸. The paper offers no way to pick them in advance, and the more likely draw is one of the 56% who did worse than the market⁸.

What it takes to compete

Being the caller, or early. The measured profits go to organisers and to those who buy first¹⁶¹². The tracking firm above reports that the median market value at a tracked caller’s first buy in August 2026 was $4,654, and the next tracked caller bought a median of 9 seconds later⁷⁸. One tweet in our collection states an entry at about $230,000¹. A person buying by hand after a post is not in the same trade.

Speed and automation. The one edge found after fees came from a program trading within the hour⁷². The 2026 paper on copying Pump.fun wallets found copier returns “remaining negative despite accurate wallet identification” for its simpler models, and “an average copier return of 3% per meme coin investment” only for its most elaborate system⁶⁵. This is coding and data work.

A way to pick callers that is not audience size. Skilled stock tipsters had fewer followers, and smaller memecoin promoters did better on average⁸⁷. On copy platforms, follower counts did not go with returns⁸⁰. And a group cannot be judged by how its messages read¹⁵. What is left is a record you can check yourself: a public wallet, or a platform that publishes every trade including the losers.

Calls that come with reasons, in a group not being marketed for growth. That is the condition under which the WallStreetBets research posts had value, and the one that had gone when the value went⁸¹.

Capital. At the best measured rate, $1,000 a month needs about $38,500 (Step 3). A $500 pot cannot carry a $50 subscription (Step 4).

For owners: an audience. The large store in Step 6 shows 124,417 joined against 173 on paid products¹⁸. As an illustration from that one store, not a rule, 21 paying subscribers on the same ratio would need about 15,000 people joined (our estimate: 21 / 0.0014). Owners also carry the licence and disclosure risks above.

How to tell early which side you are on.

• Write down every call the group makes for a month, at the price you could have bought at, not the price in the post. Compare your log with what the group celebrates. The hands-on test found over half of losing calls were never mentioned³².
• Look at the chart for the hour before each call. If the price had already run, you are being called in late. The coin we traced peaked before the tweet².
• Ask how the caller is paid. If the answer is a referral link, the caller earns on your trading volume, not your profit²⁶.
• Add up the subscription and fees as a share of your pot (Step 4). If it is more than a few percent a month, the calls must be exceptional to cover it.
• If the group tells everyone to buy at the same moment, the studies above say who that pays¹²¹⁶.

What we did not verify

• How many call group members make or lose money. No study or dataset was found. The “realistic good result” is our inference.
• The multiples in the quoted tweet. No coin addresses are given and tickers are not unique¹. We do not know whether @yeaknoya runs or promotes a group; the tweet does not mention one.
• For the traced coin: whether the author bought as stated, and whether or when the author sold.
• How any account in the collection is paid, whether its group charges, and whether it holds the coins it mentions. We did not open the groups or the X profiles. The tweets’ images, videos, threads and replies were not viewed.
• The full text of most academic papers. For⁸,¹⁰,¹¹,¹³,¹⁴,¹⁵,⁸⁰ and⁸¹ we read abstracts; for⁹ and¹², institute or authors’ summaries.
• The dashboards behind the wallet figures³³³⁴³⁵. They are news reports from different periods and should not be added together.
• What Whop’s “joined” count and the numbers beside its products mean, so the 0.14% and the $6,350 to $8,650 figures are rough¹⁸. Whether the $150 community posts calls; its page says only “Learn to trade”²¹.
• Discord’s cut of server subscriptions, what a buyer pays per Telegram Star, and whether subscription Stars pay $0.013.
• How Trojan’s referral share splits across its five layers²⁵.
• The outcome of the Hong Kong appeal⁵², the state of the SEC’s civil case against the eight US defendants, and the Department of Justice press release, which we did not re-open today⁶⁴.
• The current wording of X’s policy; the live page refused both routes⁴⁷.
• Prices in the paid-group round-up²⁹, the EU warning at source⁵⁶, and the study behind the exchange blog’s copy-trading figures⁷⁷.
• How MQL5 verifies the accounts it lists, and what its subscribers earned⁸².
• Quotes from pages read through the reader service r.jina.ai, or a tool that summarises pages, and not re-read today:^{19244150747678}.
• Taxes outside the US, and the cost of advertising to fill a group.

Sources

1. Does It Pay collection of 5 tweets on this scheme, collected 2026-10-06. Tweet dates worked out from tweet IDs. Read from the project’s stored file on 2026-10-10.
2. GeckoTerminal API (api.geckoterminal.com), one-minute and hourly price history for the traced coin’s trading pool. The pool address is withheld because it identifies an account below our naming threshold; it is available on request. Data for 6 to 10 October 2026. Read at source on 2026-10-10. Percentages are our estimates.
3. DexScreener API (api.dexscreener.com), token data for the traced coin. Address withheld as in source 2. 10 October 2026. Read at source on 2026-10-10.
4. DexScreener API (api.dexscreener.com), token data for the coin named in a second tweet. Address withheld as in source 2. 10 October 2026. Read at source on 2026-10-10.
5. Joseph Pacelli, article on ProMarket about the paper “Crypto-influencers” (see source 79). https://www.promarket.org/?p=36952 . 17 April 2023. Co-author’s article read at source on 2026-10-10 through r.jina.ai.
6. Futurity, university write-up of the same paper. https://www.futurity.org/crypto-influencer-advice-investment-3260122 . 5 December 2024. Secondary.
7. CoinWire, study of memecoins promoted by X accounts. https://coinwire.com/76-percent-twitter-influencers-promote-dead-memecoins-study/ . 20 November 2024. Read at source on 2026-10-10 through r.jina.ai. The $399 figure is the source’s own estimate.
8. Swiss Finance Institute, Research Paper 23-30, “Finfluencers” (Kakhbod, Kazempour, Livdan, Schuerhoff), abstract. https://www.sfi.ch/en/publications/n-23-30-finfluencers . 25 April 2023. Read at source on 2026-10-10.
9. Swiss Finance Institute, “Finfluencers under the microscope: advice vs. entertainment”. https://www.sfi.ch/en/about-us/news/finfluencers-under-the-microscope-advice-vs.-entertainment . No date seen on the page. Read at source on 2026-10-10 through r.jina.ai.
10. Dhawan and Putnins, “A New Wolf in Town? Pump-and-Dump Manipulation in Cryptocurrency Markets”, Review of Finance 27(3), abstract. https://ideas.repec.org/a/oup/revfin/v27y2023i3p935-975..html . 2023. Abstract read at source on 2026-10-10 through r.jina.ai.
11. Li, Shin and Wang, “Cryptocurrency Pump-and-Dump Schemes”, Journal of Financial and Quantitative Analysis, abstract. https://www.cambridge.org/core/journals/journal-of-financial-and-quantitative-analysis/article/abs/cryptocurrency-pumpanddump-schemes/97149DCD519BAF838F269F98DC76D682 . 4 April 2025. Abstract read at source on 2026-10-10.
12. Li, Shin and Wang, authors’ summary on the Columbia Law School blog. https://clsbluesky.law.columbia.edu/2019/01/07/cryptocurrency-pump-and-dump-schemes/ . 7 January 2019. Read at source on 2026-10-10 through r.jina.ai.
13. arXiv 2309.06608, long-run study of 765 pumped coins, abstract. https://arxiv.org/abs/2309.06608 . 12 September 2023. Abstract read at source on 2026-10-10.
14. arXiv 2610.10149, “Pump-and-Dump meets Honeypot Tokens”, study of 83 Telegram channels, abstract (preprint). https://arxiv.org/abs/2610.10149 . 7 October 2026. Abstract read at source on 2026-10-10.
15. arXiv 2609.01176, study of 14,499 public Telegram channels, abstract (preprint). https://arxiv.org/abs/2609.01176 . 1 September 2026. Abstract read at source on 2026-10-10.
16. Xu and Livshits, “The Anatomy of a Cryptocurrency Pump-and-Dump Scheme” (USENIX Security 2019). https://arxiv.org/abs/1811.10109 . 25 November 2018, revised 17 August 2019. Read at source on 2026-10-10.
17. US Court of Appeals for the Fifth Circuit, United States v. Constantinescu, No. 24-20143, opinion. https://www.ca5.uscourts.gov/opinions/pub/24/24-20143-CV0.pdf . 2 October 2025. Read at source on 2026-10-10.
18. Whop, Potion Alpha products, store and reviews pages. https://whop.com/potion-alpha/products , https://whop.com/potion-alpha and https://whop.com/potion-alpha/reviews . As served 10 October 2026. Read at source on 2026-10-10 through r.jina.ai. Revenue figures are our estimates.
19. Whop, ZYG VIP Insider Access products. https://whop.com/vip-discord-access-be/products . As served 10 October 2026. Read at source on 2026-10-10 through r.jina.ai.
20. Whop, ZYG store page. https://whop.com/vip-discord-access-be . Posts dated August 2026. Read at source on 2026-10-10 through r.jina.ai.
21. Whop, Triton Trades products. https://whop.com/tritontrades/products . As served 10 October 2026. Read at source on 2026-10-10 through r.jina.ai.
22. Whop, “Fees”. https://docs.whop.com/fees . Undated. Read at source on 2026-10-10 through r.jina.ai.
23. Whoptrends, “Whop creator earnings data 2026”. https://whoptrends.com/blog/whop-creator-earnings-data-2026 . 15 February 2026. Read at source on 2026-10-10. Third-party estimates.
24. Trojan, FAQs. https://docs.trojan.com/faqs.md . Undated. Read at source on 2026-10-10.
25. Trojan, referral system. https://docs.trojan.com/trojan-arena/referral-system.md . Undated. Read at source on 2026-10-10.
26. GMGN, referral page and fee settings. https://docs.gmgn.ai/index/cooperation-referral-refer-friends-to-earn-rebate-30-rebates-easily-earn-over-usd8000-monthly and https://docs.gmgn.ai/index/gmgn-fees-settings . Undated. Read at source on 2026-10-10.
27. GMGN, “Invite bots to Telegram group to earn trading commissions automatically”. https://docs.gmgn.ai/index/cooperation-referral-invite-bots-to-telegram-group-to-earn-trading-commissions-automatically . Undated. Read at source on 2026-10-10.
28. Bitget, affiliates page. https://www.bitget.com/affiliates . Undated. Read at source on 2026-10-10.
29. Plisio, round-up of crypto signal groups. https://plisio.net/crypto/best-crypto-signals . 22 April 2026. Secondary.
30. Telegram, content creator rewards terms. https://telegram.org/tos/content-creator-rewards . Undated. Read at source on 2026-10-10 through r.jina.ai.
31. Telegram blog, “Superchannels, Star Reactions and Subscriptions”. https://telegram.org/blog/superchannels-star-reactions-subscriptions . 14 August 2024. Read at source on 2026-10-10.
32. DayTrading.com, test of 7 signal groups. https://www.daytrading.com/about-us/media/signals-groups . 16 January 2026. Read at source on 2026-10-10 through r.jina.ai. Self-reported.
33. Decrypt, report of Pump.fun wallet data. https://decrypt.co/300403/pump-fun-traders-millionaires . 10 January 2025. Secondary.
34. The Defiant, report of Pump.fun trader data. https://thedefiant.io/-everyone-but-traders-make-money-from-memecoins . 21 August 2024. Secondary.
35. Cryptopolitan, report of Solana memecoin wallet data. https://www.cryptopolitan.com/6-solana-meme-traders-profit-in-90-days/ . 19 August 2026. Secondary.
36. Solidus Labs, report on Solana tokens. https://www.soliduslabs.com/reports/solana-rug-pulls-pump-dumps-crypto-compliance . About May 2025 (no date on the page). Read at source on 2026-10-10 through r.jina.ai.
37. Chainalysis, market manipulation report. https://www.chainalysis.com/blog/crypto-market-manipulation-wash-trading-pump-and-dump-2025/ . 29 January 2025. Read at source on 2026-10-10 through r.jina.ai.
38. European Securities and Markets Authority, notice on CFDs and binary options. https://www.esma.europa.eu/press-news/esma-news/esma-agrees-prohibit-binary-options-and-restrict-cfds-protect-retail-investors . March 2018 (measures agreed 23 March 2018). Read at source on 2026-10-10 through r.jina.ai.
39. FBI Internet Crime Complaint Center, 2025 annual report. https://www.ic3.gov/AnnualReport/Reports/2025_IC3Report.pdf . Published 2026. Read at source on 2026-10-10.
40. Commodity Futures Trading Commission, release 9005-24. https://www.cftc.gov/PressRoom/PressReleases/9005-24 . 31 October 2024. Read at source on 2026-10-10 through r.jina.ai.
41. Securities and Exchange Commission, press release 2022-221. https://www.sec.gov/newsroom/press-releases/2022-221 . 14 December 2022. Read at source on 2026-10-10.
42. Telegram, Stars terms. https://telegram.org/tos/stars . Undated. Read at source on 2026-10-10.
43. Telegram, terms of service. https://telegram.org/tos . Undated. Read at source on 2026-10-10.
44. Telegram, moderation page. https://telegram.org/moderation . Live counter. Read at source on 2026-10-10 through r.jina.ai.
45. Discord, deceptive practices policy explainer. https://discord.com/safety/deceptive-practices-policy-explainer . Undated; guidelines effective 29 September 2025. Read at source on 2026-10-10 through r.jina.ai.
46. Discord, monetization policy. https://support.discord.com/hc/en-us/articles/10575066024983-Monetization-Policy . Last updated 1 March 2024. Read at source on 2026-10-10 through r.jina.ai.
47. X, financial scam policy, archived copy. http://web.archive.org/web/20240423203654/https://help.twitter.com/en/rules-and-policies/financial-scam . Policy dated September 2019; copy of 23 April 2024. Secondary (the live page refused our agents).
48. Cointelegraph, report of X account suspensions. https://cointelegraph.com/news/pump-fun-x-account-suspended-crypto-account-blitz . 17 June 2025. Secondary.
49. FBI, public service announcement I-070325. https://www.ic3.gov/PSA/2025/PSA250703 . 3 July 2025. Read at source on 2026-10-10 through r.jina.ai.
50. FINRA, “Investment group imposter scams”. https://www.finra.org/investors/insights/investment-group-imposter-scams . 9 December 2025. Read at source on 2026-10-10 through r.jina.ai.
51. Securities and Exchange Commission, staff statement on meme coins. https://www.sec.gov/newsroom/speeches-statements/staff-statement-meme-coins . 27 February 2025. Read at source on 2026-10-10.
52. Mondo Visione, republished Hong Kong Securities and Futures Commission release. https://mondovisione.com/media-and-resources/news/hong-kong-securities-and-futures-commission-secures-first-custodial-sentence-aga-2025117/ . 7 November 2025. Secondary.
53. Australian Securities and Investments Commission, media release 25-315MR. https://asic.gov.au/about-asic/news-centre/find-a-media-release/2025-releases/25-315mr-market-riggers-sentenced-in-asx-pump-and-dump-case/ . 22 December 2025. Read at source on 2026-10-10 through r.jina.ai.
54. Australian Securities and Investments Commission, Information Sheet 269. https://asic.gov.au/regulatory-resources/financial-services/giving-financial-product-advice/discussing-financial-products-and-services-online/ . Date not captured. Read at source on 2026-10-10.
55. Financial Conduct Authority, press release on finfluencers. https://www.fca.org.uk/news/press-releases/fca-leads-international-crackdown-illegal-finfluencers . 6 June 2025, updated 20 February 2026. Read at source on 2026-10-10.
56. Euronews, report of an EU regulator’s warning. https://www.euronews.com/business/2024/02/06/finfluencers-face-prosecution-for-misleading-investment-advice-says-eu-watchdog . 6 February 2024. Secondary.
57. Securities and Exchange Commission, investor alert on social media. https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-alerts/social-media-and-investment-fraud-investor-alert . 29 August 2022. Read at source on 2026-10-10.
58. Commodity Futures Trading Commission, customer advisory on pump-and-dump groups. https://www.cftc.gov/LearnAndProtect/AdvisoriesAndArticles/beware_virtual_currency_pump_dump.html . No date on the page as fetched. Read at source on 2026-10-10.
59. Commodity Futures Trading Commission, release 8770-23. https://www.cftc.gov/PressRoom/PressReleases/8770-23 . 29 August 2023. Read at source on 2026-10-10.
60. CertiK, report on two Telegram bots. https://www.certik.com/blog/maestro-and-unibot . 2 November 2023. Read at source on 2026-10-10.
61. Decrypt, report on a Solana Telegram bot. https://decrypt.co/224371/solana-telegram-trading-bot-shut-down-users-drained-523k . 1 April 2024. Secondary.
62. The Block, report on a Telegram bot. https://www.theblock.co/post/317282/banana-gun-telegram-bot . 19 September 2024. Secondary.
63. Internal Revenue Service, “Digital assets”. https://www.irs.gov/filing/digital-assets . Undated. Read at source on 2026-10-10.
64. Department of Justice, press release on the indictment. https://www.justice.gov/opa/pr/eight-men-indicted-114-million-securities-fraud-scheme-orchestrated-through-social-media . 14 December 2022. Read at source on 2026-10-10 through r.jina.ai.
65. arXiv 2601.08641, paper on copying wallets in Pump.fun-style markets (preprint). https://arxiv.org/abs/2601.08641 . 13 January 2026. Read at source on 2026-10-10.
66. Department of Justice, case page for United States v. Constantinescu et al. https://www.justice.gov/criminal/criminal-vns/case/united-states-v-constantinescu-et-al . Updated 5 August 2026. Read at source on 2026-10-10.
67. Dynamis LLP (a law firm acting for the defence), summary of the March 2024 dismissal. https://www.dynamisllp.com/media/114m-discord-case-tossed . 21 March 2024. Secondary.
68. Securities and Exchange Commission, press release 2022-183. https://www.sec.gov/newsroom/press-releases/2022-183 . 3 October 2022. Read at source on 2026-10-10.
69. Pump.fun, “Fees”. https://pump.fun/docs/fees . Page last updated 8 October 2026. Read at source on 2026-10-10 through r.jina.ai.
70. The Block, report on undisclosed paid promotion. https://www.theblock.co/post/368956/zachxbt-says-over-100-crypto-influencers-accepted-promo-deals-without-disclosing-paid-ads . 1 September 2025. Secondary.
71. Federal Trade Commission, refund release in the Raging Bull case. https://www.ftc.gov/node/80608 . 6 March 2023. Read at source on 2026-10-10 through r.jina.ai.
72. Electronic Markets (Springer), “Wisdom of the crowd signals”, study of 28,700 crypto trading signals, with its Table 9. https://link.springer.com/article/10.1007/s12525-025-00815-6 and https://link.springer.com/article/10.1007/s12525-025-00815-6/tables/9 . 11 July 2025. Read at source on 2026-10-10 through r.jina.ai.
73. Kolscan, leaderboard, “Daily” tab. https://kolscan.io/leaderboard . Live page; snapshot of about 12:15 UTC, 10 October 2026. Read at source on 2026-10-10 through r.jina.ai.
74. MadeOnSol, “What is a KOL in crypto”. https://madeonsol.com/blog/what-is-a-kol-in-crypto . 5 June 2026, updated 15 September 2026. Read at source on 2026-10-10. Vendor data.
75. eToro, CopyTrader page. https://www.etoro.com/copytrader/ . Undated. Read at source on 2026-10-10 through r.jina.ai.
76. The Motley Fool, Stock Advisor page. https://www.fool.com/services/stock-advisor/ . Returns as of 10 October 2026. Read at source on 2026-10-10. Seller’s own figure.
77. KuCoin blog, citing a 2025 copy-trading study. https://www.kucoin.com/blog/is-crypto-copytrading-profitable-in-2026 . 2026. Blog read at source on 2026-10-10 through r.jina.ai; secondary for the study it cites.
78. MadeOnSol, Solana KOL trading report for August 2026. https://madeonsol.com/blog/solana-kol-trading-report-august-2026 . 28 August 2026. Read at source on 2026-10-10. Vendor data.
79. Merkley, Pacelli, Piorkowski and Williams, “Crypto-influencers”, Review of Accounting Studies. https://link.springer.com/article/10.1007/s11142-024-09838-4 . Published 2024. Read at source on 2026-10-10 through r.jina.ai.
80. Dorfleitner and others, “To follow or not to follow – An empirical analysis of the returns of actors on social trading platforms”, Quarterly Review of Economics and Finance 70, abstract. https://ideas.repec.org/a/eee/quaeco/v70y2018icp160-171.html . 2018. Abstract read at source on 2026-10-10.
81. Bradley, Hanousek, Jame and Xiao, “Place Your Bets? The Value of Investment Research on Reddit’s Wallstreetbets”, Review of Financial Studies 37(5), abstract on the University of Kentucky’s publication record. https://scholars.uky.edu/en/publications/place-your-bets-the-value-of-investment-research-on-reddits-walls/ . May 2024. Abstract read at source on 2026-10-10 through r.jina.ai.
82. MQL5, trading signals marketplace. https://www.mql5.com/en/signals . As served 10 October 2026. Read at source on 2026-10-10 through r.jina.ai.

Corrections

If you are quoted or named on this page and think something is wrong, want your reply shown beside it, or want to be removed, write to [email protected]. We aim to reply within 14 days, and always within one month. Factual errors are corrected with a dated note here. A sentence that is seriously disputed comes down while we check it.

Change log: (empty)