Skip to content
← Research

· Learned

Do LPs win when a stock is quoted in SPY?

Answer

Not for single stocks. On Robinhood Chain, all nine single-stock/SPY pools paid LPs less in fees than they lost to rebalancing (0.29 to 0.71 of LVR). The same stocks quoted in dollars did better. QQQ/SPY won.

  • 0.29–0.71fees ÷ LVR, single-stock/SPY pools
  • $419Mvolume in the nine SPY pools (Aug 3 to Nov 11)
  • $183K / $370Kfees to LPs / estimated LVR
  • 1.53QQQ/SPY: the pair moves together, so LPs win

How we measured

We took 23 pools (stocks quoted in SPY and in USDG) and read every swap from Robinhood Chain's public RPC. We used the fees actually paid, and estimated LVR from each pool's own price path in 1-hour buckets (Milionis et al., for concentrated liquidity in range).

What else we found

  • The same stocks quoted in USDG: 0.88 to 3.72, and each stock's best USDG pool beats its SPY pools.
  • At the same 0.05% fee, NVDA/SPY 0.66 against NVDA/USDG 1.15.
  • Quoting a single stock in SPY removes less of its variance than you'd hope. And SPY pools' fee tiers (0.025–0.05%) are too low for what's left.

What we take from it

So we don't say that quoting in SPY brings LPs.

What holds is narrower: it works for pairs that move together, or when fees are set to the variance that's left.

Limits

  • LVR is a model; there is no TVL history.
  • About 100 days (44 for the pools opened 2026-09-28); other DEXes excluded.
  • The pool list file was not kept: the pool ids are in fetch_all.sh, so a rerun needs it rebuilt.

Measured with an AI (Claude). It wrote and ran the scripts, and every number here is a line of the committed output.