Week of 26 September 2022: Bitcoin gains 1.32% while Ether loses 1.38%
When Bitcoin and Ether still move closely hour by hour but end the week in opposite directions, what explains the divergence and what can price data alone not tell us?
BTC rose from 18,809.13 to 19,056.8 (+1.32%) while ETH fell from 1,294.62 to 1,276.72 (-1.38%), a gap of 2.70 percentage points. Hourly correlation stayed close at 0.89, and the ETH/BTC ratio fell 2.66%. Both closed in the lower third of their weekly ranges.
What the data shows. Both assets peaked at the same hour, 12:00 UTC on Tuesday 27 September (BTC 20,385.86, ETH 1,400.0), then sold off. BTC's low came at 02:00 UTC on Wednesday 28 September (18,471.28) and ETH's eight hours later (1,253.2), so the low came after the high for both: the week opened near its low, rallied, and gave back most of it. BTC's close sat 30.6% into its 10.37% range, ETH's only 16.0% into its 11.71% range — both in the lower third. The divergence is a volume story as well: BTC dollar volume was 1.70 times its prior twelve-week average, while ETH's was 0.62 times, i.e. below its own average. Weekly returns ranked at the 51st percentile for BTC and 42nd for ETH out of 241 weeks, so neither was unusual in isolation.
What it does not show. The data records where prices went, not why the two assets separated by 2.70 percentage points in a week. An hourly correlation of 0.89 says they moved together most hours; it does not say which asset led, who traded, or whether the gap reflects different holder bases. The weekly close near the bottom of the range does not indicate the direction of the following week.
What a careful desk would check. Realised volatility was similar for both (63.5% annualised for BTC, 64.2% for ETH), so neither was materially wilder than the other over the week, though similar volatility does not by itself explain the divergence in returns. Both remained deep below their 52-week highs (-72.23% BTC, -73.66% ETH). A desk comparing hedges would note that an ETH/BTC ratio that fell 2.66% in one week makes a BTC-denominated ETH position the risk, and would ask whether the prior four weeks (up twice, down twice for each asset) still describe the current regime.
What would change this reading
- If ETH continues to fall faster than BTC in later weeks, the 2.70-point gap reads as the start of a trend rather than a one-week rotation.
- If BTC volume returns towards its prior twelve-week average while ETH stays below 0.62 times, the divergence was driven by selling in one book rather than buying in the other.
- If either asset trades back above this week's high — 20,385.86 for BTC or 1,400.0 for ETH — the lower-third closes read as a pause, not a rejection.
- If funding, near zero at 0.0057% (BTC) and 0.003% (ETH) per eight hours, moves sharply in one direction, positioning rather than spot flow may be driving the pair.
Terms
- hourly correlation
- how closely two prices moved together hour by hour, from -1 to 1; 0.89 means they usually moved in the same direction
- ETH/BTC ratio
- the price of one Ether expressed in Bitcoin, showing whether ETH is gaining or losing ground against BTC
- realised volatility
- how much the price actually swung over the period, expressed as a yearly percentage