Week of 21 February 2022: an 18% range in BTC ends 1.79% lower
When a week travels far up and down but finishes near where it started, what does the path reveal about where the risk actually sat?
BTC fell from 38,386.89 to 37,699.07 (-1.79%) and ETH from 2,621.99 to 2,616.79 (-0.20%), despite ranges of 17.56% and 25.13% from low to high. Both assets set their low on Thursday 24 February and their high on Saturday 26 February, then gave back gains on Sunday. Dollar volume ran above the prior 12-week average for both.
What the data shows. Both assets bottomed in the same hour, 05:00 UTC on Thursday 24 February (BTC 34,322.28, ETH 2,300.00), and peaked in the same hour, 00:00 UTC on Saturday 26 February (BTC 40,348.45, ETH 2,878.00). The low came before the high, so most of the week's 17.56% (BTC) and 25.13% (ETH) range was a recovery, not a decline. Then Sunday 27 February reversed it: the worst day of the week (BTC -3.62%, ETH -5.85%) also carried the largest hourly move (BTC -3.36%, ETH -4.75%) in the hour starting 19:00 UTC, late in the UTC day. Both closes sat in the middle third of their weekly ranges. Volume was 1.25 times (BTC) and 1.31 times (ETH) the prior twelve-week average. The two moved closely (hourly correlation 0.92) and the ETH/BTC ratio rose 1.62%, so this was a shared move rather than one asset breaking away.
What it does not show. A small weekly return after a wide range says nothing about cause, and the data identifies no participants or flows. The Thursday low and the Saturday high sit two days apart in the week's sequence; the data cannot say whether Thursday's turn was a durable floor. Both assets remained far below their prior peaks (BTC -45.07%, ETH -46.01% versus their 52-week highs), and this week's return ranked at the 31st percentile for BTC and 45th for ETH across 210 weeks, so the finish was unremarkable even though the path was not.
What a careful desk would check. Realised volatility, the size of the swings that actually happened scaled to a yearly figure, reached 88.1% for BTC and 116.3% for ETH. Stops and margin buffers sized on calmer weeks would have been tested by the 17.56% and 25.13% ranges. Sunday's largest hourly move arriving late in the UTC day raises a specific question: whether positions held across that hour could have been exited at all, and whether overnight gaps between daily candles would have cleared a stop before the market reopened.
Funding. Perpetual funding, the periodic payment between long and short holders that keeps contract prices near spot, was effectively flat. BTC averaged 0.0003% per eight hours (range -0.016% to 0.01%) and ETH 0.0035% (range -0.014% to 0.01%). Neither side paid a persistent charge, so the week's swings were not accompanied by measurable crowding in funding.
What would change this reading
- If a later week closes below the Thursday low of 34,322.28 (BTC) or 2,300.00 (ETH), the mid-week recovery reads as a bounce inside a decline rather than a floor.
- If the pattern of a late-UTC largest hourly move repeats, the risk in this market sits in the hours when few desks are staffed.
- If funding moves materially away from zero in either direction, the flat-crowding reading from this week no longer holds.
- If ETH falls faster than BTC, the 1.62% gain in the ETH/BTC ratio reverses and the shared move becomes a relative one.
Terms
- realised volatility
- how much the price actually swung over a period, expressed as a yearly percentage
- funding rate
- the periodic payment between long and short holders of a perpetual futures contract; positive means longs pay shorts
- close location
- where the week's closing price sat between its own low and high, as a percentage