Weekly brief · week of 24 September 2018

Week of 24 September 2018: Bitcoin slips 1.2% while Ether falls 4.74% on above-average volume

Both Bitcoin and Ether fell this week, Ether by 3.54 percentage points more: was this a market-wide drop or an Ether-specific one, and what can price and volume data alone show?

This matters to a desk or lender setting risk limits: ETH's weekly range (gap from high to low) was 20.92%, BTC's 7.74%. BTC fell 1.2% and ETH 4.74% this week. Both hit their lows on Tuesday 25 September and had their best day on Thursday 27 September. ETH's dollar volume (value traded) was 1.45 times its prior twelve-week average; BTC's was 0.84 times.

What the data shows. Both assets hit their weekly low on the same day: BTC at 6,325.02 at 18:00 UTC on Tuesday 25 September, ETH at 203.61 at 11:00 UTC that day. BTC then rose to its high of 6,814.80 at 07:00 UTC on Friday 28 September. ETH's high of 246.20 came earlier, at 02:00 UTC on Monday 24 September, before its low. Thursday 27 September was the best day for both (BTC +3.42%, ETH +6.88%) and held the week's largest one-hour moves, in the hour starting 19:00 UTC (2.18% for BTC, 4.61% for ETH). Hourly price changes tracked each other closely, with a correlation of 0.78 (1 would mean moving in lockstep). That is consistent with a market-wide move. ETH's weekly range was 20.92% against BTC's 7.74%. Its realised volatility was 99.2% against BTC's 40.9%; these figures measure the week's hourly price swings, scaled up to a yearly rate. Neither weekly return was extreme. BTC's ranked at the 44th percentile and ETH's at the 47th against 32 weeks of history, near the middle of that record.

What it does not show. The data records prices and volumes, not the reasons behind them. It contains no order books, positions or open interest. Dollar volume shows how much value changed hands, not who traded or whether selling was forced. ETH's turnover was 1.45 times its prior twelve-week average and BTC's 0.84 times. That gap does not by itself establish that investors moved money from one asset into the other. The funding-rate fields are empty for this week, so the data says nothing about funding, the payments between long and short holders of perpetual futures contracts.

What a careful desk would check. Close location, where the close sat inside the week's high-low range, was 61.6% for BTC, in the middle third, and 68.9% for ETH, in the upper third. Both closed well above their Tuesday lows. That describes this week only; one week cannot forecast the next. The ETH/BTC ratio, the price of one Ether in Bitcoin, fell 3.59% over the week, so ETH lost ground against BTC. A desk would compare ETH's 20.92% range with its stop distances, the price moves at which it closes a losing position. It would then ask whether positions sized on BTC's volatility would stay within limits in an ETH week of this width.

What would change this reading

  • If ETH keeps falling faster than BTC in later weeks, this week's 3.59% fall in the ETH/BTC ratio would suggest the start of a trend rather than a one-week divergence.
  • If BTC trades back below its weekly low of 6,325.02, this would suggest the Thursday recovery was a pause inside a broader decline.
  • If ETH volume falls back towards its prior twelve-week average while its price stays weak, this would suggest the heavier turnover this week was a one-off.

Terms

percentage points
the plain difference between two percentages; a fall of 4.74% is 3.54 points more than a fall of 1.2%
dollar volume
the total value of the asset traded during the week, measured in dollars
weekly range
the distance between the week's highest and lowest price, as a percentage
realised volatility
how much the price actually swung, measured here from the week's hourly prices and scaled up to a yearly percentage
percentile
a rank against past weeks; the 44th percentile means the week sits a little below the middle of the recorded weeks
correlation
a measure from -1 to 1 of how closely two prices move together; 1 means they move in lockstep
close location
where the week's closing price sat inside the week's high-to-low range, as a percentage
ETH/BTC ratio
the price of one Ether measured in Bitcoin, showing which of the two is gaining or losing against the other
funding
a regular payment between long and short holders of a perpetual futures contract, a futures contract with no expiry date
stop distance
how far the price may move against a position before the holder closes it to limit the loss

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