The “moderate buy” rating nobody assigned
Consensus stock ratings feel authoritative but are just an average of guesses, refreshed by software, not judgment.
By The Signal · · 5 min read
"Moderate Buy" is not a rating any analyst gives. No one at Bank of America or Piper Sandler types those words into a report. It is an average, computed by a data vendor, of individual ratings that came out of dozens of separate spreadsheets — and it now gets published as a standalone news story, several times a week, for thousands of tickers that generated no other news at all.
What actually happened
Align Technology (Nasdaq: ALGN) and Amkor Technology (Nasdaq: AMKR) both received wire stories this week announcing a "consensus recommendation" of Moderate Buy, sourced to MarketBeat, a financial data and content company based in Kentucky. Neither story reports a new rating action. No analyst changed a price target, upgraded a stock, or issued a fresh note that week. What changed, if anything, was the input mix behind an average that MarketBeat recalculates continuously and republishes as headline news whenever its automated system decides the item is worth a wire push.
The mechanism is simple enough to describe in one sentence: take every buy/hold/sell rating issued on a stock in the trailing twelve months, weight them, average the result, and label the output with a tier — Strong Buy, Moderate Buy, Hold, Moderate Sell, Strong Sell. MarketBeat's methodology page assigns numeric values (Buy = 4, Hold = 3, Sell = 2 in its scale) and takes the mean. A score of 3.4, say, rounds into "Moderate Buy." The label is arithmetic, not opinion.
Who pays, who gains
Nobody pays for this content in the way readers usually mean. That is the point. MarketBeat runs on a freemium model: free consensus-rating articles function as top-of-funnel content that drives traffic toward a $29.95-a-month "MarketBeat All Access" subscription, and the free stories get syndicated onto Google News, Yahoo Finance and dozens of regional outlets that license the wire feed cheaply because it is inexhaustible — there are more than 6,000 tickers on major U.S. exchanges, and each one can generate a fresh "consensus" story every time a single new analyst note nudges the average.
Publishers gain volume at near-zero marginal cost: a templated story about ALGN looks structurally identical to one about AMKR, because it is the same script with different tickers piped in. Readers searching a company name get an article that reads like news but discloses no event. The analysts whose names appear in these pieces — the ones MarketBeat lists as contributing to the average — did their own work weeks or months earlier and have no relationship to this particular republication.
The mechanism
Think of it as a railway timetable board that updates itself from a feed of other people's timetables, then gets reprinted as if a train had actually arrived. The underlying analyst ratings are real — a research desk at, say, Piper Sandler genuinely did put "Overweight" on Align Technology in a note tied to specific guidance, margins or earnings data. That work has a date, a named author, and a rationale a reader could follow back to a specific data point, like clear aligner shipment volumes or gross margin guidance for a coming quarter.
What MarketBeat and similar aggregators (TipRanks, Zacks) do is pool those individual notes — sometimes a dozen, sometimes three — into a single blended score, refreshed as a batch process, and then treat any movement in that blended number as a discrete news event worth a headline. The "consensus" is a statistical artifact of when ratings happen to be issued, not a judgment anyone actually rendered. If two analysts initiate coverage in the same week, the average shifts, and the software generates a story, even though nothing about the company changed and no analyst reconsidered anything.
This differs from a genuine rating action — an actual upgrade or downgrade — the same way a poll-of-polls differs from an election result. FiveThirtyEight-style aggregation is a legitimate statistical technique when the underlying inputs are disclosed and the averaging method is transparent. The problem here is presentation: these stories run in the "News" tab, formatted with a company name, ticker and verb ("Receives"), which implies an action took place. The verb is doing work the content doesn't support.
What happens next
The volume of this content is rising, not falling, because the cost of producing it is close to zero and Google News treats it as legitimate financial journalism eligible for the main feed rather than a lower-tier aggregator category. Align Technology's own week illustrates the asymmetry: alongside the consensus-rating story, the company also won a genuine, reportable event — a $1.4 million judgment in a legal dispute, per DrBicuspid.com's dental-industry trade coverage — and that story, with an actual number tied to an actual court decision, got a fraction of the syndication reach of the templated ratings piece, because ratings stories are keyed to ticker symbols that trigger automatic alerts on brokerage apps and stock-tracking widgets.
Regulatory scrutiny of this format has been limited. The SEC's rules on research-analyst conflicts (Regulation AC, adopted 2003) govern disclosure by the analysts issuing original ratings, not the aggregators repackaging them, so the wire-service layer sits in a compliance gap. The likelier correction is behavioral rather than regulatory: as more readers notice that a "Moderate Buy" headline discloses no new information, click-through rates on these stories should decay, the way they eventually did for algorithmically generated local-weather and sports-recap content in the mid-2010s. That decay has not shown up in the syndication volume yet.
FAQ
Is a "consensus rating" useless information? No — it can be a legitimate summary statistic if you know the sample size and the recency of the underlying notes. The problem is treating a recalculated average as a news event rather than a snapshot you'd check manually, the way you'd check a batting average rather than expect a headline every time it moves by a point.
Who actually writes these stories? MarketBeat discloses a templated, largely automated production process on its site; the byline attribution on syndicated versions varies, and Google News frequently strips it entirely, leaving only the outlet name.
Does this affect the stock price? Rarely, on its own. Genuine rating changes from a single influential analyst can move a stock a few percentage points intraday; a recalculated consensus label, absent a new underlying rating, is not documented to have comparable effect, because algorithmic trading systems are generally tuned to react to individual analyst actions, not aggregator relabeling.