Quoting both sides through a 49.79% range without carrying the week's inventory
Should the desk widen its ETH spreads and halve posted size to keep inventory small, or quote tighter at full size to capture volume 1.77 times the prior 12-week average?
The market facts are real; the decision maker is illustrative.
The week and the position
In the week of 7 November 2022, ETH fell 22.11%, from 1,568.25 to 1,221.49, with a 49.79% range and 155.4% annualised realised volatility. The worst day was -17.38% on Wednesday, then the best day was +17.82% on Thursday, which also carried the largest 1-hour move of 7.0% mid-day. Trading volume ran 1.77 times the prior 12-week average, and the week closed in the lower third of its range, 74.37% below the 52-week high. Perpetual funding averaged -0.0117% per 8 hours, so holders of long positions were paid.
Judgment
A desk that must hold small inventory is exposed here by the reversal, not by the trend: an unhedged long bought near Wednesday's low of 1,073.53 faced a 17.82% single-day move the next session, and the 7.0% hour can strand a quote before a hedge lands. Widen spreads and halve posted size through the week, and skew any residual inventory towards the long side since negative funding pays holders of long positions. Accept the forgone volume; the desk's product is a quote that survives, not a fill rate.
What would change it
If realised volatility and hourly range contract towards the prior four weeks' pattern, and funding returns near zero, the desk can tighten spreads and restore size. If funding turns positive, the long-side skew should be dropped immediately.
What sets this case apart
The volume ran 1.77 times the prior 12-week average, yet the desk's binding constraint is inventory: a 17.82% best day and a 7.0% hour can strand a quote mid-fill.