The Bitcoin Pi Cycle Top currently shows no crossover (October 18, 2026 close): the 111-day average ($73,031) is 0.48× twice the 350-day average ($153,580).
Two moving averages of the daily close: the 111-day, and the 350-day doubled. The short one almost always trails far behind. When it catches up, price has spent months running well ahead of its yearly trend, and in 2013 and 2017 that happened within days of a cycle high. From its publication in 2019 to 17 Oct 2026, it fired once.
“Bitcoin Pi Cycle Top: the 111-day average ($73,031) is 0.48× twice the 350-day average ($153,580), 18 Oct 2026 close.” NodeWitness, https://nodewitness.com/en/charts/bitcoin-pi-cycle-top-chart/
Think of a rev counter. The 111-day average is the engine’s recent speed; twice the 350-day average is the redline. The needle only gets there after months at high revs, never on a single week’s burst, which is why the indicator goes quiet for years at a time. The ratio panel shows how far the needle is from the redline: 1 is a crossover, 0.5 means the short average is halfway there.
Start with the lower panel. The ratio is the 111-day average divided by twice the 350-day, and the red line marks 1: above it, the short average has crossed. Dashed verticals mark each upward crossover on both panels. The top panel shows the same averages over price, on a log scale, so you can place each crossover within its cycle.
A crossover describes a rally that has already happened. For the 111-day average to reach twice the 350-day, price has to have spent months far above trend. It says nothing about valuation, on-chain activity or who holds the coins, and, as the name gives away, it was built for tops, not bottoms [1].
Philip Swift published it on Look Into Bitcoin in April 2019 [1], with the 2013 and 2017 tops already on the chart. His write-up says it caught cycle highs within three days and explains the name: 350 divided by 111 is 3.153, close to pi. In our series, the crossovers of 6 Apr 2013 and 16 Dec 2017 land three days and zero days from the highest close of their run; the 6 Dec 2013 one comes two days after the 4 December high.
Windows picked by looking at those tops will fit those tops. That’s overfitting, and the real test is what happened after publication. One crossover since, up to 17 Oct 2026: 12 Apr 2021, close $59,835, a day before the spring high ($63,554 on 13 Apr), back below on 23 Apr. The cycle’s highest close came on 8 Nov 2021 at $67,562, with the ratio at 0.57. The next cycle hadn’t crossed as of 17 Oct 2026: our series’ highest close, $124,777 on 6 Oct 2025, also came with the ratio at 0.57, and from January 2024 to that date the ratio didn’t go above 0.74. Swift’s own page now warns that ETFs and Bitcoin’s integration into the financial system may make the indicator irrelevant [1].
Each row is a day the 111-day average closed above twice the 350-day after being below it. The table is computed from the chart's series every time the page loads.
We start from our USD daily close series, which begins on 17 Aug 2010 [2]. For each day we take the simple average of the last 111 closes and of the last 350, and double the second, as in the original definition [1]; the ratio is the first divided by that doubled figure. Averages are strict: a missing day resets the window. That’s why the 111-day average starts in December 2010 and the doubled 350-day on 1 Aug 2011.
An upward crossover is the first close on which the 111-day average sits above twice the 350-day after being at or below it; a downward crossover is the reverse. Whatever state the series starts in doesn’t count as a crossover. Today only counts once it has closed, so the latest point is yesterday’s.
We compared the four upward crossovers in our series (6 Apr 2013, 6 Dec 2013, 16 Dec 2017 and 12 Apr 2021) with a public reference built on the same price source [3]: the day, the price and both averages match to the dollar.
Four crossovers in fifteen years is a tiny sample, and three of them predate publication and were used to design it. It only reacts to fast, sustained rallies: a high reached more slowly may produce no crossover, as in November 2021 and at our series’ highest close in October 2025. Between crossovers it says nothing, and the ratio has no intermediate levels with a meaning of their own. This page describes the series; it is not investment advice.
We go through all four crossovers, and what followed each, in our Pi Cycle Top recomputation. At heart it’s a moving-average crossover with well-chosen windows, which is exactly where overfitting creeps in. For where it stands today against its own history, next to ten other cycle indicators, see Bitcoin cycle top and bottom indicators today.
Not in our series, at least up to the 17 Oct 2026 close: the last upward crossover was on 12 Apr 2021. The following cycle’s highest close, $124,777 on 6 Oct 2025, came with the ratio at 0.57, nowhere near 1. The latest close, and how far the short average is from crossing, are shown above the chart.
That rally was slower. After April’s crossover, price fell until 20 Jul 2021 and the 111-day average sank with it. The second leg made a higher close, $67,562 on 8 Nov 2021, but not fast enough or for long enough to pull the short average back up to twice the long one. From July to December 2021 the ratio never got above 0.79.
Its creator presents it that way. We don’t: four crossovers in fifteen years, three of them used to choose the parameters, is too thin a record for a rule, and the indicator describes how fast price has risen, not what comes next. We show it as history, not investment advice.
Because 350 divided by 111 is 3.153, close to pi (3.1416). It’s a quirk of the name; pi plays no part in the calculation.
The rest of the cycle metrics are on Cycle & Score.