Weekly brief · week of 19 February 2018

Week of 19 February 2018: Bitcoin and Ether each fall about 7.6% after a higher prior week

When two assets fall by almost identical amounts in the same week, what does the shared shape of the move tell us, and what can a single week of history not establish?

BTC fell from 10,375.01 to 9,590.00 (-7.57%) and ETH from 911.00 to 841.19 (-7.66%). Both highs came on Tuesday 20 February and both lows after them. The gap between the two weekly returns was 0.09 percentage points, and hourly correlation was 0.89.

What the data shows. BTC opened at 10,375.01 and closed at 9,590.00 (-7.57%); ETH opened at 911.00 and closed at 841.19 (-7.66%). Both moved in the same direction, and the difference between their returns was 0.09 percentage points. The peaks came on Tuesday 20 February: BTC reached 11,786.01 at 22:00 UTC, late in the UTC day and 13.6% above its open; ETH reached 961.00 at 07:00 UTC, early in the UTC day and 5.49% above its open. In both cases the low followed the high — BTC's 9,274.80 on Sunday 25 February at 18:00 UTC, ETH's 785.99 on Friday 23 February at 01:00 UTC. Weekly ranges were 27.08% and 22.27%. The largest hourly move was -3.67% for BTC and -3.41% for ETH, in the same hour beginning 12:00 UTC on Thursday 22 February. Hourly correlation was 0.89, and both assets closed in the lower third of their weekly range. BTC finished 18.20% below its 52-week high, ETH 27.43% below.

What it does not show. The record contains one week, so there is no percentile rank, no comparison of dollar volume (3.66bn for BTC, 0.71bn for ETH) against prior weeks, and no funding data. Prices record how far and how fast the market moved, not who traded or why. Candles cannot show whether the Sunday close was a turning point or a pause.

What a careful desk would check. Realised volatility, the size of the actual swings scaled to a yearly figure, was 106.4% for BTC and 95.8% for ETH. A desk would compare its stop distances with the -3.67% hour on Thursday 22 February rather than with an average week, and ask whether exits inside that hour could have been filled. It would also note that the largest hourly move fell in the middle of the UTC day, while both weekly extremes came late in the day for BTC and early for ETH.

The prior week. The input lists one prior week for each asset, marked as up, with a return of 28.77% for BTC and 12.67% for ETH. This week's decline gives back part of that gain, but one listed prior week is not a trend.

What would change this reading

  • If a later week trades below BTC's 9,274.80 or ETH's 785.99, the Sunday and Friday lows read as levels that failed rather than as support.
  • If a later week rises while the ETH/BTC ratio moves by more than the 0.10% change seen here, the two assets have stopped moving as one block.
  • If more weeks of history are recorded and this -7.57% week ranks low against them, the decline belongs to a broader slide rather than to a pullback after a 28.77% week.

Terms

realised volatility
how much the price actually swung over a period, expressed as a yearly percentage
hourly correlation
how closely two prices moved together hour by hour, from -1 to 1
close location
where the weekly close fell within the week's high-to-low range

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