Week of 1 September 2025: Bitcoin gains 2.67% while Ether falls 1.95%
When Bitcoin and Ether move in opposite directions over a week, what does the split reveal about where pressure sat, and what can it not settle about the weeks ahead?
BTC rose from 108,246.36 to 111,137.34 (+2.67%), ending three straight down weeks, while ETH fell from 4,391.83 to 4,306.19 (-1.95%). The ETH/BTC ratio dropped 4.5%. Volume for both ran below their prior 12-week averages, and funding stayed near zero.
What the data shows. BTC set its weekly low of 107,255.00 at 05:00 UTC on Monday 1 September, 0.92% below the week's open of 108,246.36, then climbed to 113,384.62 on Friday 5 September before closing at 111,137.34. The close sat in the middle third of the weekly range (63.3%). This reversed a weak run: the four prior weeks for BTC were 4.45%, -1.58%, -3.33% and -4.62%. ETH did the opposite, setting its high of 4,491.84 on Monday at 08:00 UTC, in the middle of the UTC day, and closing lower than it opened. The two assets still moved together hour by hour (correlation 0.84), but Ether's weekly loss against Bitcoin's gain left the ETH/BTC ratio down 4.5%. Both printed their largest one-hour move in the same hour, 14:00 UTC on Friday 5 September: -2.22% for BTC and -3.81% for ETH.
What it does not show. The data records prices and volumes, not causes. Nothing here identifies who was selling Ether or whether the divergence reflects position shifts, flows or rebalancing. Volume for both assets came in below their prior 12-week averages (0.83x for BTC, 0.84x for ETH), so the split happened in a thinner tape than usual, which can exaggerate moves without implying conviction. The data also says nothing about what follows.
What a careful desk would check. BTC sits 10.26% below its 52-week high and ETH 12.74% below its own. Realised volatility, the size of actual swings scaled to a yearly figure, was 37.4% for BTC and 58.5% for ETH. A desk would ask whether this week marks a genuine rotation from ETH back to BTC, or one week of noise, and would want several more weeks of the ratio before treating it as a trend. The Friday 14:00 UTC hour, where both markets fell sharply at once, shows the pair can still drop together even in a week when their totals diverged.
What would change this reading
- If the ETH/BTC ratio keeps falling in later weeks, this week's divergence reads as the start of a rotation rather than a one-week fluke.
- If BTC holds above its weekly open of 108,246.36, the three prior down weeks look absorbed; if it breaks the weekly low of 107,255.00, the recovery reads as a pause.
- If volume returns towards or above the prior 12-week average while the split persists, the divergence gains weight it did not have on thin volume.
- If funding moves further from zero in either direction, the cost of carrying positions starts to matter more than it did this 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
- ETH/BTC ratio
- the price of Ether measured in Bitcoin; it rises when Ether outperforms Bitcoin and falls when it lags