주식 평균 단가 계산 원리와 물타기 전 점검 사항

Average cost is a weighted average

When you buy the same stock in several lots, your broker shows one average cost. It is not the plain average of the prices; it is weighted by quantity: average cost = sum of (shares × price) ÷ total shares. Buy 100 shares at 50,000 and 100 at 40,000 and you have put in 9,000,000 for 200 shares, so the average is 45,000. Only when quantities are equal does it land midway; otherwise it is pulled toward the lot you bought more of.

Say you hold 100 shares at 50,000 and add 300 at 40,000. Total cost is 5,000,000 + 12,000,000 = 17,000,000 for 400 shares, an average of 42,500, much closer to 40,000 than to 45,000. How much a purchase lowers your average depends on how many shares you buy, not just how cheap they are.

그림: 두 번에 나눠 산 가격이 수량 비중대로 섞여 평균 단가가 됩니다. 수량이 같으면 평균은 두 가격의 정중앙입니다.

Try it yourself

The interactive tool below (Korean UI) works like this: enter one row per purchase (shares and price), then optionally the current price, extra shares to add and a target average. It returns the average cost, the unrealized gain or loss, the rise needed to break even and the number of shares required for a target average. Nothing you type is sent to a server.

Fees, taxes and currency

The formula above ignores fees. Your broker's displayed average can differ because brokers differ in whether buy commissions are folded into the average, how fractions are rounded and, for foreign stocks, which exchange rate is used. The broker's statement is the reference. Tax rules vary widely by country, product and holding period, so this article does not cover them; check the official guidance of your tax authority or broker.

To include fees yourself, enter each purchase at its all-in price (total paid ÷ shares). Selling costs come on top, so the real break-even rise is slightly higher than the figure shown.

Recovering a loss is asymmetric

A common misconception is that a 20% drop needs a 20% gain to recover. After a drop of x, the gain required is 1 ÷ (1 − x) − 1. From 100 to 80 you need 25% to get back to 100. The deeper the hole, the steeper the climb. This is pure arithmetic, so no outside source is needed.

That asymmetry makes the plan "sell when I'm back to even" less realistic as losses deepen. Averaging down seems to shorten the distance, but it is not free.

What averaging down changes, and what it does not

You bought 100 shares at 50,000 and the price is now 40,000, down 20%. Your unrealized loss is 100 x 10,000 = 1,000,000 and you need +25% to break even. Buy another 100 at 40,000 and the average becomes 45,000, so the rise needed falls to 12.5% (45,000 / 40,000 - 1). On paper that looks better.

But more than the break-even distance changed. Money invested went from 5,000,000 to 9,000,000, and the amount riding on this one stock doubled from 4,000,000 to 8,000,000 at the current price. Right after the purchase the unrealized loss is 200 x 5,000 = 1,000,000, the same as before. If the price then falls to 30,000, the loss without the add would be 100 x 20,000 = 2,000,000; with it, 200 x 15,000 = 3,000,000. The average fell, but your loss became more sensitive to each further drop. Averaging down does not reduce loss; it puts more money on the same direction.

Lowering the average versus adding to what is working

Where extra money goes depends on one question: is your view of this stock better than when you first bought? If the only reason is that the price is cheaper, that price may reflect new information. If your original thesis still holds and your conviction has grown, an add has a reason. Either way, the test is not "is it below my average" but "would I buy this at this price and this size if I were starting today?" The market does not know your average cost.

Adding to a rising position confirms the thesis before increasing size, which is different, but it also raises your average and total exposure. Neither is clearly better; what matters is whether the added amount is affordable and whether you wrote down your reasoning.

Decide position size first

A typical path to trouble is a position growing without a plan. A simple guard is to set the maximum loss you will accept before buying. Divide that loss by (entry price minus the price where you would exit) to get shares. With 50,000,000 of investable assets and a 1% (500,000) limit per stock, entering at 50,000 with a reference exit of 45,000 gives 500,000 / 5,000 = 100 shares. These figures are illustrative, not a recommended percentage.

The point is to set the limit before you buy; after a loss the limit tends to become something to rationalize. If you average down, first calculate whether total size and potential loss still fit within the original limit. If not, you are changing the plan, and that should be a separate decision.

Psychological traps

• Sunk cost. Money already spent is gone, yet it distorts the next decision. Judge only on what you expect from here.
• Anchoring on your average. Fixing the goal at "sell above my average" ties it to the past, while the market has no idea what you paid.
• Loss aversion. Losses feel worse than equal gains, which encourages postponing the decision and taking bigger risks to avoid realizing a loss.
• "Scaling in is safe." Spreading purchases over time spreads entry timing, not the risk of a stock that keeps falling.
• Confirmation bias. As a position falls, you collect only supporting evidence. Deliberately read opposing views and objective figures such as earnings and filings.

These are widely used behavioral-finance terms explained in plain language; no numerical claims are made about them.

Planned scaling-in versus improvised averaging down

Both look like "buying in pieces", but they are different things. A planned scale-in fixes the total target size, the number of purchases and the condition for each step in advance, so total exposure sits within the limit from day one; a falling price simply triggers the next scheduled step. Improvised averaging down decides amount and count after seeing the drop. It is easily driven by the pain of a loss and has no ceiling on exposure. The test is simple: were the amount and conditions of the add written down before the decline? If not, it is a reaction, not a plan.

Also check for an exit condition. Without a price or event at which you stop adding and reassess, averaging down has no endpoint. Write down a cap on the number of adds and the facts that would reverse your thesis, and the average-cost calculation becomes a management tool rather than a rationalization.

Questions to ask before adding

1. Is the original reason for owning this still valid? What changed?
2. Would I buy this price and size if I were starting today?
3. What share of my total assets would it be after adding, and is it inside my limit?
4. Is the money separate from living costs and emergency funds?
5. Could I hold if the price fell another 20 to 30%? What is my plan then?
6. Is the goal to lower the average, or do I have real conviction?
7. Would I reach the same conclusion tomorrow?

Where the Multagi app fits

Multagi, made by SETLOG, is an iPhone calculator for running these numbers quickly. Per its App Store description it does two things: compute the average price automatically, and compute how many additional shares are needed to reach a target average. SETLOG's privacy policy states that all calculation happens offline on the device and no personal or financial data is collected. There are no screenshots here; the diagram and the tool above show the same math. See the Multagi guide for usage.

Sources and further reading

1. The figures here (weighted average, 1/(1-x)-1 break-even gain, position-size formula) are arithmetic and need no outside source; verify them with the tool above. All example numbers are hypothetical.
2. Multagi guide — app features and limits (per App Store description)
3. SETLOG privacy policy — the statement on offline calculation

FAQ