Week of 10 May 2021: Bitcoin falls 20.28% while Ether slips only 8.71%, a split
When Bitcoin drops more than a fifth in a week while Ether holds up much better, what does the split, and the still-positive funding, reveal about where the pressure actually sat?
BTC fell from 58,240.83 to 46,431.50 (-20.28%), its 2nd-percentile week in 169 weeks of history. ETH fell only 8.71% (3,923.95 to 3,582.10), so the ETH/BTC ratio rose 14.51%. Funding stayed positive for both assets all week, and both assets set their lows at 20:00 UTC on Sunday.
What the data shows. BTC opened near its weekly high of 59,500.00 (Monday 04:00 UTC) and closed at 46,431.50, only 16.6% of the way up from its low of 43,825.39 — the week ended near its bottom, and that low arrived late, at 20:00 UTC on Sunday 16 May. The damage concentrated on Wednesday 12 May: BTC's worst day at -12.42% and its largest hourly move, -6.25% in the hour starting 23:00 UTC, in which ETH fell 9.69%. ETH set its own high later, at 4,372.72 on Wednesday 12 May at 14:00 UTC, then swung violently into the weekend: +9.60% on Friday, -10.51% on Saturday. ETH's realised volatility (the size of the swings that actually happened, scaled to a year) reached 158.4% against BTC's 108.2%. Dollar volume was 1.30 times the prior twelve-week average for BTC and 2.82 times for ETH — ETH's dollar volume (38.36 billion) exceeded BTC's (35.21 billion) in a BTC-led selloff.
What it does not show. The candles record the split, not its cause. An hourly correlation of 0.82 shows the two prices moved together closely on an hourly basis, though the data does not identify who traded or why. BTC sat 28.10% below its 52-week high and ETH 17.33% below; the data cannot say whether Sunday's lows were final.
What a careful desk would check. The prior four weeks split by asset: BTC had already fallen 6.41% and 12.61% before two up weeks (15.31%, 2.94%), while ETH had risen 27.11% and 33.05%. BTC's -20.28% return ranked at the 2nd percentile of the 169 weeks of history, while ETH's -8.71% return ranked at the 15th percentile of those same weeks, so the same calendar week carried very different historical severity for each asset. ETH's best and worst days came on consecutive days at the end of the week, so a desk would size positions for those swings rather than for the week's headline -8.71%.
Funding. Both assets paid positive funding on average (BTC 0.0332%, ETH 0.0349% per eight hours) and neither print fell below 0.01%. Longs were paying shorts throughout a week in which prices fell hard; the data shows the cost, not the size of the positions behind it.
What would change this reading
- If a later week trades below BTC's 43,825.39 low, Sunday's print reads as a pause inside a longer decline rather than a floor.
- If the ETH/BTC ratio gives back the 14.51% gain while BTC recovers, the week's split was relative positioning unwinding, not a change in ETH's trend.
- If funding stays above 0.01% while prices keep falling, longs are continuing to pay to hold positions that are losing value.
- If ETH's realised volatility stays near 158.4%, its daily swings of 9.60% and -10.51% set the sizing standard, not BTC's.
Terms
- realised volatility
- how much the price actually swung over a period, expressed as a yearly percentage
- close location
- where the week's close sits between the week's low and high; a low percentage means it finished near the bottom
- percentile
- the share of past weeks with a less extreme value; BTC's 2nd percentile means only 2% of the prior 169 weeks were worse