A once-a-day converter meets a 26.32% day and a 12.04% day in the same week
Should a payments company keep converting customer receipts into ETH once a day at a fixed time, or lengthen the interval to batch receipts across several days while volatility is this high?
The market facts are real; the decision maker is illustrative.
The week and the position
In the week of 24 May 2021, ETH rose 13.80%, from 2,096.63 to 2,385.98, inside a 39.91% range (low 2,079.94 early Monday, high 2,910.00 on Wednesday at 08:00 UTC). Monday gained 26.32% and Friday lost 12.04%; the largest one-hour move was 5.22% on Friday starting 12:00 UTC. Realised volatility ran at 175.6% annualised and volume was 1.53 times the prior 12-week average. An illustrative payments company holds receipts in USD and converts them into ETH once a day at a fixed time, and the asset sat 44.93% below its 52-week high.
Judgment
Keep the once-a-day conversion at a fixed time and in fixed amounts, and do not lengthen the interval to batch receipts. Within this week, a conversion delayed from Monday to Friday traded a 26.32% gain for a 12.04% loss, so discretion about timing adds variance without adding an edge. What must change is the limit on one day's receipts: size each conversion so a 5.22% hourly move against it, like Friday's, does not force repricing what customers were quoted. Accept that individual conversions will sometimes land near the week's extremes.
What would change it
If realised volatility and the largest hourly move fall back to well below this week's levels for several consecutive weeks, batching two or more days of receipts becomes defensible. If a single hour exceeds 5.22% or the weekly range exceeds 39.91%, add a second daily conversion window rather than adding discretion over timing.
What sets this case apart
The company holds no market view; the decision is whether discretion over conversion timing helps when one day gained 26.32% and the following Friday lost 12.04%.