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

Monday’s analyst research calls — Block, Boise Cascade, Equifax, FirstEnergy, Lockheed Martin, Lumentum, NetApp, Roper, Spotify, and assorted technology names — landed in inboxes the way they always do: as a bundle of price-target adjustments dressed up as analysis. Treat them accordingly.

The statistically honest way to consume sell-side calls is to strip them to their signal components: rating direction, target delta, and timing. Price-target changes lag price moves more often than they lead them. Upgrades that follow a 20% run are momentum confirmation, not information. If you’re codifying this, a simple pipeline in Python — pulling ratings history via an API, computing post-upgrade drift with event-study windows (say, CAR over +1, +5, +21 days), and testing against a benchmark-neutral control — will tell you whether these calls carry any alpha in your specific universe. Spoiler for most large-cap coverage: the drift is thin and transaction costs eat it.

Sector context matters more than the individual ticker. Lockheed Martin sits in a name where analyst opinion has historically been noisy relative to defense-budget cycles — the actual driver is appropriations timing, not a target bump from $1 to $X. Equifax and Boise Cascade are cyclical plays wearing analyst-rating costumes; their factor exposures (credit conditions, housing starts, construction spend) explain more variance than any single desk’s conviction. A quick regression of each name against Fama-French factors plus relevant macro series via `statsmodels` or `alphalens`-style attribution will show you which analyst narrative is actually just beta with a press release.

Lumentum and NetApp are the interesting ones from a systematic standpoint — optical and data-infrastructure names where earnings-surprise momentum and revision breadth have historically been among the stronger post-earnings-announcement-drift candidates. If revision counts across the analyst cluster are inflecting in the same direction, that’s a measurable signal. If it’s one desk reiterating a target, it’s noise. Build the breadth metric: net revisions divided by total analysts, rolling 90 days. Threshold-cross, position-size, stop defined. Boring, repeatable, effective.

FirstEnergy rounds out the list as a regulated-utility name where analyst calls have near-zero predictive content — rate-case outcomes and the 10-year yield dominate. Anyone grading analyst accuracy on utilities is mostly measuring interest rates with extra steps.

The actionable takeaway: never trade a headline call, but do log every one. A structured database of rating changes, targets, and analyst identity becomes a feature store — and after enough observations, you can score each desk’s historical hit rate like any other model. Some analysts have genuine stock-picking skill; most have a beta. The data will tell you which is which, and the data is always more polite about it than I am.