This theme groups names whose value is driven by digital-asset prices and trading volume — not by the tokenization of real-world assets (covered separately under Asset Tokenization). It is deliberately a factor-exposure bucket, not a structural thesis: the question it answers is “which listed equities give crypto beta, and through what mechanism.” Two sub-themes split that beta cleanly between miners and non-miners.
Post-halving economics broke the model: hashprice collapsed to ~$30/PH/day in Q1 2026 against a weighted-average cash cost near $80k/BTC, leaving a large share of the fleet underwater. The sector has pivoted hard into AI/HPC — CoinShares reports >$70B in announced AI/HPC contracts and estimates listed miners could derive up to ~70% of revenue from AI by end-2026 (from ~30%). Core Scientific (CoreWeave, $10.2B), TeraWulf ($12.8B contracted), Hut 8 (multi-billion AI leases), and IREN (a ~$9.7B Microsoft GPU deal) are being re-underwritten as infrastructure landlords. The “mining” label increasingly understates an AI-datacenter business.
The MSTR flywheel — issue equity at a premium, buy BTC, repeat — depends on trading above mNAV. That premium fell from ~3.9x at the 2024 peak to ~0.94x by late May 2026, as spot Bitcoin ETFs removed the need for an equity wrapper and copycat treasuries (Metaplanet, others) ended MSTR’s monopoly on the “public BTC proxy” trade. Treasury names now carry dilution and refinancing risk on top of raw crypto beta.
Within the non-miner bucket the exposures differ sharply: an exchange (COIN) earns on volume; a treasury vehicle (MSTR, SBET, BMNR) is a levered holdings proxy; a diversified digital-asset financial-services firm (GLXY) blends trading, asset management, and — increasingly — AI datacenter revenue. Conflating these under one “crypto” label is the trap the sub-theme split is designed to avoid.
The miner re-rating is a bet that signed leases become recognized revenue at ~85% margins. The first clean quarter where a CORZ or WULF shows AI revenue dominating and margins hitting target validates the “infrastructure landlord” thesis; a construction-milestone slip (a known risk for CORZ) cuts the other way.
With BTC trading well below the ~$80k weighted cash-cost line at points in 2026, a sustained move back above ~$100k restores mining margins and hashprice; a sustained move below ~$70k could trigger capitulation that paradoxically benefits survivors via lower difficulty. The single biggest swing factor for the pure-mining names.
A durable move back to a premium re-arms the equity-issuance flywheel; a durable discount forces the shift to preferred issuance (already underway) and raises the tail risk of forced BTC sales to service convertible debt. Watch the weekly close around prior support levels as the technical tell.
CoinShares flags clean-balance-sheet operators (HIVE, CLSK) as potential acquirers of distressed or power-rich peers. Consolidation would reshape the membership of the miner sub-theme and is the most likely near-term structural event.
Each incremental spot-ETF or direct-access product erodes the “equity wrapper” premium for treasury vehicles. Conversely, friendlier US crypto market-structure rules would lift exchange-volume names (COIN) and the broader complex.
Highly cyclical and substantially priced — these names are among the highest-beta equities in the market (MSTR beta ~3.5), so the theme is a crypto-cycle expression rather than a slow-burn structural call. Net direction is a Tailwind when the crypto cycle and AI-datacenter demand are rising, but the exposure is two-sided and cycle-dependent. Time horizon is 3–5 years, tracking the crypto cycle and the AI-pivot build-out. Theme conviction is Medium — deliberately below the structural themes, because this is factor/cycle beta, not a durable competitive-advantage thesis. The most durable economics in the group may end up being AI-datacenter revenue, not crypto at all.
The in-universe roster is 17 names — 12 miners (MARA, RIOT, CLSK, CORZ, IREN, HUT, WULF, CIFR, BTDR, BTBT, HIVE, CAN) and 5 non-miners (MSTR, SBET, BMNR, GLXY, COIN). A few adjacent names and classification notes worth tracking:
Japanese BTC-treasury vehicle and the archetype of the MSTR copycats whose proliferation compressed the treasury premium. Non-US listing; tracked as context, not added as an edge.
Included in miners as the ASIC-hardware “picks and shovels” exposure rather than a mining operator — its mechanism is miner capex, which moves with BTC-price and hashprice expectations. Next-gen hardware cycles (sub-10 J/TH rigs) are its own catalyst.
Not exhaustive — see Neo4j for the full edge list.