Analyst Calls Monday: Block, Equifax, Lockheed, Spotify—Signals or Just Sell-Side Noise?

Monday’s analyst call roundup — Block, Boise Cascade, Equifax, FirstEnergy, Lockheed Martin, Lumentum, NetApp, Roper, Spotify — is less a research product than a rebalanced sentiment feed. Treating it that way is the only honest approach.

Sell-side ratings changes are, at best, a weak factor. The academic record on analyst recommendation revisions shows post-revision drift exists but decays fast, and by the time a headline hits your RSS reader, the latency-adjusted edge is mostly gone. If you’re serious about harvesting it, you’d build a scraper (feedparser plus a lightweight parser over broker RSS), normalize ratings to a numeric scale, then regress forward returns against revision sign and magnitude with proper event-study alignment. Most people skip the regression and just buy what Upgraded. That’s not systematic investing; that’s headline-following with extra steps.

Some structural observations worth noting without inventing numbers. Defense-adjacent names like Lockheed and infrastructure names like Boise Cascade and FirstEnergy tend to attract upgrades clustered around macro narratives — defense budgets, rate cuts, housing. Narrative clusters are a contrarian input: when sell-side enthusiasm concentrates in one theme, cross-sectional dispersion of forward returns among those names typically widens, not narrows. Sector-level herding is measurable; use it as a feature, not a thesis.

Names like Lumentum and NetApp are interesting for a different reason: optical and data-infrastructure exposure makes them high-beta proxies for AI capex sentiment. Their price action correlates more with hyperscaler earnings commentary than with their own fundamentals. If your model includes them, lag their signals against the big-cap AI complex — the lead-lag is usually one-directional.

The correct trade on any analyst-call digest: log it, timestamp it, backtest the signal class over your own window with transaction costs included, and only then decide it belongs in the feature set. In my experience, sell-side revisions survive a costs-inclusive backtest less often than momentum does — and momentum barely survives.

The upgrade is free. The alpha isn’t.