How long does a pump.fun takeover run before it peaks?

Updated 11 October 2026 · by AtroX · Belphor Research · the data behind this guide

As of 3 October 2026, the median pump.fun takeover in Belphor's 60-day record that rose at least 10% peaked 9 hours 18 minutes after the alert; a quarter peaked within 54 minutes and a quarter after 19 hours 25 minutes. That is 9 takeovers, a small sample, and the spread matters more than the median: 33% peaked inside the first hour and 22% were still making highs after 24 hours.

When the peak came

Fetched from the public JSON on 3 October 2026. The live table refreshes daily; this article is a dated snapshot of it. How each column is measured: how Belphor measures outcomes.

The sample is the 9 takeovers whose highest price in the 36 hours after the alert was at least 10% above the alert price, out of 12 with at least 30 minute-level observations within the 36-hour window; a coin that never rose has no peak to time. A coin still rising at 36 hours counts in the last row, and peaks after 36 hours are not tracked.

The distribution is wide. 33% peaked inside the first hour and 44% within six hours, while 22% were still making new highs after a full day. The median peak, 9 hours 18 minutes after the alert, describes few of them, since only 1 of the nine peaked between six and twelve hours; half the peaks came within 54 minutes or after 19 hours 25 minutes, the quarter points of the spread.

The late peakers were the large ones. The 2 coins that peaked between 24 and 36 hours reached a median of 10.99x, against 1.54x for all nine. Two coins decide nothing on their own, and the record says so.

By market cap at the takeover

Each row holds two to four takeovers, so the table shows the shape of the record rather than a rule.

And how long until it dies?

Belphor does not publish a clock for that, because a death has no clean timestamp: a coin bleeds, bounces and bleeds again. What the record does publish is where coins stand at a later reading. On the entry-timing table, 25% of 12 takeovers bought at the alert were at 20% of the alert price or lower at the end of the 36-hour window. On the outcomes table, 32% of the 22 takeovers with a reading at least six hours later were down 80% or more at it. Different windows, different samples, the same message: a takeover that runs can also collapse, and in the entry-timing record a quarter did so within a day and a half.

pump.fun replaced the takeover instruction on 12 September 2026. Belphor recorded the new instruction, admin_cto, unpromoted while it was measured against the old record (n 38: 47 percent reached +50 percent against 48 to 54 under the earlier instruction; 21 percent were down 80 percent or more against 18 to 21). From 11 October 2026 its new-creator form is alerted and counted as a takeover, behind liquidity, mint-safety and operator gates. Holder-reward events are recorded as their own type and are not takeovers.

What the spread means for exits

A fixed-time exit can miss a later observed peak or follow an earlier one. Whether a trailing rule improves results requires a paired, after-cost replay including non-risers. A rule that follows the price, selling when the coin gives back a set share of its gain from the peak, is the idea discussed in trailing stops and profit taking and when to sell a memecoin. Belphor's Exit Guard, a product setting, watches each alert for 12 hours by default; that window is a choice, and this table does not test it. In this record 22% of the nine peaked after 24 hours, outside that window.