Robot Money Blog
Announcements, research, and updates from the team.
Channel divergence: is macro→crypto breaking?
Three transmission indicators that measure whether risk-on macro is still reaching crypto. Bitcoin's beta to risk-appetite has collapsed from 3.3 to 0.7 since March 2026, BTC/Nasdaq sits at the 1st percentile of its 3-year window, and the stablecoin-vs-Nasdaq flow proxy has turned sharply negative. Three angles, same conclusion: the channel is breaking right now.
Late-cycle signals: how late in the rally are we?
Four slow-moving gauges inspired by BofA's 'signposts of a market peak' checklist, rebuilt from free data: index concentration (SPY/RSP), M&A activity (SEC S-4 filing counts), broker-dealer margin debt, and University of Michigan consumer sentiment. The current configuration is unusual — late-cycle in equities (concentration high, margin debt growing 20%/yr, M&A elevated) but consumer sentiment at multi-decade lows.
AI ate the bull market
Stocks are up a lot in 2025-2026. Bitcoin is not. That hasn't been true in any previous cycle. Seven free-data indicators tell the same story from different angles — beta collapse, BTC/Nasdaq at the 1st percentile, stables-flow turning negative, index concentration at extremes, leveraged speculators on top of recession-level consumer confidence. A plain-English walkthrough with explanatory charts.
Adding an equity factor panel: how the 3-panel /regime improves on the legacy 2-panel
Side-by-side backtest of the current 3-panel regime classifier (macro + on-chain + equity factor) and the legacy 2-panel /regime_2panel. Same 8 years, same point-in-time weights. The 3-panel composite wins on every portfolio — and the factor panel alone produces the best single-panel Sharpe in two of three tests.
Backtesting honestly: what survives when you remove hindsight
Our regime model weights indicators by inverse correlation. We discovered the backtest computed those weights with full-history hindsight — so we rebuilt it point-in-time and re-ran every comparison. Here's what held up, what didn't, and why the honest version is the one worth trusting.
Conservative vs aggressive: combining macro and on-chain regime signals
Most regime classifiers average macro and on-chain panels into a single composite. We backtest two alternatives — combine them conservatively (any panel off → off) or aggressively (any panel on → on, unless contradicted) — across 8 years of ETH/cash allocation.
Treasury Allocation for On-Chain Businesses
8-year backtest comparing regime-based allocation, HODL, and stable yield across a full market cycle.
Regime Detection
Prior art, methods, and data sources for risk-on / risk-off regime classification
Smart Contract Risks
A field guide to how DeFi vaults get exploited — chronological case studies for protocol developers
peaq Announces Partnership with Robot Money
peaq announced a partnership with Robot Money to bring autonomous treasury management to 3.3M+ machines on the peaq network. The first robot — a Hong Kong robo-farm — has already invested.
The Institute for Zero-Human Companies and Generative Ventures Are Building Robot Money
Autonomous Treasury Management for the Agent Economy. A partnership between ZHC Institute and Lex Sokolin's Generative Ventures to build a managed vault for AI agents.
Legal Disclaimer
Important legal disclaimers — read carefully before interacting with the Robot Money protocol