A 116.7% realised-vol week and an -8.87% hour test a desk that sold volatility
Should the desk cut the size of its sold-volatility exposure after a 13.82% down week, or hedge it dynamically and keep collecting the premium?
The market facts are real; the decision maker is illustrative.
The week and the position
In the week of 6 September 2021, ETH fell 13.82% from 3,950.28 to 3,404.21, inside a 32.11% range, with annualised realised volatility of 116.7%. The worst hour lost 8.87% on Tuesday at 14:00 UTC, the worst day fell 12.53%, and the best day gained 4.2% on Sunday, lifting the close to the middle third of the week range. Volume ran 1.23 times the prior 12-week average. An illustrative options desk is short gamma and short vega, with funding still positive at 0.0217% per eight hours and ETH trailing BTC by 2.75 points.
Judgment
A short-gamma book is hurt by exactly the realised path already recorded, and Sunday's 4.2% bounce is where the premium looks most attractive to sell and least safe to keep. Reduce the sold-volatility size to a level that a repeat of Tuesday's -8.87% hour, not the weekly average, can absorb in one session. Hedge the tail with defined-cost protection rather than a full delta hedge, which would sell ETH near the middle third of the range and lock in the loss. Accept paying for that protection out of the remaining premium.
What would change it
If eight-hour funding drops back towards its 0.01% minimum or turns negative while dollar volume falls below the prior 12-week average, the conditions recorded alongside this week's move are gone and the original size can be restored. If the ETH/BTC ratio stops falling from its -3.09% week, the idiosyncratic leg of the hedge can be dropped.
What sets this case apart
The decision is about how much short-gamma exposure to keep after realised volatility has already printed, not about predicting direction; the hedge cost is set by Tuesday's hour, not by the -13.82% weekly return.