Week of 27 July 2020: Bitcoin gains 11% and Ether 20%, both reversing inside Sunday's high hour
When two assets post weekly returns at the 89th and 95th percentiles but close far below their highs, what does the week's final hours add to how those gains should be read?
BTC rose from 9,931.54 to 11,071.35 (+11.48%) and ETH from 311.02 to 372.42 (+19.74%). Both set weekly highs in the hour starting 04:00 UTC on Sunday 2 August, and that same hourly candle closed 6.92% (BTC) and 8.65% (ETH) below its open. Funding stayed positive all week, averaging 0.0663% per eight hours for BTC.
What the data shows. The week rose from its open. BTC's low of 9,917.21 came in the first hour on Monday 27 July, and its best day (+11.06%) was that same day. ETH's low of 305.86 arrived on Tuesday 28 July, and its best day (+11.75%) came on Saturday 1 August. Both assets then printed their weekly highs in the hour starting 04:00 UTC on Sunday 2 August — BTC 12,123.46, ETH 415.85 — and that same hourly candle carried the week's largest hourly move for both: -6.92% for BTC and -8.65% for ETH, so the high and the sharpest drop sit in one hour. Closes landed in the middle third of each weekly range (BTC at 52.3%, ETH at 60.5%). Dollar volume ran 1.81 times the prior twelve-week average for BTC and 3.09 times for ETH. The weekly returns ranked at the 89th percentile for BTC and the 95th for ETH against 128 prior weeks. ETH beat BTC by 8.26 percentage points, lifting the ETH/BTC ratio 7.41%.
What it does not show. The candles show how far and how fast the market moved, not why. Nothing here identifies who was buying through the week or who sold in Sunday's hour. The drop from the Sunday high does not establish whether those highs were final, and positive funding tells us longs paid shorts, not how large those positions were.
What a careful desk would check. Realised volatility, the size of the swings that actually happened scaled to a yearly figure, reached 87.8% for BTC and 116.3% for ETH. A week that gains double digits and still ends mid-range is a week in which entry and exit levels mattered more than direction. A desk would compare its stop distances and fills against the -6.92% and -8.65% Sunday hour, and against the fact that the largest moves for both assets clustered in the early UTC hours.
Funding. BTC funding stayed positive in every eight-hour window, averaging 0.0663% with a maximum of 0.1395% and a minimum of 0.0248%; ETH averaged 0.0808% with a maximum of 0.1636% and a minimum of 0.0331%. At these rates, a long in a perpetual contract pays a running cost every eight hours, and the week's positive rates indicate that cost was paid throughout.
What would change this reading
- If BTC closes a later week above its Sunday high of 12,123.46, the Sunday high-hour drop reads as a pause rather than a turn.
- If funding stays positive while prices fail to reclaim those Sunday highs, longs keep paying carry without further progress.
- If ETH continues to outperform, the ratio's 7.41% weekly gain extends and the choice of which asset to hedge becomes more important.
- If hourly moves near -6.92% repeat while volume stays above the prior twelve-week average, risk limits sized on quieter weeks remain too loose.
Terms
- funding rate
- the periodic payment between long and short holders of a perpetual futures contract; positive means longs pay shorts
- percentile
- the share of past weeks with a lower value; the 95th percentile means 95% of the 128 prior weeks were lower
- realised volatility
- how much the price actually swung over a period, expressed as a yearly percentage