InverseOne

Did buying the KOSPI at every new thousand-point mark leave you stuck at the top?

Published 2026-09-13 23:21 KST

The KOSPI first closed above a new thousand-point mark seven times from 3,000 through 9,000, and six of those crossings fell within 234 days. Buying 10M won of KODEX KOSPI on each crossing day leaves 70M won worth 90.37M won (+29.1%) at the 2026-09-11 close. The three accounts opened at the 7,000, 8,000 and 9,000 crossings sit below principal, with losses from 6.7% to 24.2%. The 3,000 account now has the largest gain at +129.7%, yet it spent 900 trading days below principal after the purchase.

Hypothesis

"Buying on the day the KOSPI first crosses a new thousand-point mark leaves you stuck at the top with a loss."

Experiment setup

On the first day the KOSPI closed above each thousand-point mark from 3,000 through 9,000, 10M won of KODEX KOSPI was bought at that day's close. Only closing crossings counted, not intraday ones, and nothing was sold afterwards. Distributions, fees, taxes and tick sizes are excluded, and fractional shares are assumed.

First purchase
2021-01-07
Valuation date
2026-09-11
Total invested
₩70M
Stocks
KODEX KOSPI

Result

Final value

₩90.37M+29.1%

Half right
Seven accounts combined₩90.37M
Total principal invested₩70M

Trend chart

Unit: KRW
Seven accounts combinedPrincipal invested
150M75M2021-01-292023-11-302026-09-11

Month-end closes. The dotted line steps up 10M won at each crossing, and until October 2025 only the 3,000 account exists.

Verdict

Half right

The four accounts opened at the 3,000 through 6,000 crossings are in profit, and the three opened at 7,000 through 9,000 are at a loss. All seven together are up 29.1%.

Seven crossing days

The KOSPI first closed above a new thousand-point mark seven times. It crossed 3,000 on 2021-01-07 at 3,031.68, 4,000 on 2025-10-27 at 4,042.83, 5,000 on 2026-01-27 at 5,084.85, and 6,000 on 2026-02-25 at 6,083.86. It crossed 7,000 on 2026-05-06 at 7,384.56, 8,000 on 2026-05-26 at 8,047.51, and 9,000 on 2026-06-18 at 9,063.84. The climb from 3,000 to 4,000 took 1,754 days, while 4,000 to 9,000 took 234 days, and the last two crossings came 20 and 23 days apart. At each of these seven closes, 10M won of KODEX KOSPI was bought.

Four in profit, three at a loss

KODEX KOSPI closed at 70,815 won on the 2026-09-11 valuation date. The 3,000 account is worth 22.97M won (+129.7%), the 4,000 account 17.12M won (+71.2%), the 5,000 account 13.55M won (+35.5%), and the 6,000 account 11.29M won (+12.9%). The 7,000 account is worth 9.33M won (-6.7%), the 8,000 account 8.53M won (-14.7%), and the 9,000 account 7.58M won (-24.2%). The rule was fixed before calculating: if all seven accounts showed a loss the hypothesis would hold, and if none did it would be rejected. Three show a loss, so the verdict is Half right. Taken together, 70M won grew to 90.37M won, a gain of 29.1%.

All seven accounts were in profit on June 22

On 2026-06-22 the KOSPI set its record close of 9,114.55, and KODEX KOSPI also closed at its high of 93,650 won. Even the 9,000 account was 0.2% above principal that day, and the seven accounts were worth 119.52M won in total (+70.7%). The KOSPI then fell 38.6% from its peak to 5,593.56 on 2026-07-30. That day the 9,000 account was worth 6.15M won (-38.5%), the 8,000 account 6.92M won (-30.8%), and the 7,000 account 7.57M won (-24.3%). On the valuation date the KOSPI stood at 6,909.91, 24.2% below its peak.

The 3,000 account's 900 trading days

The 3,000 account now has the highest return, but it spent 900 of its 1,370 trading days below principal. Its longest such period lasted 888 trading days, from 2021-10-01 through 2025-06-23. At the low on 2022-09-30, its 10M won was worth 7.12M won (-28.8%). By the same measure, the 4,000 account was below principal for 19 trading days, the 5,000 account for just 1 day, and the 6,000 account for 33 trading days. Among the three accounts now at a loss, the 9,000 account never rose above principal from 2026-06-23 through the valuation date. The same was true for the 8,000 account from 2026-07-07 and the 7,000 account from 2026-07-13.

Key numbers

Seven accounts combined
90.37M won (+29.1%)
Accounts in profit
4 (3,000, 4,000, 5,000, 6,000)
Accounts at a loss
3 (7,000 -6.7%, 8,000 -14.7%, 9,000 -24.2%)
From 4,000 to 9,000
234 days (1,754 days from 3,000 to 4,000)
KOSPI from its peak
-24.2% (peak 9,114.55, 2026-06-22)
Days the 3,000 account spent below principal
900 trading days (of 1,370 held)

FAQ

Would the dates differ if the first intraday crossing were used?
Three dates would differ. On an intraday basis, 3,000 was first crossed on 2021-01-06, 5,000 on 2026-01-22, and 8,000 on 2026-05-15. The other four crossings fall on the same day either way. This report uses closes so that each mark has a single date.
Would buying KODEX 200 instead of KODEX KOSPI produce the same result?
The pattern is the same, with four accounts in profit and three at a loss. Starting with the 3,000 account, the returns are +162.6%, +92.9%, +46.4%, +20.5%, -3.3%, -13.6%, and -25.5%. The seven accounts are worth 97.99M won in total (+40.0%).
How far are the three losing accounts from principal?
Based on the valuation-date close of 70,815 won, the 7,000 account's purchase price of 75,900 won is 7.2% higher. The 8,000 account's purchase price of 83,000 won is 17.2% higher, and the 9,000 account's purchase price of 93,450 won is 32.0% higher. These gaps measure price alone and exclude distributions.

How this was calculated

Daily closing prices for the KOSPI index (^KS11) and KODEX KOSPI (226490.KS) from Yahoo Finance were used. A crossing day is the first day the KOSPI closed above 3,000, 4,000, 5,000, 6,000, 7,000, 8,000 or 9,000, and intraday highs were not used. On each crossing day 10M won of KODEX KOSPI was bought at the close, and every account was valued at the 2026-09-11 close of 70,815 won. Entry prices were 30,830 won, 41,365 won, 52,245 won, 62,720 won, 75,900 won, 83,000 won and 93,450 won in order. Distributions, commissions, taxes and tick sizes are excluded, and fractional share purchases are assumed. Yahoo closes do not reflect distributions, and all seven accounts were held through a distribution date, so reinvesting would put every actual result above the figures here. The verdict rule was fixed before calculating: the hypothesis holds if all seven accounts show a loss on the valuation date, is rejected if none do, and is Half right otherwise. The KODEX 200 (069500.KS) comparison used the same dates and the same rule.

Data sources

  1. [1] KOSPI index (^KS11) historical prices, Yahoo Finance, 2026-09-11
  2. [2] KODEX KOSPI (226490.KS) historical prices, Yahoo Finance, 2026-09-11
  3. [3] KODEX 200 (069500.KS) historical prices, Yahoo Finance, 2026-09-11

Stocks in this report

These are hypothetical results computed from past prices. A real trade would differ with fees, taxes, lot sizes and execution timing, and past returns say nothing about future returns. This is not a solicitation to buy or sell any security. The money flow and stock signals this service shows are estimated, for-reference information based on price and volume. It is not a solicitation to buy or sell any security, nor investment advice. It is a prior-close daily batch, not real-time quotes, and all investment decisions and responsibility are your own. The Trade button is a convenience link to your chosen brokerage's app/website; orders are placed directly at the brokerage and InverseOne is not involved in any order.