Monday’s Analyst Calls: Block, Equifax, Lockheed—Signal or Noise? My Model Weighs In
# Monday’s Analyst Calls Are Noise Until You Quantify Them
The morning ticker of Wall Street research calls — Block, Boise Cascade, Equifax, FirstEnergy, Lockheed Martin, Lumentum, NetApp, Roper, Spotify — reads like a portfolio of opinions dressed up as analysis. Sell-side price target revisions are a factor, technically speaking. It’s just a factor with a documented decay curve and a questionable alpha stream once you control for momentum.
The measurable reality: analyst upgrades and downgrades clustered on the same morning tend to be reactions to Friday’s earnings prints or weekend macro developments, which means the informational content is largely stale by the time the market opens. Any residual drift — the post-earnings announcement drift analog for ratings changes — has historically been small, and it shrinks further when you filter for liquidity and execution costs. If your backtest can’t clear slippage on an upgrade-driven gap, the signal was never an edge.
Block and Spotify drawing fresh calls is unsurprising; both names carry high short interest and high beta, making them catnip for analysts chasing visibility. High-dispersion names also produce high-dispersion price targets, which is a polite way of saying the targets are close to noise. Lumentum is the more interesting case on this list — optical transceiver demand tied to AI datacenter capex is at least a measurable driver. If you want exposure, don’t trade the rating; build a signal off hyperscaler capex guidance and component order data. That’s an actual causal chain, unlike a rating change that mostly encodes the last 20% of price action.
For the industrials — Lockheed, FirstEnergy — analyst calls function mostly as sentiment markers around defense budgets and rate sensitivity. Both are tradeable via factor exposure rather than headline-chasing: value and low-volatility tilts have historically explained more of their return variance than any individual research note.
The systematic takeaway is blunt. If you’re building a ratings-based signal, backtest it properly: event-study framework, abnormal return estimation against a factor model, decile sorts on rating delta, and an honest look at whether the effect survives transaction costs. In my experience running these tests, sell-side ratings changes add little beyond what momentum and revisions-in-aggregate already capture. A single Monday’s call sheet is entertainment. A statistically validated revisions factor across the entire coverage universe is a strategy. One of these belongs in your pipeline; the other belongs in your morning skim, right after the coffee and before the serious work begins.