Pi Cycle Top

How to read it

The Pi Cycle Top indicator watches two moving averages of very different lengths. When the fast one crosses above twice the slow one, price has risen steeply enough, for long enough, that previous cycles have topped within days. It has called several cycle tops with unusual precision.

Price is drawn on the same axis, thin and in orange. It is what makes the indicator legible: a crossover on its own is just two lines meeting, but a crossover sitting on a price peak is the claim the chart is making. It also shows you the years of nothing in between, which the crossovers alone hide.

How it's calculated

The 111-day moving average plotted against twice the 350-day moving average, both derived from price and so sharing its dollar axis. The signal is the crossover of the former above the latter.

What it did in past cycles

The indicator has fired four times in the whole record. Computed from the series on this page:

• 2013-04-05 — the April 2013 local top, correct
• 2013-12-05 — one day after the $1,135 cycle top, correct and almost exact
• 2017-12-16 — the $19,641 cycle top, to the day
• 2021-04-12 — twelve days before the April 2021 local top, correct for that high

That is the whole list, and the omissions matter more than the entries. The November 2021 cycle top produced no crossover — on 2021-11-08 the 111-day average sat at $49,510 against $87,260 for twice the 350-day, nowhere near a cross. The October 2025 top produced no crossover either, and was further away still: $113,700 against $199,800.

So the honest scorecard is two exact calls on the 2013 and 2017 cycle tops, one correct call on an interim high, and two consecutive cycle tops missed entirely. The pattern is consistent with the rest of this site's evidence: later cycles have risen less steeply, and an indicator tuned to the steepness of 2013 no longer reaches its trigger.

What it doesn't tell you

Three or four observations is not a sample. The specific numbers — 111, 350, and the multiplier of two — were chosen because they fit past tops, which is the definition of curve fitting. It also produces no useful signal at all in the years between tops.

And it has now missed two cycle tops in a row. Any account of this indicator that lists its 2013 and 2017 successes without mentioning November 2021 and October 2025 is selecting its evidence.

Reading it with other metrics

Because it fires so rarely, Pi Cycle is worth nothing on its own as a monitoring tool. MVRV and AVIV Z-Score produce a reading every day and have degraded less badly.

The power-law rainbow fails in the same direction and for the same underlying reason — both are calibrated on early cycles whose steepness has not recurred. Seeing them fail together is a better lesson than either individually. The cycle page shows the halving-relative framing the whole idea rests on.