Independent Crypto Research for Serious Investors

No paid conclusions. No guaranteed-price claims. No invented charts. Every research article is dated, sourced and written with the downside in view.

Latest research

• A $400 Million Ponzi Was Sitting on a Public Blockchain the Whole TimeJuly 30, 2026
  Goliath Ventures CEO Christopher Delgado pleaded guilty to a $400M Ponzi scheme built on fake DeFi liquidity pool claims. The pools were public. Here’s what checking them would have shown.
• 
  Hot Wallet vs Cold Wallet: Use Both, but Never for the Same JobJuly 29, 2026
  Hot wallet vs cold wallet is not a winner-takes-all choice. Use a replaceable hot wallet for activity and a separate cold wallet for meaningful savings.
• 
  How to Move Crypto Off an Exchange Without Losing ItJuly 29, 2026
  A safe withdrawal is not one click. This field guide shows how to verify the asset, network, address, test transaction and final balance before moving meaningful funds.
• 
  The Morgan Stanley Bitcoin ETF Call, GradedJuly 28, 2026
  Why Morgan Stanley’s Bitcoin ETF matters less for its first-day flow and more for the regulated distribution network it opens to investors.
• 
  Three Arrows Capital Didn’t Die From Luna. It Died From a Trade Two Years Older.July 28, 2026
  3AC didn’t collapse from the Luna crash. It collapsed from a $2.36 billion undercollateralized GBTC arbitrage loan built two years earlier. Tracing the cash through the actual court filings.
• 
  Seed Phrase vs Passphrase: The Difference That Can Lock You OutJuly 25, 2026
  A passphrase and a seed phrase are not the same thing. Here’s how BIP-39 passphrases actually work, and why a small typo can permanently lock you out.

What we publish

Evidence before conviction

• Time-stamped evidence: market figures include a clear observation period and data cut-off.
• Source hierarchy: filings, regulators, protocol documentation and direct datasets take priority over commentary.
• Reproducible charts: original charts state the instrument, venue, interval, transformation and source.
• Uncertainty disclosed: estimates, proxies and incomplete datasets are labelled rather than presented as facts.
• Corrections preserved: material errors are corrected transparently with a dated note.

Start with how this research works.