Question

Will zero fee perp venues take held positions from Hyperliquid, or only turnover?

This matters because open interest is where funding is paid and where a venue’s depth is tested. A venue that wins volume but not open interest is winning the trades that cost nothing to make, and that is a different business from the one Hyperliquid runs.

Observation

From the SwanBluechip venue snapshot of 2026-10-16 14:00 UTC, six venues, open interest as the sum of position size times mark price.

• Bybit: open interest 12.14 billion dollars, 47.4 percent of the tracked total; 24 hour volume 13.24 billion, 53.5 percent; 798 markets; taker 5.5 basis points, maker 2.
• Hyperliquid: 11.75 billion, 45.9 percent; volume 5.79 billion, 23.4 percent; 178 markets; taker 4.5, maker 1.5 at the base tier.
• Variational: 1.02 billion, 4.0 percent; volume 3.03 billion, 12.2 percent; 575 markets; taker 0, maker 0.
• Lighter: 0.54 billion, 2.1 percent; volume 1.75 billion, 7.1 percent; 213 markets; taker 0, maker 0.
• Extended: 0.09 billion, 0.4 percent; volume 0.57 billion, 2.3 percent; 329 markets; taker 2.5, maker 0.
• Pacifica: 0.06 billion, 0.2 percent; volume 0.38 billion, 1.5 percent; 77 markets; taker 4, maker 1.5.
• Total: 25.6 billion dollars of open interest and 24.7 billion dollars of 24 hour volume.

The two venues with zero fees on both sides held 6.1 percent of open interest and produced 19.3 percent of volume. Hyperliquid turned over about half of its open interest in a day; Variational turned over three times its open interest.

Hypothesis

The prediction, stated so it can be checked: on 2027-04-16, in the SwanBluechip venue snapshot nearest 14:00 UTC, Hyperliquid’s share of open interest across the same six venues (Bybit, Hyperliquid, Variational, Lighter, Extended, Pacifica) will be at least 40 percent.

If a venue is removed from tracking, the share is computed across the remaining venues. If a new venue is added, it is excluded from the check so that the denominator matches the one used here.

Why

Open interest is a held position, and a held position needs depth to enter and exit at size, funding that is not punitive and a venue that is expected to exist next week. Zero fees reduce the cost of a trade to the spread, which attracts turnover, but turnover does not require depth. The snapshot shows exactly that pattern: zero fee venues with a volume share three times their open interest share.

A second reason is that Hyperliquid’s base tier fee of 4.5 basis points is already close to Bybit’s, and lower with volume, so fee competition against it is a fight over a few basis points, not an order of magnitude.

Supporting Evidence

• Open interest and volume shares above, from https://swanbluechip.com/api/venues/summary at 2026-10-16 14:00 UTC.
• Fee schedules as recorded on the same snapshot, with each venue’s published source.
• The ratio of volume share to open interest share of about three for the zero fee venues, against about one half for Hyperliquid.

Counter Arguments

The strongest case against is that points programmes change the calculation. A venue that pays points for held positions, not just for volume, can rent open interest for as long as the programme runs, and six months is within the life of a programme.

A second objection is that the six venue set is small and includes one centralised exchange. If a large centralised venue is added or Bybit is removed, the share moves for reasons that have nothing to do with the question. The method above handles additions and removals, but not a change in what counts as tracked open interest.

A third objection is that six months is long in this market, and one venue failure or one listing of a very large market can move ten points of share in a week.

Current Conclusion

The prediction stands at 0.60. The 40 percent threshold leaves about six points of room below the current reading, which is a real move but not a collapse. The check is one JSON snapshot on one date.

Unknowns

• Whether any tracked venue is paying points for held open interest rather than volume, and at what rate.
• How the snapshot computes open interest for venues that report it differently.
• Whether the venue set will change before the check date.

What Would Change My Mind

• Confidence down before the date: Hyperliquid’s share below 42 percent for 30 consecutive daily snapshots.
• Confidence up before the date: a zero fee venue’s volume share falling while its open interest share stays flat, which would show that turnover left without positions following.
• Resolved: the reading on 2027-04-16, recorded in this document’s changelog with the snapshot time.

Next Research

• Record Hyperliquid’s open interest share monthly in the Observation section from the same endpoint.
• Add the points programme terms of each venue, from SwanBluechip’s points page, and note which pay for held positions.
• Write the resolution on 2027-04-16 whatever it is.

Implications

For traders, a venue’s volume share is not a measure of where positions are held, and depth should be checked at the size to be traded, not inferred from volume.

For venue builders, winning open interest requires depth and credibility that zero fees do not buy, so the fee is a weak lever against an incumbent with similar fees.

For Kaikim Labs, this is the first dated prediction with a written check. It is the pattern in Instrument, Experiment, Institution, Measurement, One Research Loop for Several Products applied to the instrument’s own data.

Related Research

• The Stablecoin Deposit Premium Over Treasury Funds Is Mostly Reward Tokens
• Instrument, Experiment, Institution, Measurement, One Research Loop for Several Products

References

1. SwanBluechip venue snapshot JSON: https://swanbluechip.com/api/venues/summary
2. SwanBluechip venues compared: https://swanbluechip.com/venues/
3. SwanBluechip points programmes: https://swanbluechip.com/venues/points/