Discoveries: what volume actually costs
Every run on Hyperider measures something no exchange publishes about itself: the all-in cost of volume on a real account — venue fees, spread, and funding, per venue mix and asset. This page is those measurements, aggregated anonymously across riders and republished automatically as new runs land. Not modelled. Not marketed. Measured.
BTC on Perpl — earned $0.45 per $10,000 of volume across $798 measured.
- P1BTC · Perplearns $0.45 per $10ksolo ride2 runs · $798 measurednot venue-checked yet
- P2BTC · Arcus × Robinhood Lighterearns $0.35 per $10khedged pair11 runs · $3,299 measurednot venue-checked yet
- P3BTC · BULK≈ $0 per $10ksolo ride3 runs · $3,198 measurednot venue-checked yet
- P4BTC · Robinhood Lighter≈ $0 per $10kconvoy1 run · $300 measurednot venue-checked yet
- P5BTC · Arcus × N1$1.12 per $10khedged pair7 runs · $77,174 measured25% rests filled✓ 87% venue-checked
- P6BTC · N1$1.49 per $10ksolo ride4 runs · $4,293 measurednot venue-checked yet
- P7BTC · Arcus$2.03 per $10ksolo ride8 runs · $24,423 measurednot venue-checked yet
- P8BTC · Arcus × Robinhood Lighter$2.61 per $10khedged pair2 runs · $24,691 measurednot venue-checked yet
- P9BTC · Arcus × Robinhood Lighter$2.69 per $10kconvoy11 runs · $5,807 measurednot venue-checked yet
- P10SPY · Arcus$2.69 per $10ksolo ride3 runs · $17,700 measured✓ 11% venue-checked
- P11QQQ · Arcus × Robinhood Lighter$2.86 per $10khedged pair1 run · $6,399 measured✓ 100% venue-checked
- P12AAPL · Arcus$3.64 per $10ksolo ride1 run · $300 measurednot venue-checked yet
the full table
| venue mix | asset | surface | runs | volume measured | net | per $10k | rests filled | venue-checked |
|---|---|---|---|---|---|---|---|---|
| Perpl | BTC | solo ride | 2 | $798 | +$0.04 | earns $0.45 | — | not yet |
| Arcus × Robinhood Lighter | BTC | hedged pair | 11 | $3,299 | +$0.11 | earns $0.35 | — | not yet |
| BULK | BTC | solo ride | 3 | $3,198 | +$0.00 | ≈ $0 | — | not yet |
| Robinhood Lighter | BTC | convoy | 1 | $300 | +$0.00 | ≈ $0 | — | not yet |
| Arcus × N1 | BTC | hedged pair | 7 | $77,174 | −$8.62 | $1.12 | 25% | ✓ 87% |
| N1 | BTC | solo ride | 4 | $4,293 | −$0.64 | $1.49 | — | not yet |
| Arcus | BTC | solo ride | 8 | $24,423 | −$4.96 | $2.03 | — | not yet |
| Arcus × Robinhood Lighter | BTC | hedged pair | 2 | $24,691 | −$6.46 | $2.61 | — | not yet |
| Arcus × Robinhood Lighter | BTC | convoy | 11 | $5,807 | −$1.56 | $2.69 | — | not yet |
| Arcus | SPY | solo ride | 3 | $17,700 | −$4.77 | $2.69 | — | ✓ 11% |
| Arcus × Robinhood Lighter | QQQ | hedged pair | 1 | $6,399 | −$1.83 | $2.86 | — | ✓ 100% |
| Arcus | AAPL | solo ride | 1 | $300 | −$0.11 | $3.64 | — | not yet |
| Arcus × Robinhood Lighter | HYPE | hedged pair | 9 | $4,606 | −$1.98 | $4.30 | — | not yet |
| Hyperliquid | IO:OAI | solo ride | 3 | $2,316 | −$1.00 | $4.30 | — | not yet |
| Arcus × Robinhood Lighter | QQQ | hedged pair | 2 | $28,032 | −$12.13 | $4.33 | — | not yet |
| Arcus × Robinhood Lighter | HYPE | hedged pair | 2 | $1,000 | −$0.81 | $8.11 | 0% | not yet |
| Arcus × Robinhood Lighter | CRCL | hedged pair | 4 | $7,150 | −$9.54 | $13.34 | 27% | not yet |
| Arcus | CASHCAT | solo ride | 2 | $1,132 | −$3.91 | $34.59 | — | not yet |
| Arcus × BULK | PUMP | hedged pair | 1 | $10,143 | −$37.86 | $37.32 | — | not yet |
The findings, in plain words
- BTC on Perpl (solo ride) EARNED $0.45 per $10,000 of volume across $798 measured over 2 runs — funding capture exceeded every trading cost.
- BTC on Arcus × Robinhood Lighter (hedged pair) EARNED $0.35 per $10,000 of volume across $3,299 measured over 11 runs — funding capture exceeded every trading cost.
- BTC on BULK (solo ride) broke exactly even across $3,198 measured over 3 runs — the volume cost ≈ $0 all-in.
- BTC on Robinhood Lighter (convoy) broke exactly even across $300 measured over 1 run — the volume cost ≈ $0 all-in.
- BTC on Arcus × N1 (hedged pair) cost $1.12 per $10,000 of volume across $77,174 measured over 7 runs, all-in — fees, spread and funding included. 25% of its order crossings filled as resting maker orders (≈ $0 each). 87% of it is checked against the venues' own records.
- BTC on N1 (solo ride) cost $1.49 per $10,000 of volume across $4,293 measured over 4 runs, all-in — fees, spread and funding included.
- BTC on Arcus (solo ride) cost $2.03 per $10,000 of volume across $24,423 measured over 8 runs, all-in — fees, spread and funding included.
- BTC on Arcus × Robinhood Lighter (hedged pair) cost $2.61 per $10,000 of volume across $24,691 measured over 2 runs, all-in — fees, spread and funding included.
- BTC on Arcus × Robinhood Lighter (convoy) cost $2.69 per $10,000 of volume across $5,807 measured over 11 runs, all-in — fees, spread and funding included.
- SPY on Arcus (solo ride) cost $2.69 per $10,000 of volume across $17,700 measured over 3 runs, all-in — fees, spread and funding included. 11% of it is checked against the venues' own records.
How these numbers are made
Every run is delta-neutral: long on one venue, short on another, so price moves cancel and what remains is the honest economics — prices + funding − fees and spread = net, computed per venue leg from venue-confirmed fills. Orders rest as makers first wherever maker orders are free (a filled rest costs ≈ $0), and the rest-fill rate is measured, never assumed. Only volume whose outcome was actually measured counts toward a cost figure; unknown outcomes are excluded rather than guessed. “Robinhood Lighter” and “Lighter Mainnet” are tracked as the different venues they are. Aggregates are anonymous by construction — no names, no wallets, only totals.
Questions these numbers answer
What does delta-neutral volume farming actually cost?
It depends entirely on the venue mix and asset. Measured on real runs, the cheapest mixes on Hyperider run in the low single-digit dollars per $10,000 of volume — and when the funding gap between the two venues is fat enough, the position earns more than it costs. The table on this page shows the live measured numbers per mix.
Is volume on a zero-fee venue really free?
No. Crossing any order book also pays the spread, which on a thin market is larger than most fees. A '0% fee' venue can still cost several basis points per order. That is why every number here is all-in: venue fees plus spread plus funding, measured on real fills — never just the fee schedule.
How can a delta-neutral position earn money?
A hedged pair is long on one venue and short on another, so price moves cancel. Perp venues continuously pay funding between longs and shorts, and the two venues' rates differ — the pair collects that difference around the clock. When the collected funding exceeds the entry, exit and recycle costs, the volume was better than free.
What is maker-first resting, and why does it matter?
Instead of crossing the book immediately (paying the taker fee and the spread), an order first rests at the best price for around 20 seconds. If the market comes to it, that fill costs approximately $0 on venues with free maker orders. Hyperider measures how often rests actually fill per mix — the rest-fill rates on this page are measured, not assumed.
Where do these numbers come from?
From real runs by Hyperider riders, aggregated anonymously — no names, no wallets, only totals per venue mix and asset. Net is computed as price moves plus funding minus fees and spread, per venue leg, and only volume whose outcome was actually measured counts toward the cost figures. Wherever a run has been checked against the venues' own fills and funding records, the venues' number replaces the app's own count — the venue-checked column shows how much of each mix that covers. The page republishes itself as new runs land.