Day three versus day sixty

A result prints, and the stock moves. Watch it for three days and you see one thing; watch it for sixty and you often see another. The move in the first three days is fast, loud and emotional — it is the headline hitting the wire, the algorithms trading the beat-or-miss keyword, the fast money adjusting positions before anyone has read the notes. The move over the next thirty to sixty days is slow and quiet, and it is made of something else entirely: analysts rebuilding their models, the concall being digested, the quality of the result being understood, and the price settling toward what the business is actually now worth.

These two moves are different animals, and they need not point the same way. A headline that looked like a miss but was actually fine gets sold on day one and repriced up by day sixty; a genuine surprise that the market under-reacted to barely moves on day three and then drifts in the same direction for weeks. This module is about that gap — why the reflex and the verdict diverge, and why the reader who did the pre-result work (108) can act into the crowd's day-three mistake, or wait for the day-sixty confirmation, instead of being ruled by the tape.

The price of a stock is not repriced in an instant, because the information in a result is not absorbed in an instant. On results day the market has a headline, a numbers table, and a wall of algorithmic and fast-flow trading that reacts to keywords — "record profit," "misses estimate" — in milliseconds. What it does not yet have is the considered reading: the concall listened to in full, the segment detail reconciled, the guidance weighed against delivery, the analysts' models torn up and rebuilt. That considered reading takes days and weeks to arrive, and it is what drives the durable reprice long after the knee-jerk has spent itself.

This slow absorption has a name and a direction. When a result carries a genuine surprise, the market tends to under-react to it at first and then keep moving in the surprise's direction for weeks — post-earnings-announcement drift, one of the most durable and best-documented patterns in markets. The mechanism is exactly the slow price discovery above: a real surprise is incorporated gradually, not at once, so the drift is the market finishing a job the knee-jerk only started. It is under-reaction stretched out over time — and it is why a considered reader who correctly judged the result can be paid over the following weeks even after the first move has happened.

The knee-jerk, then, can do one of two things over the following weeks. It can reverse — a headline miss that a full reading shows was actually fine gets sold in the reflex and bought back as the understanding arrives. Or it can continue — a real surprise the market under-reacted to keeps drifting the same way as the models catch up. Which one happens is not knowable from the day-three move itself; it is knowable only from the work of judging the result's true quality, which is the entire apparatus Part Nine has been building.

Lay the reaction out on a timeline and its two halves separate cleanly:

• Days 1-3: the knee-jerk. Driven by the headline keyword, index and passive flow, options hedging, and fast money adjusting. It is emotional and mechanical at once, and it reads the top line, not the disclosure. Its size tells you how surprised the fast market was; its direction is often unreliable, because it is reacting to the framing rather than the substance.
• Days 4-30: the digestion. The concall transcript circulates, analysts publish revised notes, the segment and cash detail get reconciled, and the embedded expectation the price carried gets reset to a new level. This is where a knee-jerk starts to reverse or extend.
• Days 30-60: the considered reprice. The models are rebuilt, next year's estimates re-based, and the stock settles toward what the business is now understood to be worth. For a genuine surprise, this is the bulk of the drift; for a headline that fooled the reflex, this is where the reversal completes.

The practical consequence is a choice, not a formula. Having done the pre-result work, you can act on day three into the crowd's mistake — buy the good result the reflex sold, or sell the flattered one the reflex bought — accepting that you are early and the reprice may take weeks. Or you can wait for day sixty, letting the concall and the model revisions confirm your read before committing, trading some of the edge for more certainty. The three-day knee-jerk is the noise; the sixty-day considered reprice is the signal — and the whole skill is refusing to let the loud, fast number stand in for the quiet, slow verdict.

The size of the day-3-to-day-60 gap is not the same for every stock, and that is the module's inversion: the same genuine surprise drifts by wildly different amounts depending on how fast the stock is priced. A heavily-covered index heavyweight is repriced almost fully within days — dozens of analysts, deep liquidity, near-instant price discovery — so the day-3 move is close to the day-60 move and patience earns little. An under-covered mid-cap absorbs the same surprise over weeks, because few analysts rebuild models and information spreads slowly, so the drift is long and the edge to the reader who did the work is largest. And a narrative-driven, retail-heavy name flips the logic entirely: instead of under-reacting, it over-reacts, spiking past what the result justifies and then mean-reverting — the day-3 move overshoots and the day-60 move gives part of it back.

The instruction that carries across every cell is to read how fast this stock gets priced alongside the result itself. The identical surprise is a small drift in the heavyweight, a long drift in the mid-cap, an overshoot to fade in the narrative name, and a cycle-dependent reversal in the cyclical. Point the drift logic at the wrong kind of stock — ride the overshoot as if it were under-reaction, or wait patiently for a drift in a name already fully priced — and the edge turns into a trap.

A composite under-covered mid-cap specialty chemicals maker, Meridian Chemicals illustrative, reports — followed by just two analysts, thinly traded. [illustrative] You did the pre-result work: your range had revenue at ₹500-545 crore, and the print lands at ₹540 crore with the operating margin a full point above your high end, on a genuine volume surprise in a new export product. But the headline the wire runs is "revenue mix shifts to lower-margin exports," and the reflex sells it: the stock is down 5% by day three on the scary framing and some fast-money exit.

Now the two clocks. On day three, the knee-jerk has read the headline, not the disclosure — it sold a mix worry while missing that the operating core beat comfortably on a durable surprise. Over the next weeks, the considered reprice arrives: the concall confirms the export product is structural and repeats next year, the two analysts publish upgraded models, and — crucially — because the stock is thinly covered, that repricing happens slowly, over weeks rather than days.

The reader who had a number of their own saw, on day three, that the fall belonged to the headline and the flow, not to the business — the operating result beat, and the surprise was real. That reader could buy into the crowd's day-three mistake, accepting they were early and the reprice would take weeks, or wait for the day-sixty drift to confirm before committing. Either way, the discipline is the same: the three-day move was a fact about framing and flow, and the business's worth was decided by the slow reprice that followed, not the reflex that opened it.

Knowing that the knee-jerk and the verdict diverge does not tell you which way a given day-three move will resolve. A drift and a reversal are indistinguishable on day three; both begin as a move you have not yet had time to interpret. Only your independent read of the result's quality — did the operating core genuinely beat, was the surprise real — tells you whether to expect the reflex to reverse or extend. Without that work, "the knee-jerk is often wrong" is not a strategy; it is just a shrug at the tape.

It cannot give you the timing. Drift is a tendency measured over weeks, not a scheduled event. The considered reprice can take thirty days or ninety, can stall, or can be interrupted by a fresh macro shock or a later result before it completes. Being right about the destination and wrong about the pace is a real and common way to lose money on a correct read, which is why acting on day three trades certainty for edge.

And drift is not guaranteed on any single result. Post-earnings-announcement drift is a statistical regularity across many events, not a law that binds the next one. This particular surprise may be priced faster than usual, or the stock may already have drifted before the print on leaked expectations. The pattern earns its value across many results and a disciplined process, not as a promise on the one in front of you — which is why you grade the read by the work behind it, not by whether this single drift paid.

The first trap is treating the day-three move as the verdict. The stock fell, so the result "must have been bad," and the position is sold or the thesis abandoned — when the fall was the reflex trading a headline and the considered reprice is still to come. The fast move feels like judgement because it is loud and immediate; it is usually just the framing and the flow.

The second is riding an overshoot as if it were drift. A narrative-driven, retail-heavy name spikes on a result, and the drift logic says "under-reaction, it will keep going" — but this kind of name over-reacts and mean-reverts, so the day-sixty move gives back the day-three move. Applying the mid-cap's drift edge to an overshoot name is how a good pattern becomes a loss.

The third is waiting patiently for a drift in a fully-priced stock. In a heavily-covered heavyweight, the surprise is incorporated within days, so there is little left to drift; the reader who expects a mid-cap's long, slow reprice in an index name waits for an edge that was already gone by the time they could act. The size of the gap depends on the speed of price discovery, and in the most-watched stocks that speed leaves almost nothing on the table.

The fourth is confusing being early with being wrong — and its mirror, confusing a slow reprice with a failed read. Act into the day-three mistake and the position can sit red for weeks before the drift confirms; abandon it because it has not paid yet, and you convert a correct read into a realised loss precisely when the verdict was about to arrive. The two clocks demand patience: the work is judged on day sixty, not day three, and a read that is early is not the same as a read that is wrong.