Bitcoin is sitting almost exactly on its 100-day simple moving average this morning: $76,389.68 on Coinbase at 07:04 UTC against a line that CryptoTicker puts at about $76,540 and that our own Yahoo Finance calculation puts at about 76,646 (CryptoTicker, 3 November). Chip stocks have lost about $3.5 trillion since 6 October and Americans are voting, so a close below the line tonight is a live possibility. The question this study answers is narrow: when bitcoin first closed below its 100-day average after a sustained run above it, what did the next 7 and 30 days look like, compared with any random day?
The short answer is that the next week was better than average and the next month was not. Over 22 earlier crossings since 2015, the 7-day return after the crossing averaged +4.6% against +1.4% for all days, and the share of positive weeks was 15 of 22 (68%) against 54.7%. The 30-day average, +11.9%, looks even stronger, but it is carried by three crossings in 2017; the median 30-day return was +3.7%, only 12 of 22 were positive (55%) against 57.7% for all days, and without 2017 the average falls to +2.0% with a median of −0.4%.
How we defined a crossing
We used Yahoo Finance BTC-USD daily closes (UTC) from 17 September 2014, 4,431 observations, and began measuring after the first 120 days so every date has a 100-day average behind it. A crossing is the first daily close below the 100-day simple average of closes after at least 20 straight closes above it. That filter removes the repeated whipsaw closes that happen when price lingers around the line. It finds 22 crossings with a complete 30-day follow-up; the closest two are 44 days apart, so their 30-day windows do not overlap. Today’s close is not counted because the day is not finished. Bitcoin had closed above the line 76 days in a row before today.
The results
| Horizon | Crossings: mean | Crossings: median | Crossings: up | All days: mean | All days: median | All days: up | Chance of mean this high by luck* |
|---|---|---|---|---|---|---|---|
| Next 7 days | +4.58% | +3.32% | 15 of 22 | +1.39% | +0.70% | 54.7% | about 5.5% |
| Next 30 days | +11.89% | +3.72% | 12 of 22 | +6.40% | +3.07% | 57.7% | about 12.9% |
*Permutation test: we drew 20,000 random samples of 22 start days from all days and counted how often the sample mean was at least as high as the crossings’ mean. It is a rough guide, not a verdict: it ignores that overlapping daily windows are correlated, and we also looked at several slices of the data, so a single 5.5% should not be read as significance.
At 7 days the signal is on the edge of what chance would produce; at 30 days it is not distinguishable from chance, and the median test gives about 44%, which is unremarkable. The honest summary is a mild tendency for a short bounce after the line breaks, with no reliable edge beyond it.
Why 2017 matters so much
Three crossings in 2017 (18 March, 15 July and 14 September) were followed 30 days later by +22.6%, +116.4% and +84.8%. Bitcoin was in the steepest part of the 2017 advance and the dip below the 100-day line was a buying opportunity each time. Remove them and the 19 remaining crossings give a 30-day mean of +2.0% and a median of −0.4%, with 9 up and 10 down. At 7 days the same exclusion leaves a mean of +2.4%, a median of +3.3% and 13 of 19 positive, so the short-term bounce survives better than the 30-day claim.
Other splits that matter today
- Since 2021 (12 crossings). The 30-day returns were −21.6, +52.9, −5.2, −26.8, +40.4, +16.6, −0.4, +15.9, −7.9, −15.3, +2.9 and −15.9: a mean of +3.0% but a median of −2.8%. The 7-day mean was +3.9% and the median +0.5%. In the modern, ETF-era market the median outcome after a crossing was flat to negative.
- Above or below the 200-day average. In 16 of the 22 crossings, price was still above its 200-day average; those had a 30-day mean of +15.9% (heavily 2017) and median of +5.3%, with 10 of 16 up. In the six crossings that happened below the 200-day, the 30-day mean was +1.2% and the median −5.9%. Today bitcoin is above the 200-day, which Yahoo-based closes put near 73,150 and CryptoTicker puts near $73,160.
- Downside risk. In 9 of the 22 cases (41%) the lowest close within the next 30 days was at least 10% under the crossing close, against 34.5% of all days. Slightly worse than a random day, not dramatically.
Video: a Schwab Network debate on Nvidia and the 200-day average as a line in the sand for tech stocks, published 30 October 2026. It is about equities; the moving-average logic is the same.
The 22 crossings
| Crossing close (UTC) | BTC close ($) | 100-day avg ($) | Next 7 days | Next 30 days |
|---|---|---|---|---|
| 18 Aug 2015 | 211 | 256 | +5.0% | +8.9% |
| 15 Jan 2016 | 364 | 367 | +5.0% | +11.8% |
| 2 Aug 2016 | 547 | 580 | +7.4% | +4.5% |
| 18 Mar 2017 | 974 | 992 | −0.1% | +22.6% |
| 15 Jul 2017 | 1,999 | 2,039 | +40.6% | +116.4% |
| 14 Sep 2017 | 3,155 | 3,218 | +15.1% | +84.8% |
| 22 Jan 2018 | 10,931 | 11,144 | +3.3% | −2.2% |
| 28 Aug 2019 | 9,754 | 10,057 | +8.6% | −15.4% |
| 8 Mar 2020 | 8,108 | 8,429 | −33.5% | −11.5% |
| 5 Sep 2020 | 10,170 | 10,261 | +2.7% | +6.1% |
| 25 Apr 2021 | 49,004 | 49,815 | +15.6% | −21.6% |
| 21 Sep 2021 | 40,694 | 40,776 | +0.8% | +52.9% |
| 26 Nov 2021 | 53,570 | 53,937 | +0.1% | −5.2% |
| 11 Apr 2022 | 39,522 | 41,535 | +3.3% | −26.8% |
| 10 Mar 2023 | 20,187 | 20,269 | +35.8% | +40.4% |
| 24 May 2023 | 26,335 | 26,637 | +3.4% | +16.6% |
| 17 Aug 2023 | 26,665 | 28,554 | −1.9% | −0.4% |
| 1 May 2024 | 58,254 | 59,545 | +5.0% | +15.9% |
| 14 Jun 2024 | 66,011 | 66,754 | −2.9% | −7.9% |
| 11 Feb 2025 | 95,747 | 96,081 | −0.2% | −15.3% |
| 25 Aug 2025 | 110,124 | 111,547 | −0.8% | +2.9% |
| 1 Jun 2026 | 71,320 | 73,193 | −11.5% | −15.9% |
Source: Bitcoin Mastery calculation from Yahoo Finance BTC-USD daily closes. The 100-day average is the simple mean of the previous 100 closes including the crossing day.
What this means on election day
If bitcoin closes below the 100-day average tonight, history says a modest bounce within a week has been the more common result and a stronger rebound within a month has not. That is a description of 22 events, not a forecast: the previous crossings happened in very different markets, ranging from a sub-$1,000 asset to a trillion-dollar one. The more useful reading for today is about risk: the 20-day and 50-day averages were already lost at the end of October, so the 100-day line is the last of the shorter averages still in play, and below it the next reference is the 200-day near 73,150. The US election, AMD’s results and Binance’s next steps are all inside the same 24 hours, which makes the single daily close a noisier signal than usual. Our earlier work on the 100/200-day golden cross and a guide to reading bitcoin’s moving averages covers what the lines are; our study of semiconductor-ETF down days covers the equity link.
Video: a creator’s take on bitcoin after past midterms, published 2 November 2026; it is opinion rather than data.
Limits of the study
- Twenty-two events over twelve years is a small sample, and the distribution is dominated by a handful of extreme months.
- We tried a 20-day filter, a 30-day horizon and a 7-day horizon, and several subsets; each additional look raises the chance that one of them looks special by accident.
- Closes are Yahoo Finance UTC daily closes; other data providers set the day boundary differently and produce slightly different averages, as the $76,540 and 76,646 figures above show.
- The study says nothing about volume, leverage or news, which on 30 October and 2 November were large drivers of the move.
Sources: CryptoTicker, CoinDesk; calculations by Bitcoin Mastery from Yahoo Finance data. Related: today’s market report.
Method: bitcoin prices, highs, lows and closes are Coinbase Exchange candles (UTC day) and the Coinbase spot ticker pulled by Bitcoin Mastery at about 07:04 UTC on Tuesday 3 November 2026; the 3 November candle is still open at that time, so the latest price is the 76,389.68 ticker and the last completed daily close is Monday 2 November (76,674.13). Long-history bitcoin figures are Yahoo Finance BTC-USD daily closes (UTC). Third-party figures are attributed to their source and date. Moving averages and event returns are our own calculations from Yahoo Finance daily closes pulled on 3 November 2026; the 3 November close in that data set is an open candle and is excluded from all statistics.
Disclaimer: This article is for informational purposes only and does not constitute investment advice. Bitcoin and other cryptocurrencies are volatile and you can lose some or all of the money you put in. Nothing here is a recommendation to buy, sell or hold bitcoin, any exchange-traded fund, any listed security or any other asset, and the technical levels, probabilities and scenarios discussed are descriptions of published data, not forecasts. Do your own research and consult a licensed financial advisor before making investment decisions.