A 12.35% day and a 3.31% hour test a market maker's inventory limit
Should the desk tighten its quotes to capture 2.18 times normal volume, or widen and skew them so inventory stays small through hours like Thursday's 3.31% move?
The market facts are real; the decision maker is illustrative.
The week and the position
In the week of 6 November 2023, ETH rose 8.09% to close 4.11% below its 52-week high, with a 15.49% range. Thursday gave the best day of 12.35% and the week's largest 1-hour move, 3.31%, in the middle of the UTC day; Friday then fell 2.06%. Dollar volume reached 6.43bn, 2.18 times the prior 12-week average, and 8-hour funding averaged 0.0175%. ETH beat BTC by 2.23 points at a loose 0.51 hourly correlation. An illustrative desk quotes both sides of ETH and must keep inventory small.
Judgment
Inventory is exposed on the hour a 3.31% move lands, not on the week's close. Keep both sides quoted, but widen the spread and skew quotes towards reducing accumulated inventory, and lower the per-hour size limit ahead of the middle-of-the-UTC-day window where this week's largest move occurred. Accept a lower fill share and thinner capture. With volume 2.18 times its 12-week average, the missed spread is smaller than the cost of holding a full inventory through such an hour.
What would change it
If 1-hour moves stay well under 1% for two consecutive weeks while weekly dollar volume remains near 6.43bn, spreads and per-hour size limits can be restored. A week closing in the lower third of its range would reverse the quote skew.
What sets this case apart
This is inventory timing inside a rising week: the binding risk is the size of one hour, not the direction of the week.