Week of 20 July 2020: Ether rises 30.06% while Bitcoin adds 7.85%
When one asset gains four times as much as the other in the same week, what does the gap say about their relationship, and what can the week's data not settle?
BTC rose from 9,208.99 to 9,931.54 (+7.85%), the 81st percentile of 127 recorded weeks. ETH rose from 239.12 to 311.00 (+30.06%), the 98th percentile, outperforming by 22.21 percentage points. Hourly correlation was 0.68, loose by this desk's standard, and funding stayed positive for both.
What the data shows. BTC rose from 9,208.99 to 9,931.54 (+7.85%), the 81st percentile of the 127 weeks of history recorded. ETH rose from 239.12 to 311.00 (+30.06%), the 98th percentile of the same history, beating BTC by 22.21 percentage points and lifting the ETH/BTC ratio 20.6%. Both assets set their low before their high, on Monday 20 July: BTC 9,131.00 at 10:00 UTC, ETH 234.05 at 19:00 UTC. Both peaked on Sunday 26 July (BTC 10,111.00, ETH 318.75) and closed in the upper third of their range (81.7% and 90.9%). The largest hourly moves differed in size and timing: +6.07% for ETH at 22:00 UTC on Wednesday, +1.68% for BTC on Sunday. Volume tells the same split story: ETH's $1.68bn was 1.66 times its prior twelve-week average, while BTC's $3.30bn was 0.8 times its own.
What it does not show. Hourly correlation was 0.68, below this desk's 0.7 threshold for moving together. That measures co-movement, not cause: nothing here explains why ETH pulled ahead, who was buying, or whether the gap persists. ETH finished 1.65% below its 52-week high; BTC finished 18.73% below its own, so the two were at very different distances from their prior peaks.
What a careful desk would check. Realised volatility, the size of the swings that actually happened scaled to a yearly figure, was 80.0% for ETH against 28.9% for BTC. A 30.06% week at the 98th percentile invites a reversion question, and the 6.07% ETH hour shows how fast that could run. Sizing for BTC-style volatility would leave an ETH position several times too large.
Funding. Longs paid shorts all week: ETH funding averaged 0.0365% per eight hours (high 0.0944%), BTC 0.0126% (high 0.0654%). The cost of holding ETH longs was roughly three times BTC's.
What would change this reading
- If ETH gives back much of the 22.21-point gap over later weeks, this week reads as a single-asset divergence rather than a new relationship.
- If hourly correlation returns above 0.7 while both assets hold this week's closes, the loose reading was temporary.
- If ETH trades below its weekly low of 234.05, the 98th-percentile week becomes a local top in hindsight.
- If BTC closes a later week above 10,111.00, its 18.73% distance from the 52-week high narrows sharply.
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; below 0.7 is treated as loose
- funding rate
- the periodic payment between long and short holders of a perpetual futures contract; positive means longs pay shorts