Averaging Down in Stocks: When the Math Works, When It Destroys Capital
Buying more of a falling stock lowers your break-even — but in Yes Bank, every average-down buyer lost more than they saved on paper.
Most discussions of averaging down stop at the break-even math: if you bought 10 shares at ₹500 and the stock falls to ₹300, buying 10 more brings your average to ₹400. True. But the calculation ignores two things: your total capital deployed in one stock just doubled, and the reason the price fell matters far more than the new average. This article works through both — with two Indian examples on opposite sides of the outcome.
| Scenario | Market panic (temporary shock) | Business deteriorating |
|---|---|---|
| Why the price fell | Macro event, sentiment, broad correction | Revenue falling, losses rising, debt piling up |
| Business fundamentals | Intact | Weakening or broken |
| Thesis for buying more | Clear: market corrects, company unchanged | Absent or hope-based |
| Historical examples | Nifty crash March 2020, HDFC Bank | Yes Bank 2018–2020, Vodafone Idea |
| Typical outcome | Averaging down rewarded | Averaging down multiplied losses |
| The honest test | ”Would I buy this fresh today?” — Yes | ”Would I buy this fresh today?” — No |
The math of averaging down
If you buy N shares at price P₁ and then M more at price P₂ (where P₂ < P₁):
New average cost = (N × P₁ + M × P₂) ÷ (N + M)
Three things are simultaneously true after you average down:
First, your break-even price is lower. Instead of needing the stock to return to ₹500, you only need it to reach ₹400.
Second, your recovery hurdle from today’s price is still real. If the stock is at ₹300, it needs to rise 33% to reach your new ₹400 average — versus 67% to reach your old ₹500. The improvement is genuine, but not as dramatic as it sounds.
Third, your total exposure has increased. If both lots are the same size, you now have twice as much capital in one stock.
The calculator below shows all three effects simultaneously. Enable “Add dates” to see XIRR: it shows whether the time-weighted return, accounting for how long each rupee was deployed, actually justifies holding.
The exposure trap: the math everyone skips
Say your portfolio is ₹10 lakh. You bought ₹1 lakh of stock A — a 10% position.
The stock falls 40%. You average down with another ₹1 lakh.
Now you have ₹2 lakh in stock A: roughly 22% of your (now slightly smaller) portfolio. The position that started as a 10% allocation is now your single largest holding.
If the stock falls another 40%, your ₹2 lakh becomes ₹1.2 lakh: a ₹80,000 loss on the averaging-down purchase alone.
The break-even improvement is real. But averaging down always trades recovery probability for concentration risk. The more you average, the more the fate of your portfolio depends on one stock being right.
Case 1: HDFC Bank, March 2020 — the business was unchanged
March 2020. Nifty 50 fell from 12,000 to below 7,600 in 40 days: a 37% collapse driven by COVID-19 panic. HDFC Bank fell from roughly ₹1,290 to ₹738 over the same period.
At ₹738, HDFC Bank’s net NPA was still below 1.5%, its capital adequacy ratio was above regulatory minimums, and its loan book had not yet shown stress. The price collapsed because the entire market collapsed: investors were selling anything liquid to raise cash.
Someone who averaged down on HDFC Bank in March 2020 at ₹738 and held through to December 2020 saw the stock recover to above ₹1,400: close to a 90% return on the averaged-down tranche in nine months.
The thesis was clear: same bank, same management, same fundamentals, price driven down by external panic. The stock’s fall was the market’s problem, not HDFC Bank’s.
Case 2: Yes Bank, 2018–2020 — the business had broken down
Yes Bank traded at ₹400 in early 2018. By late 2018 it was at ₹200. By 5 March 2020, the RBI had imposed a moratorium and the stock was trading at ₹5.
At each step of the decline — ₹400, ₹200, ₹100, ₹50 — the same surface argument applied: “the average cost is lower, the stock can’t fall much further.” Investors who averaged down at each level found themselves doubling, then tripling, then quadrupling their exposure to a bank whose loans to distressed companies were rising, whose auditors were raising questions, and whose promoter was losing credibility with regulators.
At ₹5 — a 98.75% decline from the ₹400 peak — every averaging-down buyer had lost more capital on their additional purchases than they had recovered from the lower average.
The business was not temporarily repriced by the market. The business itself was deteriorating, steadily and visibly, for two full years before the collapse.
The opportunity cost nobody calculates
When you deploy additional capital into a falling stock, that capital has an alternative. At minimum, you could have bought a Nifty 50 index fund.
Between January 2019 and March 2020: a Nifty 50 index fund fell roughly 23% from its January 2019 level. A bad outcome — but recoverable. Nifty recovered all losses by December 2020 and was up roughly 40% from January 2019 by early 2021.
Yes Bank over the same period: down 98%.
The averaging-down buyer in Yes Bank did not just lose the capital they deployed. They also lost the compounding that capital would have earned elsewhere during the same period.
Opportunity cost is invisible in the break-even calculation. It only shows up in your XIRR when you finally close the position: which is why XIRR, not break-even price, is the honest measure of an averaging-down strategy.
Three questions before averaging down
Ask all three. If the answer to any one is “I don’t know” or “no”, do not average down.
1. If I had not already owned this stock, would I buy it today at today’s price?
This is the most important question. It removes the sunk cost from the calculation. You are not buying to recover losses: you are making a new investment decision. If the stock does not pass fresh scrutiny at today’s price, averaging down is an emotional response to a prior mistake, not a rational allocation.
2. Is the business materially the same as when I first bought it?
Compare the most recent quarterly results to the results when you bought: revenue trend, net profit margin, debt levels, promoter pledging, auditor comments. If you cannot confidently say the business is intact — not “it should recover”, but “the fundamentals are unchanged” — the price fall may be informational, not noise.
3. If I average down and the stock falls another 30%, what will my portfolio look like?
Model it. If a further 30% fall would make this position more than 20–25% of your portfolio, or would force you to sell other holdings for liquidity, you are averaging yourself into a concentrated, illiquid situation. Portfolio construction matters more than break-even arithmetic.
Averaging down vs. rupee cost averaging: not the same thing
Rupee cost averaging (RCA): a fixed amount invested at regular intervals regardless of price, usually into a diversified fund. Averaging down: buying more of a specific stock that has already fallen.
The confusion is understandable: both result in buying more units when prices are lower. But the structural difference matters.
RCA works on diversified instruments because you are capturing market-wide volatility, not betting on one company. You are indifferent to whether the price fell because of market panic or because of something worse: the basket will eventually recover even if some individual companies do not.
Averaging down on a single stock requires the company-specific thesis to be intact. If it is, the mechanism can work. If it is not, the mechanism amplifies the loss.
Use RCA for your index fund SIPs. Use the three-question test before averaging into any individual stock position.
Source notes
- HDFC Bank NSE price history (March 2020 low of ₹738.35, close above ₹1,400 by December 2020): NSE India historical data, nseindia.com.
- Yes Bank NSE price history (₹400 in early 2018; RBI moratorium announced 5 March 2020; stock at ₹5 on 9 March 2020): NSE India historical data, nseindia.com.
- RBI moratorium on Yes Bank: Reserve Bank of India press release, “Yes Bank Ltd — Moratorium”, 5 March 2020, rbi.org.in.
- Weighted average cost formula: standard portfolio mathematics.
- Capital gains and cost-basis method for listed shares: CBDT Rule 8AA, Income Tax Rules 1962.
Further Reading
Newspapers & Magazines 1 articles
Investing Blogs 2 articles
The Dangers of Averaging Down
Vishal Khandelwal argues that averaging down is dangerous when the underlying business is deteriorating — a core lesson Indian retail investors repeatedly ignore.
Rupee Cost Averaging via SIP Has No Benefit Other Than Accumulating MF Units
M. Pattabiraman shows with Sensex data that the so-called averaging benefit of SIPs is a myth — returns and risk converge with the market regardless of how units were accumulated.
See the trade-off in real time
each lot = same size as your original purchase
Current price: ₹300 −40% from ₹500
The upside: lower break-even
₹400
new average cost
The downside: more exposure
+60%
more capital in one stock
If it falls another 40%, the averaged lots alone lose ₹120.