A 3x Paper Gain Where You Can Only Cash Out 30%

Deep thoughts on AI and aspirations —— ByteDance Deep Thinking Circle

In the fall of 2023, friend.Tech was crypto’s hottest product. Spend a bit of Ethereum to buy a KOL’s Key, join their private chat, and sell the Key anytime. Simple design, nearly everyone saw their paper returns double or triple, screenshots flying everywhere.

But between the price displayed on screen and the value you can actually walk away with lies a very wide gap. Breaking down this gap delivers something far more valuable than the product itself, because it appears in every closed-pool product where “everyone is making money,” and it will appear again and again.

The Moment You Enter, You’re Already Down Nearly 20%

Let’s lay out the rules first. Friend.Tech uses an extremely simple pricing curve: a Key’s price equals the number of holders squared, divided by 16,000. More holders means higher prices, and the rate of increase itself accelerates—each new entrant pushes the price higher than the last, a textbook ponzi curve. Every transaction incurs a 10% fee, split equally between the protocol and the Key owner.

On the surface, the transaction cost is 10%. Calculate it once and you’ll see this number is deceptive. Say you enter with 1.1 Ethereum. Buying incurs a 10% fee, so you can only purchase Keys worth 1 Ethereum. When you sell, you pay another 10%. In other words, from the moment you hit buy, this position’s liquidation value locks in at 0.9 Ethereum—the exit fee is guaranteed, just deferred.

Round trip, the real loss is 1 minus 0.9 divided by 1.1, approximately 19.2%. Prices need to rise 22% just to break even.

The official line is “high fees encourage long-term holding.” This explanation doesn’t hold: if the goal is to encourage holding, taxing sellers would suffice. Also taxing buyers 10% doesn’t encourage any holding, it only increases extraction. To judge a mechanism’s true intent, look at who pays fees, not how it’s explained.

Three People Split One Cow: The Paper Value Illusion

The real sleight of hand is on the second layer. A thought experiment: Zhang, Li, and Wang pool money to buy a cow, a duck, and an egg, agreeing that whoever exits first gets the cow, second gets the duck, last gets the egg. All three feel they “own” claim to a cow, but there’s only one cow. Average across all six possible exit sequences, and each person truly owns one-third of a cow, one-third of a duck, one-third of an egg.

Friend.Tech’s universal paper gains use this exact illusion. This product has only one counterparty: the pool. The TVL in the pool is the truly liquidatable capital, while displayed Key prices are paper numbers calculated by the curve. For instance, when Key count hits 40, price is 0.1 Ethereum, paper market cap is 4 Ethereum, but the pool actually contains only about 1.38 Ethereum.

Under this algorithm, once Key count exceeds roughly 20, the ratio of liquidatable value to paper value stabilizes around 30% and stops climbing. This 30% has two implications. First, anyone buying in the flat portion of the curve loses approximately 70% of expected value the instant they buy, on top of the two fee layers. Second, the Room Value displayed on screen is inherently over three times the liquidatable value. The supposed “everyone made 2-3x” is largely just mistaking a 3x paper illusion for money.

Every bit of paper profit you see comes from diluting the expected value of the next buyer. This accounting doesn’t require a collapse to be true—it’s happening every moment.

The Top Doesn’t Need a Crash, It Walks Itself Down

Next comes the most elegant and brutal part of this structure: it self-collapses without any external shock.

The reasoning chain goes like this. Assume growth caps at N participants. Then buyers entering after position N-M are guaranteed not to break even. Everyone can see this information, so rational actors won’t buy after N-M. But if no one buys after N-M, then buyers after N-M-L can’t break even either, so even earlier positions lose buyers. Layer by layer in reverse, equilibrium price continuously migrates downward. This is the classic “beauty contest” reasoning from game theory: everyone is guessing when everyone else will stop, so the stopping point keeps advancing.

The actual micro behavior: the moment net inflow slows, high-priced Keys become unprofitable first, speculative capital pivots to cheaper Keys, cycle repeats, and individual Key price ceilings drop notch by notch. From a paper value perspective breaking even looks easy—a Key bought for 5 Ethereum only needs 27 new buyers; from an expected value perspective, a Key bought for 1 Ethereum needs 115 new buyers to break even—and the supply of new buyers itself is drying up.

Making matters worse are bots. Friend.Tech is rife with bots that camp new Key launches in the low-price zone. When the equilibrium price for minting drops to a certain level, it falls directly into bot arbitrage range, shaving another layer off ordinary users’ expected value.

The popular mutual-buy etiquette in that circle (you buy my Key, I buy yours) can’t hold either. Mutual holdings are stable between two people—like exchanging hostages. Add more participants and front-running becomes profitable. Once the chain of suspicion forms, the final equilibrium is everyone losing together. This etiquette looked solid during the upswing only because paper gains masked the continuous drain of expected value. Once inflow stops, front-running immediately becomes the dominant strategy.

The Business Model Is Actually a Pump

Add up all the friction and the product’s true face becomes clear.

By the data at the time: at roughly $36 million TVL, cumulative fees had reached $24 million, of which the protocol took $12 million. Even by the most conservative measure, with roughly $48 million in capital entering, a quarter became protocol revenue in about two months. Projecting forward: with paper market cap around $110 million and 5% daily turnover, about $16.5 million gets extracted monthly, nearly 45% of TVL. All net deposits are continuously flowing to the protocol.

For scale comparison: NFT marketplace OpenSea charges 2.5% royalty plus 2.5% fees round trip, versus 20% here—friction four times higher. At this extraction rate, debating “whether this product is a ponzi” becomes secondary. It’s first and foremost a machine that continuously pumps away deposited capital. The ponzi structure is just the shell that lets the pump run a bit longer.

This arithmetic can audit any closed-pool product with three calculations:

If any of these three numbers significantly exceeds comparable markets, the on-screen yield isn’t credible.

The Exit It Could Have Had, and Why It Didn’t

To be fair, this product’s design genuinely considered an exit. The exit logically works: if real services emerged behind Keys—group owners continuously producing information, maintaining relationships, delivering actual benefits to holders—these “utility holders” would become subordinated claimants, not rushing to sell, genuinely improving remaining holders’ expected value. Back to that thought experiment: if Zhang commits to exiting last, what Li and Wang hold transforms from “one of three” to “one of two,” immediately increasing gold content.

Subsequent events gave this hypothesis a complete test. Result: falsified. After the airdrop landed, users collectively sold and exited, activity collapsed, product heat dissipated within about a year. Utility demand didn’t grow; speculative demand did jam the exit channel.

My judgment is more pessimistic than “utility didn’t grow”: speculative cold-starts actively block the path toward utility economics. Speculation prices on front-running and flipping; utility users price on service value. Two pricing logics on the same curve cannibalize each other. A “group membership” bid up to several Ethereum by speculators becomes a prohibitive price for those genuinely wanting service—the service itself might be worth tens of dollars monthly, but the admission ticket has been priced to another order of magnitude by gamblers. Earlier speculators raised the barrier for all who follow. The product quotes real customers at peak prices. Cold-starting with speculation’s force must be repaid with interest later.

This conclusion’s boundaries should be clear. The 30% gold content estimate rests on the assumption “all buyers are speculators.” If utility demand truly exists, the ratio exceeds 30%. So to judge whether a similar product has hope, ignore narratives and watch two observable variables: issuer self-holding ratio, and non-speculative holdings (those who paid for usage rights without intent to flip). If these ratios don’t rise, paper value remains forever illusory. If they rise, even closed pools can grow real business.

Back to the original question: facing any closed-pool product where “everyone is making money,” discount the displayed price by 70% before deciding whether to enter. 70% off isn’t a precise figure—it’s a reminder: when your counterparty is a pool rather than a market, the number on screen never promised it was your money.