The Tools You're Using To Trade Crypto Are Built To Extract Money From You.

There is a retail kid trying to learn crypto. He is twenty-two, or sixteen, or forty-eight. He works a normal job. He has saved up a few thousand dollars. He has heard that crypto is where wealth is being created and he wants in. He finds a funding rate scanner. The scanner has a polished interface, full of features, run by people who appear to be helpful. He clicks the 1-Click Trade button on the funding arbitrage screener because the screener implied that 1,500% returns were possible. He puts up his savings. He gets liquidated on the leveraged short during a snapshot window mismatch. He concludes, like so many before him, that crypto is a scam.

He is half right. The technology is not a scam. The industry that has grown up on top of it largely is. The kid does not have the experience to tell the difference, and the people taking his money have no incentive to teach him.

I want to be specific in this article about how that machine actually works, because most retail traders do not understand the business model of the tools they are using, and the moment you do understand it, the tools stop looking helpful.

The business model of a funding rate scanner

The popular funding rate scanners and aggregators all run on the same model. The site is free to use. The site shows you funding rates across many exchanges. The site flags arbitrage opportunities. The site offers, prominently, a way to execute on those opportunities. That executable button routes you to one of the listed exchanges through a referral link. When you open a position via that link, the scanner collects a fraction of every trading fee you generate from that point forward, often for the lifetime of the account.

That is the entire business. The site exists to convert retail attention into exchange volume routed through their referral codes. The exchange gets the user, the scanner gets the recurring revenue, and the user gets whatever happens to him in the trade.

There is nothing inherently wrong with referral revenue. Affiliate models exist all over the internet and most of them are legitimate. The problem is what the model does to the design of the tool when the tool is the only thing standing between a retail trader and a position they are about to put on with real money.

The scanner gets paid when you click. The scanner does not get paid when you walk away because the trade looked bad. So every design decision in the scanner is biased, gently or not so gently, toward making more clicks happen. Bigger numbers in bigger fonts in greener colors. Fewer caveats. Faster paths from looking to clicking. The 1-Click Trade button itself is the most extreme version of this design pressure, but the same pressure is in every other element of the page.

This is not a conspiracy. The people who build these tools are not necessarily acting in bad faith. They are responding rationally to the incentives. If clicks pay and caveats reduce clicks, the rational design is fewer caveats. The result, summed across the industry, is a category of tool that systematically misleads its users in subtle ways, all of which point toward the same outcome of more trades being placed.

What gets hidden

The omissions are not random. They are predictable consequences of the business model.

Funding interval mismatch. Almost every scanner shows a “spread” between two exchanges that subtracts the rates and ignores the fact that the two rates are paid on different clocks. The displayed spread is a fiction in cases where the intervals do not match. Surfacing this would reveal that most of the headline arbitrage opportunities are not actually capturable. Surfacing it loses clicks.

Time to next snapshot. No major scanner shows you when each leg’s next snapshot fires. Without that information, you cannot tell whether the rates you are looking at will be the rates you collect. Surfacing it would force users to wait for alignment, which means fewer trades, which means fewer clicks.

Realized vs projected APR. Every scanner shows annualized rates as if they will persist. None of them show you the realized rates over the past thirty days, which would let you see that the projected APRs almost never materialize. Surfacing realized rates would expose how far the projected rates are from reality. It would lose clicks.

Liquidation distance. Every scanner that mentions leverage shows you the projected return at high leverage without telling you what adverse price move would liquidate you. A 10x leveraged position that returns “1,500% APR if held for a year” is one that gets liquidated by a 5% move against you, and 5% moves on memecoins happen daily. Surfacing the liquidation distance would expose that the leveraged trade is structurally a coin flip with the venue holding the edge. It would lose clicks.

Slippage and fees. The math on the page assumes you fill at mid and pay no fees. In practice, on the assets where the headline numbers come from, the order book is thin enough that any meaningful size eats slippage that exceeds the expected spread capture. Surfacing realistic execution costs would shrink the headline numbers to something honest, which would not look as exciting on the homepage. It would lose clicks.

Each of these omissions is individually small. Stacked together, they make the difference between a tool that helps a retail trader and a tool that funnels him into trades the tool’s operator profits from while he does not.

What the operators tell themselves

The operators of these tools are not all moustache twirling villains. Most of them genuinely believe they are helping. The story they tell themselves is that they are democratizing access to information that was previously only available to professional traders. Funding rates are public data. Cross-venue spreads are public data. Aggregating that into one place is a service. The trader who gets liquidated did not get liquidated because of the scanner, the story goes. He got liquidated because he used too much leverage on a volatile asset, which is his own mistake.

There is enough truth in this story that it functions as a narcotic for the people who run these tools. Yes, the data is public. Yes, the trader is responsible for his own decisions. Those things are true. But the story leaves out that the tool’s design materially shapes what the trader believes he is looking at, and the design is optimized for exchange volume, not for the trader’s actual trading outcomes. A tool that helps you make better decisions and a tool that makes you make more decisions are not the same tool. The industry has converged on the second one because the second one pays.

What an honest tool would do differently

I am building an honest version of this kind of tool, so I have thought about what the differences actually look like in practice. They are not subtle.

An honest tool would surface fewer opportunities, not more. On a typical day, two or three setups might pass a rigorous bar instead of fifty. The bar is matched funding intervals, sufficient open interest on both legs, persistent spread over the past several days, and realized historical performance that justifies the projected return. Most days, the rigorous bar will reject most opportunities. That rejection is the point.

An honest tool would project realized APR rather than instantaneous APR. The number on the page would be what this kind of spread has actually paid over the past thirty days for this asset, not what it would pay in the impossible world where the current spread holds for 8,760 hours. The realized number would usually be a fraction of the instantaneous one, and that fraction would be useful information.

An honest tool would surface execution constraints alongside theoretical opportunities. If a market only has $500K of open interest on one leg and $20M on the other, the position is sized constrained on the small leg, and any meaningful size eats slippage. The tool would say so.

An honest tool would be transparent about referrals. When the tool earns money from referrals to specific exchanges, that disclosure would be on every page where those exchanges appear, in plain text. Not a buried footer link. Not a tooltip. The user would always know that the recommendation comes with a financial relationship, and the user would know what that relationship is.

An honest tool would be willing to say that an exchange is bad even when the exchange pays well. If a venue lists pump and dump memecoins, has poor liquidation engine reliability, or routinely engages in practices that harm retail traders, the tool would say so. Even if the venue’s referral program is the most generous in the industry. Even if saying so cost real money. The cost is what makes the editorial credible.

None of these are technical innovations. They are all editorial choices. They are choices that no scanner with a referral driven business model can make consistently, because each of them reduces the rate at which users put trades on, which reduces revenue.

Why I am writing this

I am building one of these tools myself. The version I am building includes the honest defaults I just described. It will not generate as much referral revenue per visitor as the casino versions do. That is fine. The point of the tool is to actually be useful to the retail traders using it, which over the long run is a better business than the alternative anyway, even if it is a slower one.

But the article is not really a pitch for my tool. The article is a request that you, the reader, look at the tools you are currently using and ask yourself what their incentives are. If the tool is free to use, the tool makes money somewhere, and that somewhere is almost always your trading volume. The design of the tool is shaped by that fact whether the operators want it to be or not. The headline numbers on the page reflect that shaping. The trade you are about to put on reflects it too.

You can still use these tools. They contain real data, mostly correctly aggregated. But you have to read them with the knowledge of what they are actually selling you, which is yourself, repackaged as an opportunity.

If you would like to see what an honest version of the same kind of tool looks like, the funding page is where I am working through it. Same data, mostly. Different defaults. Fewer opportunities surfaced. More information about what would make those opportunities actually capturable. It is a slower, less exciting version of the same category of tool. That is the entire argument.