$BTC
3 widely circulated numbers, all copied without timestamps
First: -12.34%.
This morning, a screenshot was being shared in quite a few groups. On a forex market data page, Bitcoin’s change was listed as -9,173.24 points, or -12.34%. It looked like Bitcoin had crashed 12% in a day.
I checked the page: its “Previous close” was listed as 74,314.32. But 74,314 wasn’t yesterday’s price—it was the price on November 9. In other words, the page had crammed the entire week’s decline from November 9 to November 16 into the “Change” field, while labeling it “Previous close.”
At the same time, I checked a 12:36 snapshot of Binance’s USDT-margined perpetual contract: the 24-hour change was -0.018%. A 12-percentage-point difference. Both figures were real; one described a week, the other a day.
Second: -1.05%, at 64,710.
This came from an 08:22 article by Zhitong Finance. It was written quite properly and even included the 200-week moving average from Daan Crypto Trades: 67,962, and the 200-week exponential moving average: 69,909. The numbers were fine.
The problem was the timing. That was the price at 08:22. When I checked again at 12:36, it was 64,970.01. In four hours, the price had risen 260 points, and the decline had narrowed from -1.05% to -0.018%. If you didn’t see that article until 3 p.m., you might think Bitcoin was still down 1%.
Third: Fear and Greed Index at 10.
SoSoValue reported 10 today. alternative.me reported 13 yesterday. These two figures are often strung together as “down from 13 to 10, fear deepens.”
But they aren’t points on the same scale. They come from two organizations, two sets of questionnaires, and two different weighting methods. Subtracting figures from different sources is like subtracting the markings on two different rulers. The difference may look meaningful, but it’s really a difference between the rulers.
At this point, the three mistakes all come down to the same thing: missing timestamps, missing exchange and contract details, and missing rolling-window lengths.
From now on, whenever you see a number, ask three questions: What time was it taken? Which exchange and which contract? What time window does this change cover? If you can’t answer all three, that number is just for atmosphere, not a basis for decisions.
Numbers themselves don’t lie. The people copying them just leave out the time.
Though a thousand siftings and ten thousand washings make the task hard, only when the wild sand is blown away does the gold appear.
The 12% drop you saw today—what page did you see it on, and at what time? Share the timestamp and the asset from your screenshot, and I’ll check that we’re comparing like with like. To get that table every day, complete with timestamps and contract names, join the “Eagle Soars High · Crypto Study Group” chat.
3 widely circulated numbers, all copied without timestamps
First: -12.34%.
This morning, a screenshot was being shared in quite a few groups. On a forex market data page, Bitcoin’s change was listed as -9,173.24 points, or -12.34%. It looked like Bitcoin had crashed 12% in a day.
I checked the page: its “Previous close” was listed as 74,314.32. But 74,314 wasn’t yesterday’s price—it was the price on November 9. In other words, the page had crammed the entire week’s decline from November 9 to November 16 into the “Change” field, while labeling it “Previous close.”
At the same time, I checked a 12:36 snapshot of Binance’s USDT-margined perpetual contract: the 24-hour change was -0.018%. A 12-percentage-point difference. Both figures were real; one described a week, the other a day.
Second: -1.05%, at 64,710.
This came from an 08:22 article by Zhitong Finance. It was written quite properly and even included the 200-week moving average from Daan Crypto Trades: 67,962, and the 200-week exponential moving average: 69,909. The numbers were fine.
The problem was the timing. That was the price at 08:22. When I checked again at 12:36, it was 64,970.01. In four hours, the price had risen 260 points, and the decline had narrowed from -1.05% to -0.018%. If you didn’t see that article until 3 p.m., you might think Bitcoin was still down 1%.
Third: Fear and Greed Index at 10.
SoSoValue reported 10 today. alternative.me reported 13 yesterday. These two figures are often strung together as “down from 13 to 10, fear deepens.”
But they aren’t points on the same scale. They come from two organizations, two sets of questionnaires, and two different weighting methods. Subtracting figures from different sources is like subtracting the markings on two different rulers. The difference may look meaningful, but it’s really a difference between the rulers.
At this point, the three mistakes all come down to the same thing: missing timestamps, missing exchange and contract details, and missing rolling-window lengths.
From now on, whenever you see a number, ask three questions: What time was it taken? Which exchange and which contract? What time window does this change cover? If you can’t answer all three, that number is just for atmosphere, not a basis for decisions.
Numbers themselves don’t lie. The people copying them just leave out the time.
Though a thousand siftings and ten thousand washings make the task hard, only when the wild sand is blown away does the gold appear.
The 12% drop you saw today—what page did you see it on, and at what time? Share the timestamp and the asset from your screenshot, and I’ll check that we’re comparing like with like. To get that table every day, complete with timestamps and contract names, join the “Eagle Soars High · Crypto Study Group” chat.