What Happens After Bitcoin Hits a New All-Time High?

"Bitcoin hits a new all-time high" is one of the biggest headlines in crypto. It's also when a lot of people buy for the first time.

So is a record a sign that more gains are coming, or a sign of the top? We looked at every record close on Bitstamp from April 2013 to 1 October 2026, and at what bitcoin did next.

Finding 1: new highs come in bunches

A record close almost never comes alone. 85% of bitcoin's 171 records were followed by another within a week. 91% were followed by another within a month.

They fall into fourteen bunches. Each bunch ends with a fall of 20% or more. The biggest bunch had 26 records, and the smallest had 3.

So if bitcoin closes at a record today, the odds say another one is coming soon. That's the good news. The bad news is in the gaps between the bunches.

Finding 2: the first new high usually slips back

The first record after a fall of 20% or more is a breakout. There have been thirteen since 2013.

In nine of them, bitcoin closed back below the breakout price within days. Seven slipped the very next day, and none took more than two. The dips were small. In the typical case, the lowest close in the next month was 8% below the breakout price. The deepest was 22% below, after the February 2017 breakout.

Here's the part that matters. Every one of those breakouts still went on to set at least two more records before the next big fall. A quick slip back below the old high didn't mean the breakout had failed.

Only four breakouts held above their price for a whole month: November 2013, April 2017, October 2017 and November 2024.

Finding 3: how far the run went

From a breakout to the top of its run, the median gain was 37%. The median run lasted 29 days.

The runs have got smaller. Up to 2020, the gains ran from 8% to 366%. The biggest came after the November 2013 breakout. Since 2021, every run stopped between 5% and 40%.

That fits what our halving-cycle study found: bitcoin's swings have shrunk as it has grown. A breakout today is likely to run less far than one in 2017.

Finding 4: a good month, a worse year

In the weeks after a record, bitcoin did better than usual. 63.7% of record closes were higher 30 days later, against 55.6% of all days. The median gain over those 30 days was 11.5%, against 2.6%.

A year later, it was the other way round. Only 57.4% of records were higher a year later, against 68.6% of all days. The median one-year gain after a record was 10%. After any day, it was 68%.

Why the flip? Because records bunch up near the tops. The last records of a bunch come just before a big fall, so a year later many of them are underwater.

There's a catch with the good month too. Records happen in bull markets, and every day in a bull market tends to do well. So we also compared each record with other days in the same year. The month after a record beat other days that year in only three of six years. The year after was worse in five of six. The exception was 2021. A year after almost any day that year, bitcoin was lower.

So a record tells you you're in a bull market. It doesn't tell you much more than that about the next month.

Finding 5: every crash started at a new high, but few highs started a crash

Each of bitcoin's fourteen falls of 20% or more since 2013 started at a record close. Six of them went past 50%. Our bear-market study measures how deep and how long they were.

But turn it around. Out of 171 records, only fourteen were the last one before a 20% fall. That's about 1 in 12. Only six came before a 50% fall, about 1 in 29.

So buying at a record isn't usually buying the top. But when it is, the top can be very expensive. Nobody can tell which record will be the last one until long after.

Finding 6: where bitcoin sits now

Bitcoin's last record close was $124,728, on 6 October 2025. That's 360 days ago. On 1 October 2026 it closed at $84,853, 32% below that high. To set a new record, it needs to rise 47%.

If it gets there, the past breakouts give you a rough picture, not a forecast. Most slipped back below the old high within a day or two. The median run went 37% further, and the runs since 2021 were all under 50%. Thirteen breakouts is too few to plan on.

How to read it

• Expect records to come in bunches. One new high usually means more are close.
• Don't panic if a breakout slips back below the old high. Most did within days, and every one went on to more records.
• Don't read a record as a promise for the year ahead. A year after a record, bitcoin did worse than after an average day.
• Remember every big crash started at a record. Size your position so a 50% fall from the top won't force you to sell.
• If you invest a fixed amount every month, records are just another month. You don't have to decide whether this one is the top.

Method, and what this cannot tell you

We took every daily close of bitcoin on Bitstamp, in US dollars, from September 2011 to 1 October 2026. A record close is one above every close before it. We count from April 2013, because Bitstamp's record starts after bitcoin's first big peak in June 2011. A breakout is the first record close after a fall of 20% or more, and its run ends at the last record before the next 20% fall. Binance's daily closes give the same answer on whether each breakout since 2020 slipped back or held.

• Closes only, in UTC. A record set during the day and lost by the close doesn't count here.
• The 171 record closes aren't 171 separate events. They come in 14 bunches, so the real sample is small.
• Thirteen breakouts is a small sample. The next one doesn't have to look like any of them.
• Comparing record days with all days mixes in the bull and bear years. We also compared them with other days in the same year, and the short-term edge mostly went away.
• Bitstamp was a small exchange in 2013, and prices from those years differ from one exchange to the next.
• The 20% cut-off for a fall is our choice. A different cut-off would give a different number of breakouts.