I have tried, more than once, to say something useful to them. Not "scalping is stupid" — I don't believe that, and I spent years doing it myself. Something narrower and more honest: you have been at this for a year or more, you are stressed, you have not been able to leave your job, and there is a version of this that would work better for you. Slow down. Widen the timeframe. Learn the instruments properly. It falls on deaf ears every time.
I have thought about why for a while. And when I went looking, it turns out this is not a personality problem. It is a documented, measured phenomenon, and the research on it is much more damning than anything I would have said out loud.
The people they are learning from are, statistically, the worst source available
There is an academic paper called Finfluencers, from researchers at UC Berkeley, Rice University and the University of Lausanne, published through the Swiss Finance Institute. They took tweet-level data from over 29,000 financial influencers on StockTwits and measured what actually happened to returns if you followed the advice.
They sorted them into three groups. About 28% were skilled — their calls generated roughly 2.6% monthly abnormal returns. Around 16% were unskilled, meaning following them did essentially nothing either way. And the remaining 56% were what the researchers called antiskilled: following them produced roughly negative 2.3% monthly abnormal returns.
Read that again. More than half of the financial influencers studied were not merely useless. They were reliably wrong enough that you would have done measurably better doing the opposite.
That alone would be a good enough article. But it is not the interesting part.
The worse they are, the bigger they get
Here is the finding that explains every conversation I have had with these guys.
The researchers found that the unskilled and antiskilled influencers had more followers, more activity, and more influence on retail trading than the skilled ones. The people giving the worst advice had the largest audiences. And this was not a temporary market inefficiency that competition would eventually clean up — the paper explicitly tested whether followers could identify the skilled influencers and drive the bad ones out, and concluded that they could not.
Why? Two mechanisms, both of which you can see with your own eyes on any timeline.
The first is posting volume. Skilled influencers, the study found, tweet less frequently, and tend to post data-driven assessments that are sometimes negative. Antiskilled influencers post more frequently, which increases their visibility and their perceived authority. Volume gets mistaken for expertise. The person posting nine times a day looks more like an expert than the person posting twice a week, even though the data says the opposite is true.
The second is what the researchers call homophily — people gravitate toward voices that sound like them and tell them what they already want to hear. A confident, optimistic, high-energy account promising fast money is simply more appealing than a measured one saying "this setup is marginal and I'm sitting it out." So the appealing one grows, and the useful one doesn't.
This is the thing I could never quite articulate to my friends. It is not that they got unlucky and followed a bad account. The selection mechanism is inverted. The algorithm that decides who they see is optimising for engagement, and engagement correlates with overconfidence, which correlates with being wrong.
What the flashing cash actually is
An industry audit in 2026 put numbers on the specific formats these guys live in.
On X, 61% of viral trading ideas resulted in losses if you actually followed them. On TikTok, across videos with more than 20.7 million combined views, 80% were graded C or below, and 60% received an F for risk disclosure — up from 30% the year before. The disclosure problem is getting worse, not better.
But the number that matters most came from Discord and Telegram signal groups. Auditors joined seven of them and found that 74% of signals were posted with no stop loss at all — and that 54% of losing trades were never acknowledged.
That last figure is the entire illusion in one statistic. It is not that these channels are winning. It is that slightly more than half the losses simply never get mentioned. What you are seeing when someone posts a screenshot of a five-figure day is not a track record. It is the surviving half of a record where the other half was quietly deleted.
You cannot reason someone out of a belief formed by watching a feed that has had its losses filtered out. The evidence they are drawing on has been pre-selected to be wrong.
The house always wins twice
You had a suspicion about the business model, and the adjacent data supports it.
Look at the prop firm world these traders orbit. FPFX Technology data covering more than 300,000 evaluation accounts found that roughly 14 out of every 100 traders who buy a challenge pass it — and only about 7 ever receive an actual payout. The average payout came to around 4% of the account size. In practice: a trader on a $10,000 funded account typically earned about $400, after paying roughly $100 to enter. The average user bought 1.6 challenges. Some bought as many as 18.
Notice where the reliable revenue sits in that arrangement. Not in the trading. In the entry fees.
The same structure applies to a paid channel. If someone's intraday edge genuinely produced life-changing returns at scale, the marginal hour spent editing content, running a Discord, and posting P&L screenshots would be an expensive hour. The fact that so much time goes into the channel tells you something about which side of the business is actually paying. This is not a conspiracy theory — it is just reading where the incentives point.
And the base rates underneath all of it have not moved. Both the SEC and FINRA have published investor guidance summarising the academic literature: 70–80% of active day traders lose money over any rolling twelve-month period. In the heavily-studied Brazilian futures market, only 1.1% of day traders earned more than minimum wage, and roughly 97% of those who persisted beyond 300 days were unprofitable. Traders keep losing for the same reasons over and over, and none of those reasons are fixed by a subscription.
They didn't know what the underlying was
Here is the moment that pushed me to write this.
In one of these conversations I used the word underlying, in the ordinary way you would use it — the stock beneath the option. Blank look. So I tried derivative. Nothing. This person had been actively trading, with real money, for over a year, following paid signal channels, and they didn't know what the words meant.
I want to be careful here, because that is not an insult. Nobody is born knowing this vocabulary, and there is no shame in not having learned it yet. The problem is the sequence. They had skipped straight to executing high-frequency intraday trades in leveraged instruments without the vocabulary required to describe what they were holding. You cannot evaluate a strategy you cannot name. You cannot assess whether an influencer knows what they are talking about if you do not know what they are talking about either.
That is the actual mechanism by which people stay stuck. Not stupidity. Not laziness. A missing foundation that makes every piece of incoming information impossible to filter. When you do not have the vocabulary, every confident voice sounds equally credible, and the loudest one wins by default — which, as the research shows, is precisely the worst outcome.
What actually changes the outcome
I am not going to finish this by telling anyone to quit. That is not my call, and I do not believe scalping is worthless — it taught me more about how price actually moves than anything else I have done.
But if you have been at it for a year or more and you are still where you started, the honest read is that the format is not working for you, and that adding another paid channel will not change that. Three things would.
Learn the vocabulary first. Underlying, derivative, strike, expiry, premium, intrinsic and extrinsic value, assignment. It is a weekend of work and it permanently changes your ability to filter noise. Someone who knows what extrinsic value is cannot be sold a strategy that quietly relies on it decaying against them.
Widen the timeframe. Swing trading is not a downgrade from scalping. It is where most people who make a living at this actually end up — fewer decisions, better information, less transaction cost, and the ability to sleep. The reason it does not appeal is that it does not produce a daily adrenaline hit, which is exactly the reason it works.
Get your reps somewhere that does not charge you tuition. You can replay five years of real market history and test whether an approach actually holds up before it costs you anything. Every hour spent doing that is an hour not spent paying for the privilege of learning the same lesson live.
The uncomfortable summary of the research is this: the traders posting the most are, on average, the ones performing the worst, and the audience cannot tell the difference. If your entire information diet comes from people who are financially better off selling you the dream than trading it, then the question in the back of your mind — how long am I going to keep doing this and going nowhere — is not pessimism.
It is the most accurate read you have had in a year.