Averaging down a stock: what the new average hides

The comfort of a lower number

A share you bought at ₹1,000 now trades at ₹600. Buying more at ₹600 brings your average down, and the price you need to "get back to" moves closer. Averaging down feels like fixing a mistake. Sometimes it is a reasonable decision, but the average price is a summary, and it hides most of what has changed about the position.

This post works through one example in rupees. The share is hypothetical, and nothing here is a view on any of the stocks we track.

The example

You have a ₹10 lakh portfolio. ₹1 lakh of it, 10%, is 100 shares of one company bought at ₹1,000. The share falls 40% to ₹600. The rest of the portfolio is flat.

You decide to average down and buy 200 more shares at ₹600, ₹1.2 lakh, moved from elsewhere in the portfolio. The stock average calculator gives the new figure:

(100 × ₹1,000 + 200 × ₹600) ÷ 300 = ₹2,20,000 ÷ 300 = ₹733.33

Here is what the average does and does not tell you.

The top half of the table is what averaging down promises. The bottom half is what it costs.

What the new average hides

1. The position got bigger, not just cheaper. The fall had already shrunk this share to 6.3% of the portfolio. Averaging down tripled it to 18.8%, nearly double where it started, in a company the market has just marked down by 40%. You did not lower your risk, you increased your exposure to the one holding that has been going wrong.

2. Break-even is not the same as recovery. The 22.2% rise only gets you back to ₹2.2 lakh, which is now more money than you started with in this share. The share itself still needs to rise 66.7% to reach ₹1,000. Losses and recoveries are asymmetric: a 20% fall needs a 25% rise, 40% needs 66.7%, 50% needs 100% and 75% needs 300%. The average does not change the asymmetry. It only spreads it over more shares.

3. The original price means nothing to the market. ₹1,000 is your number, not the company's. The question that matters at ₹600 is whether you would buy this share today at this price, in this size, if you held none. If the answer is yes, buying more is a fresh decision on its own merits. If the answer is "only because I'm already down", that is loss aversion dressed up as strategy.

4. A further fall hurts more. A fall from ₹600 to ₹300 costs ₹70,000 on the original position and ₹1.3 lakh on the averaged one, 13% of the original portfolio. A company that has fallen 40% can fall further. Even the broad market can: the Nifty 50 fell about 60% between January and October 2008, and a single share has no floor above zero. The maximum drawdown of one stock is not bounded by an index's.

The ladder that runs away

Averaging down rarely happens once. The same logic that bought at ₹600 wants to buy again at ₹400, usually in a bigger lot, because a bigger lot moves the average further.

By the fourth buy, a ₹1 lakh position has become ₹4.6 lakh, almost half of the original ₹10 lakh portfolio, in a share that has fallen 60%. Each step looked modest at the time. This is how a diversified portfolio turns into a bet on one company without anyone deciding it should.

When it can be reasonable

Averaging down is not wrong in itself. It is a decision to add capital, and should be judged like any other purchase:

• The case for owning it has not changed. The price fell for reasons you can name, and they don't touch the business you bought. A fall caused by bad news about the business is a different situation from a fall caused by the whole market.
• You set a size limit before you start. Decide the most you will ever hold in one share, say 5% or 10% of the portfolio, and stop there, whatever the average says. A core and satellite split is one way to fix that ceiling in advance.
• The money is new, not borrowed. Averaging down with margin adds interest and the risk of a forced sale at the bottom.
• You plan for it going lower. If the share halves again, can you hold without selling? If not, the position is already too large.

There is a reason averaging works in a SIP: rupee cost averaging into a diversified fund buys more units when prices are low without concentrating on one company that can fail. The same arithmetic applied to a single share carries that single company's risk.

Costs and tax

The average ignores charges. Each extra purchase pays STT, stamp duty and exchange fees; the brokerage calculator shows the full stack and what trading actually costs explains it. They are small on one delivery buy and add up across a ladder.

Tax ignores the average entirely. Gains and losses are worked out lot by lot, first in, first out, per demat account. Sell 100 of your 300 shares at ₹733 and the ₹1,000 lot is deemed sold first, a ₹26,667 loss on paper, even though your position shows no gain or loss. Each lot also carries its own holding period: a lot held 12 months or less is taxed at 20% as short-term gain, and a lot held longer at 12.5% above the ₹1.25 lakh yearly exemption, at FY 2026-27 rates. A loss on the old lot can be useful for tax-loss harvesting, but only if you mean to sell.

The rule of thumb

Judge the buy, not the average. Before you average down, write down three numbers: what share of the portfolio this stock will be afterwards, what you lose if it falls another 50%, and the largest size you will ever let it reach. If any of them makes you uncomfortable, the lower average was never the point. For more on why falling prices push people towards bad decisions, see the psychology of a market crash.