Weekly brief · week of 14 August 2023

Week of 14 August 2023: Bitcoin falls 10.63% and Ether 8.45% in a single week

When both assets drop more than 8% in one week after weeks of quiet, what does the timing of the fall tell us about the market's structure, and what does it leave unanswered?

BTC fell from 29,303.85 to 26,189.99 (-10.63%) and ETH from 1,840.73 to 1,685.24 (-8.45%). Both weekly lows arrived at 21:00 UTC on Thursday 17 August, in the same hour as the week's largest hourly move (-5.15% for BTC, -5.73% for ETH). Volume ran above the prior twelve-week average for both.

What the data shows. BTC's high of 29,695.32 came Monday 14 August at 16:00 UTC, 1.34% above the week's open of 29,303.85, and the low of 25,166.00 arrived Thursday 17 August at 21:00 UTC, a range of 18.00% from low to high. Thursday was also the worst day for both assets (BTC -7.33%, ETH -6.99%). The largest single hourly move fell in that same Thursday hour. The four weeks before had been mostly flat to down for BTC (-0.49%, -2.67%, -0.66%, +0.74%), so the drop followed a quiet stretch. Both assets moved closely together hour by hour (correlation 0.94), but BTC fell further, lifting the ETH/BTC ratio by 2.44%. Volume was 1.17 times BTC's and 1.14 times ETH's prior twelve-week averages. At the close, BTC sat 17.14% below its 52-week high and printed in the lower third of its weekly range (22.6%); ETH sat 20.81% below its high, closing in the middle third (44.1%).

What it does not show. The candles do not say what moved the market or who was selling. Nothing here measures whether the Thursday selling was forced or whether the low was final. Hourly correlation tells us the two assets fell together, not why either fell.

What a careful desk would check. Realised volatility, the size of the swings that actually happened scaled to a yearly figure, reached 49.0% for BTC and 62.1% for ETH. The prior four weekly returns for BTC stayed between -2.67% and +0.74%, so a position sized on those weeks would face Thursday's -5.15% hourly move without having been built for it. A desk would compare stop distances and margin buffers with that hour rather than with the average week, and check whether its exits could have been filled inside it.

Funding. BTC funding averaged 0.0058% per eight hours and ETH 0.0028%, both close to zero, with a low of -0.0175% for ETH. Near-zero funding means neither side was paying a meaningful standing cost; the ETH low reading points to a period where shorts briefly paid longs.

What would change this reading

  • If BTC trades below Thursday's low of 25,166.00 in a later week, the drop reads as the start of a move rather than a single shock.
  • If BTC recovers towards its weekly open of 29,303.85 while volume returns to its prior twelve-week average, the week reads as a liquidity event that was absorbed.
  • If the ETH/BTC ratio keeps rising after a 2.44% weekly gain, the relative weakness is shifting from Ether to Bitcoin and any hedge needs to be rebuilt.
  • If realised volatility stays near 49.0% for BTC, limits calibrated on the prior four flat weeks remain too tight.

Terms

realised volatility
how much the price actually swung over a period, expressed as a yearly percentage
ETH/BTC ratio
the price of one Ether measured in Bitcoin; it rises when Ether outperforms Bitcoin
perpetual funding
a regular payment between long and short holders of a perpetual futures contract that keeps its price close to the spot price

Research notes on real market weeks, not investment advice. Sources: Binance spot BTCUSDT and ETHUSDT 1-hour candles and USD-M perpetual funding rates; weekly figures computed by WSTiger.

RISCON Labsrisconlabs.com