The sixth green week ran into a wall at 7,300 and came back. Friday settled at 7,248, right in the middle of our rotation range. Breadth is not confirming. Four timeframes of participation data say the same thing: about half the market is going up. The rest is flat or fading.
That is not bearish. But it is fragile.
The Lie holds at +2 CONSTRUCTIVE. Selling exhaustion confirmed. Institutional positioning improving. But breadth scored zero. Until the average stock starts pulling its weight, rallies into prior highs will get faded.
This week: AMD and Arista after the bell Tuesday. If both deliver, Wednesday gaps higher and the rally broadens. If they disappoint, the rotation low gets tested. NFP Friday is the end-of-week catalyst.
The playbook is simple. Buy the rotation low. Fade the round number. Let earnings decide.
The Lie: +2 CONSTRUCTIVE. Was +2 last week. Components unchanged. Breadth is the swing variable. A close above 65% on the 20-day upgrades the score mid-week. A close below 45% downgrades it.
Breadth divergence is the feature of this market, not a bug. $800B+ in AI capex this year (BofA estimates $1T by 2027) is flowing to 20-30 stocks. The other 470 are dealing with $102 oil, fertilizer costs up 40%, and a Fed that just split 4 ways on policy direction. Two economies, one index. Goldman’s research shows narrow breadth episodes last a median of three months since 1980 — but always resolve with a leadership rotation. The divergence resolves when software joins the party or when oil breaks something. Until then, the rotation range holds.
The scorecard, six levels for the week, and two high-conviction ideas are below the fold.
