Bitcoin Sunday pump, Monday dump: does the pattern hold?
After an up Sunday, Coinbase Bitcoin fell on Monday 54.03% of the time vs 48.55% for any Monday. See other coins, each year and the weekend move.
Only weakly, and not in every year. On Coinbase's BTC-USD spot price, a Monday closed down after an up Sunday 54.03% of the time (298 up Sundays among 587 Mondays since July 2015). Any Monday closed down 48.55% of the time. On Binance's BTCUSDT perp the numbers were 52.06% and 48.25%, and in several years the gap flipped.
These are past cases, not a forecast. They show how often Monday followed Sunday, not what the next Monday will do.
What does "Sunday pump, Monday dump" mean here?
The saying is that crypto rises on Sunday and falls on Monday. We can only test direction, so we define it plainly.
A pump is any Sunday that closed higher than the day before. A dump is any Monday that closed lower than Sunday. Size does not count here: a 0.1% Sunday rise counts the same as a 5% rise. (We look at big weekend rises further down.)
Days are UTC days, the shared world clock. UTC midnight is 7 or 8 p.m. in New York, depending on daylight saving time. So some of what you call Sunday evening is "Monday" in these tables.
How often did a down Monday follow an up Sunday?
For each market we counted Mondays that came after an up Sunday, and how many of those Mondays closed down. We compare that with how often any Monday closed down.
| Market | Mondays measured | Up Sundays | Monday down after an up Sunday | Monday down, any Monday |
|---|---|---|---|---|
| Coinbase BTC-USD (from 2015-07-20) | 587 | 298 | 54.03% | 48.55% |
| Coinbase ETH-USD (from 2016-05-18) | 543 | 278 | 52.52% | 48.25% |
| Binance BTCUSDT (from 2019-09-08) | 371 | 194 | 52.06% | 48.25% |
| Binance ETHUSDT (from 2019-11-27) | 360 | 191 | 50.79% | 46.67% |
| Binance SOLUSDT (from 2020-09-14) | 318 | 160 | 53.12% | 52.20% |
All rows end October 19, 2026. In all 12 markets we measured, the down-Monday share was higher after an up Sunday, by roughly 1 to 8 points. So the claim is not made up, but the gaps are small.
Is it really about Sunday, or about any up day?
A day that follows a rise may fall a bit more often than an average day, whatever the weekday.
On Coinbase BTC-USD, across all days, the day after an up day closed down 50.26% of the time. A random day closed down 47.69% of the time. So part of the Sunday-to-Monday gap also shows up when you look at all days, not only Sundays.
So the Sunday-only story has less room than the first table suggests, though we cannot prove the whole gap is general.
Did the pattern hold each year?
No. Here is Binance BTCUSDT year by year. Each year has only 19 to 32 up Sundays, so one year can swing a lot.
| Year | Up Sundays | Monday down after an up Sunday | Monday down, any Monday |
|---|---|---|---|
| 2020 | 27 | 44.44% | 42.31% |
| 2021 | 28 | 57.14% | 53.85% |
| 2022 | 22 | 40.91% | 50.00% |
| 2023 | 28 | 57.14% | 53.85% |
| 2024 | 32 | 50.00% | 45.28% |
| 2025 | 29 | 37.93% | 42.31% |
| 2026 (to October 19) | 19 | 63.16% | 42.86% |
In 2022 and 2025, a down Monday after an up Sunday was less likely than an ordinary Monday. In 2026 it was much more likely. We left out 2019, where Bitcoin had only 9 up Sundays.
The longer Coinbase BTC-USD history swings even more. In 2017, only 26.92% of up Sundays were followed by a down Monday (26 up Sundays). In 2018, it was 82.14% (28 up Sundays). Other coins swing too: Binance ETHUSDT ranged from 44.44% to 57.14% across 2020 to 2026.
A pattern that points one way one year and the other way the next is hard to use. It may be real in some periods and absent in others, or it may be noise.
Is Monday a weak day for crypto in general?
Not in these results. Here is Monday on its own.
| Market | Mondays | Average Monday move | Middle (median) Monday move | Share of Mondays that closed up |
|---|---|---|---|---|
| Coinbase BTC-USD | 587 | +0.42% | +0.09% | 51.45% |
| Binance BTCUSDT | 372 | +0.38% | +0.14% | 51.61% |
| Binance ETHUSDT | 360 | +0.43% | +0.33% | 53.33% |
| Binance SOLUSDT | 318 | +0.23% | -0.19% | 47.17% |
Monday was not a down day for Bitcoin or Ethereum. We also compared each weekday with the other six in all 12 markets, 84 comparisons in total. The usual bar for "stands out" is a t-score above 2. A t-score says how far a result sits from what random noise would give. Only 1 of the 84 passed it (Sunday for Binance XRPUSDT, at -2.09). About 4 would be expected by chance alone. So no weekday stood out.
For how big moves are on weekends rather than which way they go, see is crypto more volatile on weekends. For every weekday side by side, see the best day of the week to buy Bitcoin.
What about a big move over the whole weekend?
Some traders watch how far Bitcoin moves while CME Bitcoin futures are shut, from Friday 16:00 to Sunday 17:00 Chicago time. We did not use CME prices. We measured how far the spot price (the coin itself, not futures) moved during those hours (Coinbase spot, with a Binance perp as a check), then looked at the Monday after.
On Coinbase BTC-USD, across 461 weekends from December 22, 2017 to October 16, 2026, the typical size of that move (the median, ignoring direction) was 1.16%. The price was higher at the reopen on 59.44% of weekends.
| Market | Weekends | Monday down, any weekend | Weekends price rose | Monday down after a rise |
|---|---|---|---|---|
| Coinbase BTC-USD | 461 | 50.76% | 274 | 52.19% |
| Binance BTCUSDT | 371 | 48.25% | 222 | 50.90% |
The differences of 1 to 3 points are within normal noise for samples this size, so we call it no real difference.
Did a bigger weekend pump bring a bigger Monday dump? No. On Coinbase, when spot rose more than 3% while CME was shut, Monday closed down 48.28% of the time (58 weekends). After a rise under 1%, it was 55.20% (125 weekends). Both groups are small, so neither number means much on its own. But a big weekend rise was not followed by more down Mondays.
We counted every weekend on the schedule, including those in 2026, without checking whether the break still applied that year. For whether price comes back to its Friday level, see does Bitcoin fill CME gaps.
What does this not show?
- Direction only. In the Sunday tables, a 0.1% rise counts the same as a 5% rise.
- Daily closes. A Monday that rose first and then fell by the close is a down Monday.
- No cause. We did not study who traded or why.
- Overlap between coins. Coins move together, so the 12 markets are not 12 separate tests.
- No trading result. We tested no rule and counted no costs.
How we measured this
We used public daily price bars for Binance USDT perps and Coinbase spot, plus hourly bars for the weekend part, all ending October 2026. Each day's move is the change from one close to the next, labeled by its UTC date. We compared the share of down Mondays after up Sundays with the share for all Mondays, by market and by year. Missing days are not filled in.
Try this in Comparand
Ask your connected AI agent: "Get daily bars for Binance BTCUSDT from 2020. For each up Sunday, tell me whether the next Monday closed down, and show the count by year."
- venue_bars returns open, high, low and close prices plus volume. It supports 1-hour and 1-day bars and returns completed bars only. It is free on every key.
- Your agent counts the Sundays and Mondays from those bars. The tool does not return a ready-made weekday report.
For a different question, ask: "Find the 10 closest past charts to the latest 48 hourly bars of Binance BTCUSDT, and show what price did 12, 24 and 48 hours later."
- shape_analogs compares a coin only with its own past on the same exchange and bar size. It returns each past case with its own path and later moves, in basis points (100 basis points is 1%). It does not return an average or a hit rate, so you read the cases yourself.
- A search costs $0.01 from your credits.
You can try 3 coins free, and the docs show how to connect an AI app. For how well look-alike charts did in a test, see do chart look-alikes predict crypto. These are past cases, not a forecast.
The short version
- After an up Sunday, a down Monday followed 54.03% of the time on Coinbase Bitcoin, against 48.55% for any Monday.
- The same direction showed in all 12 markets, but by small amounts.
- Part of the gap shows up on every weekday: any day after a rise fell slightly more often.
- Year by year the gap flipped, and big weekend rises were not followed by more down Mondays.
- Monday's average move was up, and no weekday stood out across 84 tests.
Questions people ask
Does Bitcoin dump on Mondays after a Sunday pump?
Only slightly, and not reliably. On Coinbase BTC-USD, Monday closed down after an up Sunday 54.03% of the time (298 up Sundays, 587 Mondays since 2015), against 48.55% for any Monday. On Binance BTCUSDT it was 52.06% against 48.25%, and it changed sign in some years.
Is Monday a bad day for Bitcoin on average?
No. On Coinbase BTC-USD, Monday's average move was up 0.42% over 587 Mondays, and the middle (median) Monday was up 0.09%. Monday closed higher on 51.45% of those days. The result is not strong enough to call Monday special.
Is the Sunday to Monday pattern the same for Ethereum and Solana?
The direction was the same, but the size was not. After an up Sunday, a down Monday followed 50.79% of the time for Binance ETHUSDT and 53.12% for Binance SOLUSDT. Their ordinary Monday rates were 46.67% and 52.20%.
Does a big weekend price move lead to a Monday drop?
Not in our test. After a weekend when Coinbase Bitcoin rose, Monday fell 52.19% of the time (274 weekends), against 50.76% for all Mondays (461). A difference that small is within normal noise.