Historical study
Does market making pay, once the bootstrap can run?
Not identified. Two fill models land on opposite sides of zero and neither interval touches it.
Assume a fill whenever the price touches the quote and the answer is a loss between -12,121 and -2,413 dollars a day. Assume a fill only against a print that really happened and it is a profit between 881 and 5,889. Both intervals sit well clear of zero, in opposite directions. The difference is not noise, it is the assumption, and the assumption cannot be checked from outside the exchange.
Data · 5,413,998 book snapshots · 468 tokens · 5 days · 2026-07-30 to 2026-08-03
| Touch model, dailyassumes a fill whenever the price touches our quote | [-12121.02, -2412.7] USD per day |
|---|---|
| Tape model, dailyassumes a fill only against a print that actually happened | [881.05, 5888.98] USD per day |
| Fills, touch | 12,923 |
| Fills, tape | 18,686 |
Next test · planned
Do conservative queue and execution scenarios agree on net portfolio PnL in prospective event holdouts with capital and inventory carried forward?
Research renewal planPayload snapshot 2026-10-12. These figures are frozen; the interactive page reads the same file.
Page export dates do not extend the observation period shown above.
Full report (docs/research/mm_pnl_stream-5tage.md) · Source module (src/mm_pnl.py)
Research tool only: no investment advice, no order placement, no venue affiliation. Public Polymarket and Kalshi data, provided as-is.