To The Moon

$Coinbase Global, Inc.(COIN)$  

The short answer

The share price mostly trades like leveraged Bitcoin. Over the last 52 weeks the correlation of weekly returns is 0.82, and COIN's beta to BTC is 1.19 over two years and 1.16 over one year. But the business has a second engine that runs on the Fed, not on crypto: in Q3, $430.6M of $1,215.6M total revenue (35.4%) came from USDC and interest. My read: that is the main cushion, and it is smaller than the "half of revenue is recurring" line suggests, because part of subscriptions and services (staking rewards) still moves with crypto prices.

1. The overlay: early at tops, late at bottoms

On weekly closes indexed to COIN's listing week, COIN topped on 18 Jul 2025 and BTC on 3 Oct 2025, so COIN led by about 11 weeks. In the last bear market BTC bottomed on 25 Nov 2022 and COIN on 6 Jan 2023, about 6 weeks later. The 2026 lows so far fit the same shape: BTC 26 Jun, COIN 31 Jul, about 5 weeks later.

Limits: COIN has only one full cycle, so this is n=1, and neither 2026 low may be the real bottom. This is a pattern, not a forecast.

2. The crypto engine is weak

Consumer trading revenue was $450.1M in Q3, the 6th lowest of the 23 quarters since 2021. The part of Coinbase that is truly leveraged Bitcoin is running near its lows.

3. The rate engine, in Coinbase's own numbers

The FY2025 10-K (Item 7A) says a ±150bp move in rates would have changed stablecoin revenue by $540.3M and interest on cash equivalents by $143.2M. Divide by six and each 0.25% is worth about $114M a year (my maths: static balances, symmetric as disclosed; customer custodial funds are not included).

What actually happened in Q3: lower rates cost $52.8M year on year, while higher average USDC held on Coinbase added $84.6M. So far, USDC growth has beaten rate cuts. Rate risk now depends mostly on USDC balances.

Method you can reuse: for any broker, bank or exchange, find the market-risk section, divide the disclosed shock by the number of 0.25% steps, then check the latest quarter's rate-versus-volume bridge to see which one is winning.

4. Who shares the USDC money

Under the Circle arrangement, Coinbase earns a share for USDC held on its platform and receives 50% of the Payment Base left after Circle's issuer share and payouts to other partners. In Q2, Circle paid Coinbase $324.6M out of $667.7M of reserve income, about 48.6% (my maths). In Q3 the average USDC supply was $73B, of which about $22B (~30%) sat in Coinbase products, and Coinbase renewed the arrangement on the same terms. The on-platform share is the lever to watch.

5. The counter-case, given real space

Q3 net income of $718.7M is above adjusted EBITDA of $332.9M. The gap is $715.0M of gains on crypto assets held for investment; operating income was only $61.7M. So the headline profit is mostly a crypto mark, which is the leveraged-Bitcoin story again, just on the balance sheet. Operating cash flow was $405.5M versus $709.9M a year ago. I can't calculate free cash flow cleanly because Coinbase has no separate capex line, so I won't print one.

And one honest caveat on point 1: on daily closes, COIN's lowest close of 2026 was on 12 Feb, before BTC's low. The 5-week lag holds on weekly closes only.

Narratives, kept separate from numbers

Prediction markets: "$200M+ annualized revenue" is Coinbase's own claim in its Q3 deck; there is no line item for it in the 10-Q. Base: Coinbase discloses no Base revenue, and a token is narrative only.

Tests that could prove me wrong

1. Q4 2026 consumer transaction revenue comes in below $450.1M (Q4 print).

2. COIN's next low again lags BTC's by 4-6 weeks.

I'll come back and score both, hits and misses.

Question for you: when a stock trades like leveraged crypto but over a third of its revenue runs on rates, which one do you model first?

Not advice. I may hold positions in the names mentioned.

Stop guessing. Start calculating.

Live to fight another day. 🤙

#options #tradingpsychology #riskmanagement #optionstrading