Analyst price targets follow the stock. They don't lead it.
Analyst price targets follow the stock. They do not lead it. Across 426,782 consecutive price target revisions in AnaChart’s records, the number an analyst publishes moves with the price action that has already happened, and the bigger the move, the more reliably it does.
That is not a claim about any one analyst being lazy. It is what 833,000+ recorded analyst events look like in aggregate once you line up each revision against the stock’s move since that same analyst’s previous call on that same name.
82.6% of price target revisions move in the same direction the stock had already moved. The correlation between the size of the revision and the size of the prior price move is 0.705. Median time between one call and the next on the same stock: 90 days.
What’s in this study
The pattern, and how strong it is
Take every case where an analyst set a price target on a stock and then set another one on that same stock later. That gives 426,782 pairs, drawn from 5,165 analysts across 5,695 tickers. For each pair, compare two things: which way the analyst moved the target, and which way the stock had moved in the meantime.
Those two numbers correlate at 0.705. In the 357,537 pairs where both the target and the stock actually moved, 82.6% of revisions went the same direction the stock had already gone.
Here’s the catch: that is not what a price target is supposed to be. A target is presented as a forecast, a statement about where the stock is headed. In aggregate it behaves much more like a description of where the stock has been.
The bigger the move, the harder they chase
The relationship is not flat. Sort every revision by how far the stock had moved since the analyst’s previous call, and a clear U shape appears.
Read the middle bar first. When the stock had gone nowhere, meaning it moved less than 5%, the follow rate drops to 62.0%. With no price signal to anchor to, analysts scatter. That is the closest thing in this dataset to independent judgment, and it is the only bucket where a meaningful share of revisions cut against the drift.
Now read the ends. After a fall of more than 30%, 93.7% of revisions were cuts. After a rise of more than 30%, 94.5% were raises. The more decisively the market has already repriced a stock, the more uniformly the targets follow it down or up.
They chase, but they under-chase
Here is the part that costs readers money. Following the stock would at least keep the implied upside intact. It does not.
Across the 122,266 revisions that came after a gain of more than 10%, the median raise lifted the target by 16.5% while the stock had already gained 21.5%. The target went up. The gap between the target and the price got smaller.
A raised price target usually signals less upside than the one it replaced, not more. The headline reads bullish. The arithmetic underneath it is the opposite.
This is why “analyst raises target on NVDA” is close to information free as a standalone headline. It tells you the stock went up, which you could already see.
Who does not follow the tape
The averages hide real variation, and the outliers are where this gets useful. Among analysts with 150 or more recorded revisions, the ones least tied to the prior price move look like this.
| Analyst | Revisions | Follows the prior move |
|---|---|---|
| Mathieu Robilliard | 155 | 56.8% |
| John Gerdes | 156 | 64.1% |
| Steven Seedhouse | 176 | 65.9% |
| Mani Foroohar | 165 | 66.1% |
| Ed Arce | 235 | 66.4% |
| Robert Burns | 268 | 66.4% |
| Terence Flynn | 321 | 67.0% |
| Devin McDermott | 637 | 68.1% |
Analysts with at least 150 consecutive revisions in AnaChart’s records, ranked by the share of those revisions that moved in the same direction as the stock’s prior move. Lower means less tied to the tape.
Notice what they have in common. These are largely biotech and energy analysts. In those sectors the thing that changes a fair value is an event: a trial readout, an approval, a reserve revision, a commodity curve. The share price moving 20% does not by itself change the model, so the target does not automatically move with it.
The other end of the table is just as consistent.
| Analyst | Revisions | Follows the prior move |
|---|---|---|
| Steve Barger | 203 | 95.6% |
| Rupesh Parikh | 237 | 95.4% |
| Vincent Andrews | 165 | 94.5% |
| Jeffrey Meuler | 164 | 93.9% |
| Dennis Geiger | 256 | 93.8% |
| Bobby Griffin | 156 | 93.6% |
The same measure, highest first. Consumer, restaurant and industrial coverage, where a valuation multiple applied to a steady earnings base tracks the share price closely by construction.
Neither group is doing anything improper. A consumer analyst whose model is a multiple on next year’s earnings will mechanically produce a higher target when the multiple re-rates. The point is that the two behaviours look identical in a headline and are completely different as information.
Experience does not fix it
The obvious hypothesis is that seasoned analysts anchor less to the tape. They do not.
| Analyst tenure when the call was made | Revisions | Follows the prior move |
|---|---|---|
| Under 4 yearsNewer coverage | 73,414 | 81.8% |
| 4 to 12 yearsMid career | 264,556 | 82.8% |
| 12 years or moreVeterans | 63,555 | 83.0% |
Tenure measured as years between the analyst’s first recorded price target and the call in question. The spread across two decades of experience is 1.2 percentage points, and it runs the wrong way.
Bottom line: this is a structural feature of how sell side targets are produced, not a skill gap that seniority closes. If anything the veterans chase marginally more.
We found the same thing testing conviction. Across all 625,901 price targets with a usable implied upside, the median target sits 15.5% above the price for analysts under four years, 15.6% for mid career and 15.9% for veterans, with near identical Buy, Hold and Sell mixes. Newer analysts do not swing bigger. They just move the same way everyone else does.
Even the most covered stocks
If independent analysis survives anywhere, it should be on the names with the deepest coverage and the most competition for a differentiated view. It does, but only barely.
Tesla at 77.9% and Nvidia at 78.0% are the least tape-following megacaps, which fits: both are names where the bull and bear cases rest on arguments about a future business rather than on next quarter’s multiple, so genuine disagreement survives. Meta at 87.3% sits near the other end.
Even so, the spread across the entire market is narrow. No ticker in the dataset with meaningful coverage falls below 73.9%.
Why this happens
Three mechanisms explain most of it, and none of them require bad faith.
Models take the price as an input. A target derived from a target multiple on forward earnings moves when the multiple moves, and the multiple moves when the price moves. The chasing is built into the arithmetic.
Being wrong alone is worse than being wrong together. A target far from the market invites questions from clients and from an internal committee. Staying inside the pack is professionally cheaper than defending an outlier, especially after the market has moved decisively against it.
Revisions are scheduled, not triggered. The 90 day median gap between calls is close to a quarterly reporting rhythm. Many targets are refreshed because earnings arrived, not because the analyst’s view changed, and refreshing a model after a quarter of price action mechanically pulls the target toward the price.
What to do with a price target instead
The practical response is not to ignore analysts. It is to stop reading the target on its own and start reading the person attached to it.
Two analysts can publish the same $220 target on the same stock on the same morning. One has a record of targets on that name being reached quickly and by a wide margin. The other has been marking to market for three years. The number is identical and the information content is not remotely the same.
That per stock, per analyst record is what AnaChart is built to expose: for every analyst covering a stock, how many of their targets on that name were reached, how long it took on average, and where the performance score ranks them against everyone else on the same ticker. The score rewards targets that were reached quickly and by a wide margin, and a target that was never reached scores zero.
Related reading: our study on how quickly newer analysts reach their targets compared with veterans, and why the individual analyst’s name matters more than the firm’s.
How we measured this
Source. AnaChart’s own records: 833,000+ analyst events covering 660,000+ price targets across roughly 9,686 tickers and 7,191 named analysts, with history back to 2004 and dense continuous coverage from 2013. Snapshot used for this study runs through April 2026.
Construction. For every analyst and ticker combination we ordered that analyst’s price targets by date and formed consecutive pairs. Each pair yields two figures: the percentage change in the price target, and the percentage change in the stock’s closing price between the two call dates. A revision “follows” if both moved the same direction.
Filters. Pairs more than 400 days apart were dropped, since a gap that long says little about a response to a specific move. Price target and price changes beyond plus or minus 200% were dropped as data artifacts. The 82.6% headline uses the 357,537 pairs where both the target and the price moved by more than 0.1%; the 0.705 correlation is computed across all 426,782 pairs. Implied upside figures use the 625,901 targets where both a target and a same day close were recorded, trimmed to between minus 95% and plus 500%.
Tenure. Years between an analyst’s first recorded price target anywhere in the dataset and the call being measured. This understates true career length for anyone who was publishing before AnaChart’s coverage begins, which is a conservative direction for the tenure comparison.
What this study does not claim. It does not measure whether targets were reached, how fast, or how well any individual analyst performed. Those are separate metrics that require full daily price history, and they are what AnaChart’s met ratio, average days to target and performance score are built on. This study measures one thing only: the direction of a revision against the price move that preceded it.
Limitations. Closing prices are taken from the dataset’s own record of each event date, so a stock’s path between two calls is not observed. Corporate actions are handled by the underlying split adjustment and a small number of ragged rows were excluded. Sector labels in the discussion above are descriptive, drawn from the coverage of the named analysts, not from a formal sector classification.
Related reading: who moves price targets first on a stock, and who arrives behind them.
Frequently asked questions
Do analyst price targets predict where a stock is going?
Mostly they describe where it has already been. Across 357,537 consecutive price target revisions in AnaChart’s records, 82.6% moved in the same direction the stock had already travelled since that analyst’s previous call, and the correlation between a revision and the prior price move is 0.705. The pattern gets stronger as the move gets bigger: after a stock rose more than 30%, 94.5% of the next revisions were raises.
How long after a stock moves does the price target change?
The median gap between one analyst’s call on a stock and their next call on the same stock is 90 days. So a target you read today typically reflects a quarter of price action that has already happened. That is why the date on a target matters as much as the number, and why AnaChart timestamps every one of the 660,000+ price targets in its records.
If analysts chase the stock, does the upside number mean anything?
It means less than it looks. After a stock had risen more than 10%, the median revision raised the target by 16.5% while the stock had already gained 21.5%. The target moves up, but by less than the price did, so the advertised upside quietly shrinks even as the headline number climbs. Reading the raise as fresh conviction gets this backwards.
Which analysts do not just follow the tape?
They cluster in sectors where value is set by events rather than by price. Among analysts with 150 or more revisions, Mathieu Robilliard follows the prior move just 56.8% of the time, John Gerdes 64.1%, Steven Seedhouse 65.9%, Mani Foroohar 66.1% and Ed Arce 66.4%. Those are biotech and energy names, where a trial readout or a reserve revision resets the model regardless of what the share price did last quarter. At the other end, consumer and industrial analysts such as Steve Barger (95.6%) and Rupesh Parikh (95.4%) move with the tape almost every time.
Does experience make an analyst less likely to chase the price?
No, and this is the cleanest result in the study. Splitting every revision by how long the analyst had been publishing targets, the share that follow the prior move is 81.8% for analysts under four years, 82.8% for four to twelve years and 83.0% for veterans of twelve years or more. A longer career does not change the behaviour. This is one reason AnaChart ranks analysts by a performance score built on what their targets actually did, not on tenure or on brand name.
So how should I read a price target at all?
Read it against the record of the person who set it. A raise from an analyst whose targets on that stock have been reached quickly and by a wide margin carries different information from an identical raise by someone who has been marking to market for years. That per-stock, per-analyst record is exactly what AnaChart publishes: how many of their targets on that name were reached, how fast, and how the performance score ranks them against everyone else covering it.
Look up a stock on AnaChart to see every analyst covering it, how many of their price targets on that name were reached, how quickly, and how they rank by performance score. Start with NVDA, TSLA or browse the full analyst directory. Institutions licensing the underlying data can find it at anachart.store.
Data as of April 2026. Last updated 27 July 2026.