Tokenized stock price accuracy, live onchain vs Nasdaq across 11 equities
Absolute deviation between each tokenized equity's Uniswap v4 pool price on Robinhood Chain and its real market reference price, in basis points, labeled by market session. The only live public measurement of whether onchain stocks actually track their underlying.
TL;DR. As of , NVDA leads price deviation at 0.067% (24h avg) on Tokenized stock price accuracy, live onchain vs Nasdaq across 11 equities. Source: OpenChainBench, https://openchainbench.com/benchmarks/tokenized-stock-peg.
Robinhood Chain tokenizes ~95 equities, and its Uniswap v4 pools trade them 24/7 while the Nasdaq closes every evening and every weekend. This page answers the question that setup begs. how far does the onchain price drift from the real one. Every 60 seconds we read each pool's spot price straight from the chain (StateView getSlot0, keyless) and compare it to the live reference price, labeling every sample with the market session it was taken in (pre, regular, post, closed). The regular-hours deviation is the headline ranking; the closed-state series is the weekend drift nobody else publishes: what AAPL is worth onchain on a Saturday when no market maker has a reference to arb against. Of the ~95 tokenized equities only 14 have pools with real liquidity; we measure the 11 with a listed underlying. SpaceX trades onchain too, but SpaceX is not listed anywhere, so there is no reference price to measure against, which says something about tokenized private equity all by itself.
Methodology
Every 60 seconds the harness reads the Uniswap v4 spot price of 11 tokenized equities on Robinhood Chain (one batched eth_call to the StateView contract, keyless) and fetches the reference equity price from Yahoo Finance (one batched spark call). Deviation is 10000 x |pool - reference| / reference, in basis points, published with a market_state label derived from Yahoo's holiday-aware session windows. No API keys, no transactions, both legs reproducible from public endpoints.
Frequently asked
Do tokenized stocks track their real market price?
During regular Nasdaq hours, mostly yes: arbitrageurs keep the Uniswap v4 pools within tens of basis points of the reference on the liquid names, and this page shows the live per-symbol number. Outside market hours the tether loosens, the pools keep trading around the clock with no authoritative reference, and the closed-session drift measured here is the honest answer to how far a 24/7 stock wanders from its 9:30-to-4 anchor.
What happens to tokenized stock prices on weekends?
The pools trade continuously, so the onchain price becomes pure supply and demand against the Friday close. This bench samples that drift every minute all weekend and publishes it as the closed-session series. Monday pre-market usually snaps it back; the interesting data is how far it got before that.
Why only 11 stocks when Robinhood tokenized about 95?
Because only 14 of the 95 have a USDG pool with real liquidity and swap activity, and 3 of those are unmeasurable: SpaceX has no listed reference price, and SNDK plus QQQ hold under $2k of depth. The other 80 tokens sit in placeholder pools at 90-95% fees that nobody has ever traded. The gap between tokens issued and markets that exist is one of this page's findings.
How is the onchain price read?
One batched keyless eth_call per minute to the StateView contract of the Uniswap v4 singleton on Robinhood Chain, decoding sqrtPriceX96 from getSlot0 for each pool and converting with the pool's currency ordering and the 6-vs-18 decimals gap. No indexer, no API key, reproducible with curl.
Where does the reference price come from?
Yahoo Finance's public chart API, one batched call per minute with the session windows read from currentTradingPeriod, which is holiday-aware. When the market is closed the reference is the last regular close, which is exactly the baseline the weekend drift should be measured against.
Source code github.com/ChainBench/OpenChainBench/tree/main/harnesses/tokenized-stock-peg