Measuring Executable Net Carry in Cross-Venue Perpetual Funding-Rate Arbitrage

A quoted funding-rate differential is not, by itself, a measure of the cash flow available from a cross-venue position. This paper studies executable net carry as a measurement problem: expected funding must be placed on a robust multi-day basis, evaluated at a stated position size, and reduced by fees and depth-dependent entry and exit costs before routes can be compared. Using only the methodology and reported measurements in the humarb Methodology Pack, hereafter the humarb framework, the paper formalizes a route-level framework that combines contract-identity controls, seven-day medians of paired daily funding differences, separate order-book walks for entry and exit, book-specific freshness tests, and explicit treatment of unmeasured inputs. In the full-list snapshot published at 2026-11-01 19:19 UTC, 509 of 3,164 routes had a reference estimate; among those 509 routes, 462 (90.8%) had negative estimated net income after all measured costs at USD 5,000 per side over a seven-day hold. In the separate 2026-11-01 12:00–13:05 UTC window of 343 consecutive publishes, median net-positive route counts on the reference side were 71, 69, and 106 for the reported default, USD 5K, and USD 2K categories, respectively, whereas the conservative side reported zero for all tiers. These are model-based snapshot measurements, not realized trading outcomes. The evidence shows how cost and data-quality controls alter the economic interpretation of apparent funding spreads. Website: https://humarb.ai/