Week of 2 September 2019: BTC and ETH rise after a run of down weeks, on light volume
When two assets rise together after several falling weeks but trade on below-average volume, what does the week actually establish, and what would confirm or weaken it later?
BTC rose from 9,723.59 to 10,381.18 (+6.76%) and ETH from 170.73 to 181.19 (+6.13%), after three down weeks for BTC and four for ETH. Dollar volume was 0.74x the prior twelve-week average for BTC and 0.62x for ETH. Both assets set their largest hourly move in the same Friday hour.
What the data shows. BTC rose from 9,723.59 to 10,381.18 (+6.76%) and ETH from 170.73 to 181.19 (+6.13%). BTC opened close to its weekly low of 9,712.50, which came in the opening hour on Monday 2 September, and its best day was that same day (+6.34%). BTC reached its high of 10,905.87 at 12:00 UTC on Friday 6 September, then retraced; the week's largest hourly move, -3.92%, arrived five hours later at 17:00 UTC, the hour in which ETH also posted its largest hourly move (-2.61%) and both assets logged their worst day (BTC -2.5%, ETH -2.68%). ETH's low of 165.00 came later, at 20:00 UTC on Friday, after which it recovered to a weekly high of 184.18 at 03:00 UTC on Sunday 8 September and closed in the upper third of its range. The prior four weeks were down for BTC in three of four (-10.68%, -1.59%, -4.12%) and down in all four for ETH. Trading was light: BTC dollar volume of 2.94bn was 0.74 times its prior twelve-week average, and ETH's 0.35bn was 0.62 times.
What it does not show. A rising week after a run of falling ones does not establish that the direction has changed; one week is one observation. Volume below the twelve-week average means fewer dollars changed hands than in the recent past, but the data does not say who was buying or whether the buying was new positioning or the closing of earlier shorts. Both assets still sat well below their 52-week highs (BTC -25.64%, ETH -49.37%).
What a careful desk would check. Realised volatility, the size of the swings that actually happened scaled to a yearly figure, was 50.6% for BTC and 56.3% for ETH. BTC closed at 56% of its weekly range (middle third) while ETH closed at 84.4% (upper third), so ETH finished visibly stronger within its own week. The weekly return ranked at the 81st percentile for BTC and the 75th for ETH across 81 weeks of history — a modest sample against which a percentile should be read.
Ether against Bitcoin. Both moved in the same direction, but ETH lagged by 0.63 percentage points and the ETH/BTC ratio fell 0.6%. The two tracked each other closely, though not tightly (hourly correlation 0.73), which leaves room for the gap to widen in either direction in later weeks.
What would change this reading
- If a later week closes back below BTC's weekly open of 9,723.59, this week reads as a bounce inside the prior three-week decline rather than a turn.
- If volume returns towards the prior twelve-week average while price holds above 10,381.18, the light-volume qualifier on this rise weakens.
- If ETH's weekly return again trails BTC's and the ETH/BTC ratio keeps falling from its -0.6% change, the lag is a trend rather than a single week.
- If a later week trades below ETH's low of 165.00, Sunday's high of 184.18 marks a ceiling rather than a recovery.
Terms
- realised volatility
- how much the price actually swung over a period, expressed as a yearly percentage
- close location
- where the closing price fell within the week's high-to-low range, as a percentage
- hourly correlation
- how closely two prices moved together hour by hour, from -1 to 1