← Back to blog
Methodology

Layer Rotation, Confluence Engine, Momentum Map: how our three signal tools actually connect

Oct 8, 2026 · 5 min read · AI007 Team
🔗

We get some version of this question a lot: what's the actual difference between Layer Rotation, the Confluence Engine, and the Momentum Map -- and which one should you be looking at? The honest answer is they're not competing tools. They answer three different questions, in order, and they're built to be used together.

Layer Rotation Heatmap: where is money moving, right now?

This one works within a single universe. It ranks that universe's own layers against each other by relative strength each trading day -- rank 1 is the strongest layer that day, rank N is the weakest -- over a rolling window, so you can see whether a layer's strength is a one-day blip or a real multi-day trend. Click into any layer and it drills down into the actual companies driving that day's move, not just an abstract number.

What it answers: which part of the AI trade is leading today, and has it been leading for a while or did it just show up?

Confluence Engine: is that rotation real, or just noise?

A layer moving isn't proof of anything by itself -- single-day moves happen for all kinds of reasons that don't last. The Confluence Engine cross-checks a source-universe layer's move against a related ETF-universe layer. A thesis only gets logged when a layer moves at least 2% in one direction, the paired ETF layer moves the same direction, and that ETF move comes on real volume -- at least 1.3x its 20-day average. Only then does it get marked CONFIRMED rather than DIVERGENT or NOISE, and once confirmed, it's tracked forward for 10 sessions to see whether the thesis actually plays out.

What it answers: is this rotation backed by real conviction, or could it reverse by tomorrow?

Momentum Map: which specific stocks do I actually look at?

Layer Rotation and the Confluence Engine both operate at the group level -- a layer, a sector, a theme. Neither tells you which individual stock inside that group is actually worth a look. The Momentum Map plots every real stock across all 7 universes on a six-factor momentum score (RSI, MACD, moving-average position, 1-month and 12-month rate of change, and volume versus its 20-day average) against its real 1-month return. Four corners: LEADING (strong score, strong return), IMPROVING (score turning up, return hasn't caught up yet), WEAKENING (return still positive, score rolling over), LAGGING (neither).

What it answers: once I know which layer is hot, which stocks inside it are actually leading, not just riding the average up?

All three, on the same real thesis

Here's today's data walking through all three, in order, rather than three separate screenshots. On Oct 6, Power & Energy (T9) was one of the AI Supply Chain universe's strongest layers on the Layer Rotation Heatmap. The Confluence Engine picked that up as a trigger -- T9 up 2.54%, its paired Energy & Grid Infrastructure ETF layer (E10) confirming with a 3.36% move on 1.55x normal volume -- and logged it CONFIRMED. That thesis is still active. Drop into the Momentum Map and filter to T9: Solaris Energy Infrastructure (SEI) sits furthest into the LEADING corner, momentum score 73, real 1-month return +19.5% -- with Vistra (VST) and Quanta Services (PWR) right behind it, both BULLISH with double-digit 1-month returns.

That's the actual workflow: Layer Rotation tells you where to look, the Confluence Engine tells you whether to trust it, and the Momentum Map tells you which names inside it are worth the rest of your attention. Each one on its own is a data point. Used in that order, they're closer to a process.

Want to see it yourself?

Launch App Free →

More from the blog

🔥One glance, the whole AI economy: how the AI Universe Heatmap actually works🏛️The 10-year and 30-year Treasury yields just hit multi-year highs. Tech didn't blink -- here's why⚡The Nasdaq 100 and Nvidia just hit fresh highs — here's where the rotation data says to look next