Everyone Says "Buy Low, Sell High." Nobody Says How Low Is Low Enough.
Everyone says buy the dip. Nobody says where. A 5-layer framework to find the exact price to place your bid — and when to wait.
Markets are crashing. Your watchlist stocks are 20-30% off their highs. Twitter is full of "generational buying opportunity" takes. Your broker app shows a sea of red. You know you want to buy — you've done the fundamental research, you believe in the thesis, you have capital set aside.
But at what price do you actually place the order?
This is the question nobody answers well. Most advice stops at "buy the dip" or "dollar cost average." These are fine philosophies but terrible execution strategies. DCA doesn't care whether you're buying into an accelerating selloff or a forming bottom. "Buy the dip" doesn't tell you whether the dip is 5% or 25%.
I spent the last several months building a system that attempts to answer this question with more rigor. Not "should I buy this stock?" — that's a fundamentals question I assume you've already answered. The question is: "Given that I want to accumulate this stock, where exactly should I place my limit order to get the lowest possible price?"
Here's how it works.
The Setup: Greedy Accumulation
First, the framing. This isn't a trading strategy. It's an accumulation strategy for conviction holdings — stocks you've selected for fundamentals and moat, that you plan to hold for 12+ months. The current macro destabilization (pick your crisis — oil shocks, tariff wars, geopolitical escalation) is creating price dislocations in otherwise excellent businesses.
The goal is simple: buy these great stocks at the lowest possible price. Not "a good price" — the lowest possible price. Every rupee or dollar below your average cost compounds over years.
This reframes the entire problem. You're not asking "is it time to buy?" You're asking "where is the floor, and can I trust it?"
Five Layers of Intelligence
To find the floor, you need to look at the problem from multiple independent angles. Any single indicator can mislead you. But when five different analytical frameworks point to the same price band, something meaningful is happening.
Layer 1: Technical Analysis — Reading the Selling Pressure
The most important question during a crash isn't "how far has it fallen?" It's "are sellers running out of inventory?"
A stock can be down 30% and still have further to fall if fresh sellers keep showing up. Conversely, a stock down 15% might be at its floor if the selling has exhausted itself. The absolute drawdown tells you nothing about what happens next. The dynamics of the selling tell you everything.
There are specific patterns that distinguish "active selling" from "exhaustion." When volume is declining while price performance improves — even marginally — sellers are drying up. When volume is rising while declines accelerate, new sellers are joining the party and you haven't seen the bottom yet.
ATR — Average True Range — gives you a volatility-calibrated ruler to measure how far a "normal" dip extends versus a crisis-level drop. A conservative entry is less than one ATR below current price (a routine daily fluctuation). An aggressive entry is three ATRs below (crisis territory). These bands auto-recalibrate as volatility changes, which is critical during market stress when fixed-percentage targets become meaningless.
RSI divergence, ADX trends, MACD histogram — these aren't predictive on their own. But they tell you about the character of the decline. Is momentum accelerating or decelerating? Is the trend strengthening or losing conviction? Is the selling pattern consistent with capitulation or steady liquidation?
Layer 2: Options — What the Dealers Know
Options data reveals something price action alone cannot: where institutional money is being committed.
Max Pain — the strike price where option writers (typically dealers and institutions) lose the least money — acts as a gravitational center. When max pain sits above the current price and is stable or rising, dealers are positioned for the stock to be higher. When it's falling, dealers are repositioning lower and the floor is moving.
Open interest in puts versus calls tells you about institutional commitment. A Put-Call Ratio above 1.0 with rising put OI means institutions are writing puts — they're selling insurance against further declines, which means they believe the downside is limited. They're being paid to bet on a floor. This is different from retail buying calls hoping for a bounce.
IV skew measures how much more expensive downside protection is compared to upside speculation. When IV skew is extreme, the market is pricing in fear. When it normalizes, institutions are no longer buying crash protection — they think the worst is over.
The OI support wall — the strike with the heaviest put open interest — functions as a price floor defended by dealer hedging mechanics. When dealers have sold a large number of puts at a certain strike, they dynamically hedge by buying the stock as it approaches that strike, creating genuine buying pressure at that level.
Layer 3: Macro Regime — The Tide That Lifts or Sinks All Boats
Individual stock analysis means nothing if the macro environment is actively deteriorating. You can find the perfect floor on a stock and watch it break because oil spiked another 10% overnight.
The key insight about macro analysis for accumulation purposes isn't the absolute level of any indicator — it's the rate of change. FII selling of negative 5000 crore per day that has decelerated to negative 2000 per day is a stabilizing signal, even though flows are still negative. VIX that has peaked and is flattening is different from VIX making new highs. Brent crude that's range-bound at an elevated level is different from Brent that's still climbing.
A simple regime classifier that counts bearish signals across oil, currency, volatility, foreign flows, and news sentiment gives you a quick read: are things getting worse (stay away), stabilizing (start watching), or improving (act)?
Layer 4: News and Sentiment — The Narrative Layer
News doesn't move markets directly, but it tells you about the narrative cycle. During a crisis, the progression typically follows a pattern: shock and escalation, sustained fear, fatigue, and then de-escalation or normalization.
When headlines shift from "markets plunge as crisis deepens" to "markets fall on continued concerns" to "markets stabilize despite ongoing uncertainty," the narrative is maturing. Each phase corresponds to different accumulation opportunities.
More importantly, news analysis helps you distinguish between price declines driven by temporary external shocks (a geopolitical crisis that will eventually resolve) versus structural deterioration (a company's business model breaking). The former creates accumulation opportunities. The latter doesn't.
Layer 5: Fundamentals — The Thesis Check
This layer is the foundation that everything else sits on. Before you even begin the accumulation analysis, you need to answer: does the investment thesis still hold?
Is the company still growing earnings? Has revenue trajectory changed? Has the competitive moat been impaired? If a stock is down 25% but EPS has also collapsed 25%, it's not cheaper — it's repriced to reflect a worse business. That's not an accumulation opportunity; it's a value trap.
For the stocks in my portfolio, fundamentals serve as a continuous filter. As long as PAT growth is intact, the moat is undamaged, and the business is executing, price weakness driven by macro factors is a gift. The moment fundamentals deteriorate, the thesis needs re-examination regardless of what the other four layers say.
The Convergence Zone: Where Signals Agree
Here's where it gets interesting. Each of the five layers produces floor-price candidates independently. Technical analysis gives you ATR-based entry bands and 52-week lows. Options give you OI support walls and max pain levels. Fibonacci retracements give you mathematically-derived support levels. Pace projections give you where-the-price-is-heading estimates.
These are all independent methodologies. They use different data, different math, different assumptions. When they disagree — when TA says the floor is at 680 but options say 720 and Fibonacci says 650 — you have uncertainty. Any single signal could be wrong.
But when they converge — when four or five independent signals all point to the same price band — you have something much more robust. The probability that five unrelated analytical frameworks all happen to produce the same wrong answer is much lower than any single framework being wrong.
The implementation uses ATR as a clustering window. Any two floor candidates within one ATR of each other belong to the same cluster. The cluster with the most members wins. In practice, this produces remarkably clean results. Here's a recent real example:
HDFC Bank with 7 floor candidates: ATR aggressive at 681, pace projection at 694, OI support at 700, ATR optimal at 702, ATR conservative at 719, Fibonacci 100% at 727, and 52W low at 727. The algorithm clusters these into a zone at 681-719 with 5 converging signals. The 52W low and Fibonacci sit just above the zone but outside the cluster — they're nearby reference points, not part of the core convergence.
That's your convergence zone. Five independent signals from three different analytical frameworks (ATR volatility bands, options positioning, and pace projection) agree: the floor for HDFC Bank is somewhere in the 681-719 range.
But Can You Trust the Zone?
Finding the convergence zone is the easy part. The hard part is deciding whether to act on it.
A convergence zone computed during active, accelerating selling is very different from the same zone during selling exhaustion. The math is the same; the context is different. This is where judgment comes in — and where I use an LLM (GPT-5.4) for contextual assessment.
The LLM doesn't estimate prices. It doesn't compute anything. It receives the deterministic zone along with all the live context — selling dynamics, macro regime, options positioning, news — and answers one question: will this zone hold?
The answer is one of three labels. FIRM means the zone is well-supported — selling is exhausting, institutions are defending it, macro headwinds are stabilizing. Place your limit orders with confidence. FRAGILE means the zone exists but could break under pressure — mixed signals, maybe exhaustion is starting but macro is still deteriorating. Only place orders at the low end. UNRELIABLE means the zone is unlikely to hold — fresh sellers, no institutional defense, macro actively worsening. Wait.
This two-layer architecture — deterministic computation plus contextual judgment — avoids the biggest trap of using AI for investment decisions. You don't want the AI making up price targets. You want it assessing conditions. The zone is math. The reliability is judgment. And the judgment is constrained to three clearly-defined categories with explicit criteria.
From Zone to Limit Order
The final step translates the zone and its reliability into an actual order:
When the floor is FIRM, you get three options. Place at the top of the zone for the highest fill probability (aggressive), the midpoint for balance (optimal), or the bottom for the best price if it fills (conservative).
When the floor is FRAGILE, only the bottom of the zone is recommended, with an alternative below the zone in case it breaks.
When the floor is UNRELIABLE, the advice is simply: wait. Don't try to catch a falling knife when the conditions don't support a floor.
What I Learned Running This in a Live Crash
I've been running this system through the current market turbulence, monitoring four stocks across India and the US. A few observations from real-world use.
First, the system is genuinely useful for preventing premature action. When every instinct screams "this is cheap, buy now," seeing "UNRELIABLE — new sellers still entering, volume rising" on all four stocks is a concrete, data-backed reason to wait. It overrides the emotional pull of a 25% drawdown.
Second, the convergence zone is more stable than I expected. Even as prices move day to day, the zone shifts gradually because it's anchored by structural signals (OI walls, Fibonacci levels, 52W lows) that don't move on daily noise. This gives you a relatively stable target even in volatile markets.
Third, the selling exhaustion signal is the single most important input. More than options, more than macro, more than any other layer — whether sellers are running out of inventory is the dominant factor. Everything else is context around that core question.
Fourth, there's a meaningful difference between markets where options data is available and markets where it isn't. The Indian stocks with NSE options data produce richer, more confident zones because OI support and max pain are direct measures of institutional positioning. Stocks without options data rely more heavily on TA-derived signals, which makes the zone inherently less robust.
The Uncomfortable Truth
No system eliminates risk. The convergence zone could break. The LLM could assess FIRM and be wrong. Black swan events don't respect statistical clusters.
But the alternative — buying on gut feel, buying because a stock is "down a lot," buying because someone on social media called a bottom — is strictly worse. At minimum, this approach forces you to articulate why you think a certain price is the floor, using multiple independent frameworks. Even if the zone breaks, you've made a decision based on the best available evidence rather than emotion.
The strategy works best with what I call "greedy patience." You've identified the stocks. You've computed the zone. You've assessed its reliability. Now you wait — not passively, but actively monitoring for the exhaustion signal that tells you the zone is ready to hold. When that signal fires, you place your limit and let it fill.
In a market where everyone is telling you to buy the dip, the real edge isn't in buying. It's in knowing exactly where to place your bid, and having the discipline to wait until conditions support it.