Opt-out, reversed
The asset class marketed as opting out of the banks has, in a decade and a half, replicated every concentration pattern of traditional finance plus unique scam vectors that TradFi has regulated against for 90 years. Sources: Chainalysis, Glassnode, SEC and DOJ filings, BIS, Cambridge CCAF. This page does not argue crypto is worthless; it argues the gap between the pitch and the measurable reality is where the money actually gets made, and not by most of the people putting money in.
The entry-timing problem
Pick a month and a dollar amount. The calculator walks month-end BTC closes from that date to today, and shows what you would have sat through to hold. The S&P 500 line is the alternative, not to argue it "beat" BTC (it often didn't) but to keep the opportunity-cost honest. The number to watch is max drawdown and months underwater: those are what retail buyers actually capitulate to.
Portfolio value, indexed to entry = 100
Month-end close, BTC vs S&P 500 total return from Nov 2021 onward
Slide the entry month. Late-2017 and late-2021 entries, the two cycle peaks (which is when retail attention and ad spend peaked), both show multi-year drawdowns of 70%+ and 24-30 months under water. Those are the cohorts that show up in the "73% retail loses money" figure: not because the asset class is a scam, but because human buying behavior is procyclical. Monthly closes from CoinGecko and Yahoo Finance (IVV adjusted). Data current as of March 2026.
The 2022-24 collapse chain, to scale
Every major documented loss event in the industry, sized to the same linear axis. Terra's $60B depeg dwarfs everything that came after, including FTX, which dominated the press. The color shows the failure mode; the clustering shows that 2022 was a lender-contagion wave triggered by one stablecoin, not a dozen independent scandals.
The memecoin survival problem
~99% of memecoins go to zero within a year. On-chain data puts the median lifespan under 60 days. Click "buy 10 random", each token gets $100 and a calibrated 12-month fate. Re-roll. The cumulative P/L across rolls converges to what the data says: the retail memecoin basket loses, because the rare 10-100x does not cover the 80%+ total wipeouts.
Concentration
4 itemsWho actually holds the asset. 2% of BTC wallets hold ~95% of supply. Tether, the unaudited stablecoin, facilitates ~70% of daily trading volume. BlackRock's spot-BTC ETF is now the single largest Bitcoin holder, achieved within 18 months of approval. The Trump family runs commercial crypto ventures in parallel with federal policy toward the asset class. Every concentration pattern the asset was marketed against has reappeared faster than in traditional finance.
Glassnode / NYDIG on-chain analysis: roughly 2% of Bitcoin wallets hold approximately 95% of the circulating supply, and the top 0.01% of wallets hold a majority. Comparable US equity-market concentration is lower. "Democratized money" is the marketing; Gini-coefficient measurements on BTC ownership are in line with the highest-inequality traditional financial assets.
Tether (USDT) is the dominant crypto stablecoin at ~$155B market cap, facilitating roughly 70% of total daily crypto trading volume. Tether has never produced a full independent audit of its reserves; it publishes attestations that are narrower in scope and subject to different professional-liability standards than a SOX-style audit. CFTC fined Tether $41M in 2021 for false statements about reserve backing. The entire crypto market is structurally dependent on Tether's solvency.
Spot Bitcoin ETFs were approved by the SEC in January 2024. Within 18 months, BlackRock's iShares Bitcoin Trust (IBIT) became the single largest holder of Bitcoin at roughly 700,000 BTC (~$70B+ AUM at current prices), exceeding Satoshi's presumed unmoved holdings. The institutional capture of the asset originally pitched as anti-institutional was functionally complete within the first post-ETF cycle.
The Trump family operates two crypto ventures covered elsewhere on this site: $TRUMP meme coin (May 2025 dinner-for-holders gala; ~$320M in family trading fees) and World Liberty Financial (~$550M raised, 75% of platform revenue to a Trump-family entity, including a $2B tranche to a UAE-linked sovereign-wealth fund closed alongside favorable policy decisions). The largest single case of a sitting US president with a disclosed-ownership commercial position in an asset class they directly regulate.
The losses
4 itemsThe 2022-23 "crypto winter" wiped $20B+ in US customer funds across FTX, Celsius, Voyager, Genesis, BlockFi, 3AC, and others. $3.3B more in 2024 hacks and exploits. ~$10B/yr in pig-butchering investment scams targeting US victims. ~99% of memecoins go to zero within a year. Not a single category is covered by an FDIC/SIPC equivalent. The losses the industry generates are real; the consumer-protection infrastructure around them is, structurally, zero.
FTX collapsed in November 2022 with approximately $8 billion in customer-funds shortfall. Sam Bankman-Fried convicted on seven counts of fraud and conspiracy in 2023, sentenced to 25 years. Creditors ultimately recovered most of the shortfall at pre-collapse dollar values via bankruptcy proceedings, not Bitcoin values; customers did not recover the asset appreciation they would have seen had FTX not defrauded them. FTX's insurance fund, advertised as protection, was a spreadsheet constant the exchange could overwrite at will.
Celsius (~$4.7B), Voyager (~$1.3B), BlockFi (~$1.2B), Genesis ($4B+), 3AC (~$10B), Gemini Earn (~$900M) and others all filed bankruptcy or paused withdrawals in the 2022-23 cycle. Aggregate customer losses exceed $20 billion. Most of these entities marketed themselves as "crypto banks" but held none of the FDIC-style protections that term would imply in TradFi.
Chainalysis 2025 Crypto Crime Report: $3.3 billion in documented crypto hacks and exploits in 2024, with DeFi protocol exploits the largest category. Additional ~$10B+ in investment-scam ('pig butchering') losses attributed to US victims specifically, much of it flowing to industrial-scale operations in Southeast Asia using human-trafficking labor. The losses are not covered by any investor-protection fund.
Multiple on-chain studies (Dune, Messari, Bitquery) have found that roughly 99% of memecoins launched on Solana, Ethereum, and Base in 2024 lost more than 95% of their peak value within 12 months; median lifespan is under 60 days. Pump.fun alone has hosted millions of new token launches, the vast majority designed as short-duration insider rugs. Retail investors who find these tokens through influencer marketing are structurally the exit liquidity.
Who profits
4 itemsExchanges take 0.6-1.2% per trade regardless of direction. A dozen HFT market-making firms account for over half the on-chain volume. Four mining pools control 80% of BTC hash. Trump-family crypto ventures collect platform fees while the federal government accumulates BTC as a reserve asset. The business model of the ecosystem is not the technology; the business model is volume, fees, and custody, the same as TradFi.
Coinbase (COIN) reported $1.8B net revenue in Q3 2024, >$5B annualized. Transaction fees on retail trades (0.6%-1.2% per side, higher on pro tier) are the dominant revenue source. Coinbase is essentially a regulated casino: the house takes a percentage of every transaction regardless of whether the underlying asset moves up or down. The business model is consistent with Binance, Kraken, and others in the category.
Jump Trading, Cumberland, Wintermute, and ~10 other high-frequency trading firms account for more than half of daily crypto trading volume. These firms have preferred access to exchange APIs, co-location, and order-flow data; retail traders systematically lose to them in the same way retail equities traders systematically lose to HFTs (PFOF internalizers plus market makers). "Level playing field" rhetoric in crypto mirrors the pre-Reg NMS equity market of the 1990s.
BTC proof-of-work mining concentration: four pools (Foundry USA, AntPool, F2Pool, ViaBTC) regularly control 75-85% of total hash rate. A single pool repeatedly exceeded the 51% threshold in early 2024, prompting community debate about mining centralization. Institutional mining firms (Marathon, Riot, Core Scientific) have consolidated retail miners out of the market; "anyone can mine" is historically accurate but economically no longer true.
In March 2025, Trump signed the Strategic Bitcoin Reserve executive order, using federal holdings of seized BTC (~200K coins) as the initial reserve. The policy makes the federal government a price-sensitive holder of the asset class the president's family separately operates commercial ventures in. No other US president has a disclosed direct commercial interest in an asset class they have moved to position as a federal reserve.
Pitch vs reality
4 itemsThe marketing pitches were 'opt out of the banks', 'democratize money', 'save the unbanked', 'payments efficiency'. The measured outcomes: 73% of retail investors losing money, El Salvador's legal-tender experiment functionally reversed, BTC payments using 15× the energy per transaction of Visa, and no consumer-protection architecture. None of this means the technology is worthless. It means the gap between the marketing and the measurable is the central piece of due diligence before putting savings into the asset class.
BIS (Bank for International Settlements) 2022 study: approximately 73% of US retail crypto investors who entered between 2015 and 2022 had negative returns when measured at market-cycle troughs. The time-weighted investor return is materially worse than the asset's time-weighted return because retail flows are pro-cyclical (buy high during hype cycles, sell low in crashes). The pattern holds across major equities and TradFi products; crypto's volatility just makes it sharper.
El Salvador adopted Bitcoin as legal tender in September 2021. By 2024 the government and IMF quietly unwound most of the program: <10% of households ever used Bitcoin for transactions, merchant acceptance outside government-mandated points was negligible, and the government sold much of its BTC holdings in 2024-25 during the post-ETF rally. It is the one completed experiment in country-level BTC-as-money adoption, and it functioned as a tourist/speculator attraction, not a currency.
Cambridge Centre for Alternative Finance: a single Bitcoin transaction consumes approximately 15× the electricity of a Visa transaction on a like-for-like basis. Annual BTC network energy consumption is ~150 TWh, roughly Argentina's total. Proof-of-work is secure precisely because it is expensive; the energy is not "waste" in the cryptographic sense. But the payments-efficiency pitch that dominated early BTC marketing is the opposite of the measured technical reality.
US crypto exchanges and wallets are not covered by FDIC, SIPC, or any other investor-protection backstop. If your exchange fails, customer funds are unsecured claims in bankruptcy (average FTX recovery: ~$0.80 on the dollar at 2022 USD values, $0 of the asset-appreciation recovery). The asset class is marketed as equivalent to savings and investment accounts; the protection architecture is equivalent to neither. This is a feature, not a bug: insurance funds require solvent counterparties and regulatory capital, which are the TradFi characteristics crypto was pitched against.