How the weekly spot model decides, and the markers already printed on the frozen sample.
The input is BINANCE:BTCUSDT, standard weekly candles. A week opens Monday 00:00 UTC. The model decides only after that week has closed, and the marker stays on the decision candle.
A teal band is an accumulation zone. An orange band is a distribution zone. BUY is a research marker to accumulate spot inside a confirmed lower zone. REDUCE SPOT reduces spot the model already holds in its example ledger.
A distribution phase lasts at most 16 closed weeks. A higher high, an unknown count, or a broken pair suspends it. It can resume only on a new count in the same family, and the original clock does not reset. At most four reductions are counted for the whole epoch.
Invalidation pauses new buys. The port keeps the frozen thresholds: WaveTrend, the partial count grammar, a 0.1 percent fee and 0.02 USDT of slippage inside the hidden example ledger, and four-week spacing between decisions.
| Known at | Marker | Slice |
|---|---|---|
| Aug 8, 2022, 00:00 UTC | Buy, accumulate | 1/4 |
| Dec 19, 2022, 00:00 UTC | Buy, accumulate | 2/4 |
| Sep 1, 2025, 00:00 UTC | Reduce spot | 1/4 |
| Sep 29, 2025, 00:00 UTC | Reduce spot | 2/4 |
| Nov 10, 2025, 00:00 UTC | Reduce spot | 3/4 |
The 2017 to 2026 sample was already known when R4 was frozen. It is in-sample. It has not been validated out of sample, and it is not a complete wave count.
On that sample the model printed two buy markers and three reduce-spot markers. It did not flatten the example position and it did not open another funded epoch. A later price can trade far below a buy marker. That is an adverse move, not a promise of recovery.
Nothing on this page is investment advice, a broker instruction, or a personal allocation. The example ledger stays hidden unless you open it.