What the -6.12% KOSPI Crash Really Means — Why You Have to Check the Macro Gate Before Hunting Relative-Strength Names

Fri, 15 May 2026 00:00:00 +0000

On May 15 the KOSPI fell -6.12% — the day right after the all-time high of 8,000. KOSDAQ -5.14%. The sell-side sidecar circuit breaker fired. On a day like this, investors face two temptations. One is to hunt for names that didn’t fall, looking for the next leadership stock. The other is to chase, saying “everything fell, so it’s cheap.” Both are dangerous. The right read is not stocks — it is the macro gate: US long rates, oil, USD/KRW, VIX. Until those four stabilize, chasing even a great stock is inefficient.

Key takeaways

May 15 Korea: KOSPI 7,493.18 (-6.12%), KOSDAQ 1,129.82 (-5.14%). Sell-side sidecar fired right after a fresh all-time high.
Cause: not simple profit-taking. US PPI +6.0% (largest since 2022), Japan PPI +4.9%, JGB 10-yr at the highest since 1997, US 10-yr 4.46%, Brent 108, USD/KRW near 1,500 — a global macro cycle pressuring risk assets simultaneously.
There are RS names: Hana Micron +18.6%, Jeju Semiconductor +8.9%, SFA Semiconductor +4.0%, Samsung Electro-Mechanics -1.4% (RS vs index +4.7pp). Doosan Robotics +19.3%, Robostar +9.9%.
But chasing is wrong: “didn’t fall today = next leader” is a fallacy. If macro doesn’t normalize, RS names follow the index down the next session.
The real read — macro gate: US 10-yr below 4.45%, Brent below 105, USD/KRW below 1,480, VIX below 18. At least 2–3 of those must clear before new buying is rational.
What today’s RS screen is actually for: not a “buy signal,” but a “watchlist-generation signal.”

1. What happened

1.1 Korea

May 15 close:
KOSPI: 7,493.18 (-6.12%)
KOSDAQ: 1,129.82 (-5.14%)

Notes:
- Briefly hit a fresh ATH intraday, then reversed sharply
- Sell-side sidecar (5-min suspension of program selling) fired
- Massive foreign and institutional net selling; retail absorbed
- A 6%+ single-day decline — the largest since August 2024.

This is too large to read as ordinary profit-taking.

1.2 Korea wasn’t alone

Same day, global markets:
- US Nasdaq futures down sharply
- Nikkei lower
- Mainland China sold on summit disappointment

Drivers:
US April PPI +6.0% (largest since March 2022)
US April CPI +3.8% (highest since May 2023)
Japan April PPI +4.9% (3-year high)
JGB 10-yr 2.55% (highest since 1997)
US 10-yr 4.46%, 30-yr 5.02%
Brent \~108 USD
USD/KRW near 1,500.

1.3 Is this just volatility, or is it the cycle?

The “global macro cycle” we mapped in an earlier post actually executed.

The cycle in motion:

Iran / Hormuz → oil 108 USD
 ↓
US PPI +6.0%, CPI +3.8%
 ↓
Fed-cut hopes priced out (Dec hike prob 36%)
 ↓
US 10-yr 4.46%, 30-yr 5.02%
 ↓
Japan PPI +4.9% (import prices +17.5%)
 ↓
BOJ June-hike prob 77%
 ↓
JGB at the highest since 1997
 ↓
Global discount rate up
 ↓
Risk assets squeezed simultaneously
 ↓
KOSPI -6.12%

→ Not a one-off shock —
 the "settlement day" of a multi-day cycle.

2. The trap of “RS hunting” on a day like this

2.1 What relative strength is

Relative Strength (RS):
stock return − index return

If KOSPI fell -6.12% and a stock rose +5%:
→ RS = +5% − (-6.12%) = +11.12pp
→ "11.12pp stronger than the index"

Why look at it:
→ Strong demand survives the index decline
→ Tends to lead on the recovery
→ A screening tool for next-leadership candidates.

2.2 Today’s top RS names

Rank	Name	Return	RS vs index	Turnover
1	Samti	+30.00%	+35.1pp	KRW 84B
2	Samji Electronics	+29.97%	+35.1pp	KRW 77.5B
3	Doosan Robotics	+19.29%	+25.4pp	KRW 1.08T
4	Hanson Engineering	+19.18%	+24.3pp	KRW 60.6B
5	Hana Micron	+18.61%	+23.8pp	KRW 688.8B
6	Jahwa Electronics	+16.53%	+22.7pp	KRW 84.8B
10	LG Electronics	+10.83%	+17.0pp	KRW 1.44T
12	Robostar	+9.90%	+15.0pp	KRW 188.6B
13	Jeju Semiconductor	+8.86%	+14.0pp	KRW 884.6B
17	SFA Semiconductor	+3.99%	+9.1pp	KRW 280.8B
18	Samsung SDS	+3.34%	+9.5pp	KRW 366.7B
20	Samsung Electro-Mechanics	-1.37%	+4.7pp	KRW 1.41T

2.3 Trap 1 — confusing “limit-up momentum” with “earnings strength”

The RS top list mixes two very different things:

A. Earnings-backed strength (good RS):
 - Hana Micron: 1Q26 OP KRW 72B (+513.6%)
 - Jeju Semiconductor: 1Q26 OP KRW 67.1B (+1,713%)
 - Samsung Electro-Mechanics: MLCC price hike + AI parts thesis
 → Real fundamentals.

B. Theme / news pop (risky RS):
 - Samti limit-up +30%
 - Samji Electronics limit-up +30%
 - Hanson Engineering +19%
 → One-day catalyst or flow event.

Treating limit-up names as "RS #1–2" is dangerous:
→ Likely to gap down and form a red candle the next day
→ Not real leadership candidates.

2.4 Trap 2 — without “close location,” every RS looks alike

Close location = (close − low) / (high − low)

Same +5% gain, very different read:
- Close location 90% → buyers held through the close (good)
- Close location 20% → distributed from intraday highs (weak)

In today's data:

Name Return Close loc Read
Hana Micron +18.6% 71.7% ★ buying held late session
Jeju Semi +8.9% 62.2% Good, but watch-listed risk
SFA Semi +4.0% 19.5% ✗ pushed intraday (low-quality RS)
Samsung E-M -1.4% 19.6% ✗ outperformed index but
 distribution on the candle

→ By return alone, SFA looks "strong"
→ By close location, sellers actually dominated
→ Same RS rank, very different quality.

2.5 Trap 3 — if macro doesn’t clear, RS names get dragged down the next session

Historical pattern after a major down day:

A. Scenario (macro recovers):
 → Index rebounds
 → RS names lead higher (+10–15% possible)
 → They graduate into actual leadership

B. Scenario (macro worsens):
 → Index falls more
 → RS names get chase-sold
 → Yesterday's strong names mean-revert -10–15%
 → Short-term traders stop out

The differentiator between A and B is macro, not the stock.
→ You have to read macro BEFORE you trade the stock.

3. The macro gate — what to actually watch on Monday

3.1 Gate 1: US long rates

Now:
US 10-yr: 4.46%
US 30-yr: 5.02%

Recovery signals:
US 10-yr below 4.45% (first)
US 30-yr below 5.0% (stronger)

Why it matters:
→ Anchor of the global discount rate
→ Key driver of foreign flows into Korea
→ Sets the multiple for growth / semis / robotics.

Current: 🔴 Red

3.2 Gate 2: Oil / geopolitics

Now:
Brent: 108–109 USD
Hormuz: still functionally closed

Recovery signals:
Brent below 105 USD (first)
Brent below 100 USD (stronger)
Hormuz reopening headlines

Why it matters:
→ Origin of the US / Japan inflation pulse
→ Trigger variable for the whole cycle
→ Above 100 USD, Fed-cut hopes don't return.

Current: 🔴 Red

3.3 Gate 3: USD/KRW

Now:
USD/KRW \~1,495–1,500

Recovery signals:
USD/KRW below 1,480 (first)
USD/KRW below 1,450 (stronger)

Why it matters:
→ Direct gauge of foreign-selling pressure
→ KRW weakness = lower Korean asset appeal
→ Above 1,500 invites more selling.

Current: 🟡 Yellow → Red

3.4 Gate 4: VIX

Now:
VIX \~18.6

Recovery signals:
VIX below 18 (first)
VIX below 16 (stronger)

Why it matters:
→ Gauge of global risk aversion
→ Above 20 = panic
→ Below 16 = risk-on.

Current: 🟡 Yellow

3.5 Gate summary

Status of the 4 gates:
🔴 Rates: Red
🔴 Oil: Red
🟡 FX: Yellow → Red
🟡 VIX: Yellow

→ 0 of 4 are Green
→ Inefficient to add risk
→ Even RS names should NOT be chased.

4. A macro-gated trading frame

4.1 Gate-pass conditions

To size up new buying, at least 2–3 of the 4 gates must clear.

Specifically:
1. US 10-yr falls below 4.45%
2. Brent falls below 105 USD
3. USD/KRW stabilizes below 1,480
4. VIX falls below 18
5. KOSPI / KOSDAQ holds the prior session's low and closes green

At least 2–3 met → partial buying allowed.
1 or fewer met → no buying.

4.2 Action by scenario

Scenario	Conditions	Action
A. Macro recovers	3+ of 4 gates clear	Scaled buying in RS names; max 30–40% of full size
B. Macro neutral	1–2 gates clear	Just observe the close; minimize new buying
C. Macro worsens	Gates deteriorate further	No new buying; only re-evaluate existing positions for thesis damage

4.3 Relationship between “RS buys” and the macro gate

Wrong order:
RS scan → buy → check macro

Right order:
Check macro gate → if gates clear, screen RS → then add risk

Why:
- If macro doesn't clear, RS names follow the index down
- Buying great stocks at the wrong price = bad outcome
- Real leadership isn't built in a day — it shows up over days / weeks
 of sustained relative strength.

5. If the macro gate clears — name priority

When the gates pass, work the list in this order.

5.1 #1 — Hana Micron

Fundamentals:
1Q26 revenue KRW 507.7B (+62.8% YoY)
1Q26 OP KRW 72B (+513.6% YoY)
Margin 14.2%
Beats consensus operating profit by +29.7%

Close location: 71.7% (strong end-of-day demand)
Turnover: KRW 688.8B (ample)

Read:
- Cleanest earnings-backed RS candidate
- KIS target KRW 55,000 is close (current 52,900, +4% upside)
- Prefer a pullback or sustained turnover over chasing

Entry conditions:
- Hold 52,900 on the close
- Turnover stays above KRW 340B (50% of today)
- RS vs KOSDAQ +3pp or better.

5.2 #2 — Samsung Electro-Mechanics

Fundamentals:
1Q26 revenue KRW 3.21T
1Q26 OP KRW 280.6B (+40% YoY)
MLCC price-hike cycle starting
NH target lifted to KRW 1.5M

Close location: 19.6% (intraday distribution)
Turnover: KRW 1.41T (large-cap volume)

Read:
- Structurally the best large-cap candidate
- After +23% in May, today is a down candle
- Wait for KRW 1,024,000 (prior close) to be reclaimed

Entry conditions:
- Reclaim KRW 1,024,000 (yesterday's close)
- Stronger signal: hold above KRW 1,061,000 (today's open).

5.3 #3 — Jeju Semiconductor

Fundamentals:
1Q26 revenue KRW 180.4B (+273%)
1Q26 OP KRW 67.1B (+1,713%)
Margin 37.2%

Risk:
"Investment caution" designation
Trading-halt risk on further surges
Customer concentration (A 72%, China 73%)

Read:
- Numbers are strong but trading is difficult
- Volatility makes chasing inefficient
- Want a settle in the KRW 80k zone + 2–3 sessions of digestion candles.

5.4 #4 — SFA Semiconductor

Fundamentals:
4Q25 turned profitable
1Q26 expected to stay profitable (estimate only — not yet confirmed)

Close location: 19.5% (intraday push-down)
Turnover: KRW 280.8B

Read:
- Back-end thesis is right, but earnings proof is still light
- Low close quality hurts the RS quality
- Wait for 1Q26 disclosure.

5.5 Names to exclude from chasing

High RS, but no chase:

1. Limit-up pops (Samti, Samji Electronics):
 - One-day catalyst / flow event
 - Likely to gap down

2. Needs verification (Hanson Engineering):
 - Turnover only KRW 60B
 - Could be a one-off rather than a trend

3. Low close quality (SFA, today's Samsung E-M candle):
 - RS-positive on the day
 - But low close location = sellers dominated.

6. Monday checklist

Before the open (KST 08:00–09:00):

1. Did US 10-yr drop below 4.45%?
2. Did Brent drop below 105 USD?
3. Did USD/KRW move into the 1,480s?
4. Did VIX drop below 18?

→ How many of these are Green?

Intraday (09:00–15:30):

5. Did KOSPI / KOSDAQ gap down and recover, or gap up and fade?
6. Are foreigners net-buying or at least slowing selling?
7. Are Samsung Electronics / SK hynix recovering steadily?

→ How many of these are positive?

Close (15:30–16:00):

8. Did Hana Micron hold KRW 52,900?
9. Did Samsung E-M reclaim KRW 1,024,000?
10. Did turnover stay at 50%+ of today?

→ All 3 positive AND 2–3 macro gates green = scaled buying possible
→ 1 or fewer macro gates green = no new buying.

7. The core lessons

7.1 “Didn’t fall today” ≠ “next leader”

Wrong inference:
"The market fell, but this stock didn't"
→ "It's strong"
→ "It'll be strong tomorrow too"
→ "Buy"

Right inference:
"The market fell, but this stock didn't"
→ "There was real demand today"
→ "For tomorrow to be strong too, macro has to stabilize"
→ "If macro confirms AND turnover holds, then it's a buy candidate".

7.2 Leadership is built over many sessions, not one

Strong for 1 day ≠ leadership
Strong for 1 week = candidate
Strong for 1 month = leadership

Today's RS #1 is not automatically tomorrow's leader.
Only after sustained RS over the next 5–10 sessions
does a name actually earn the leadership label.

→ Today's buying is "betting on a candidate," not "buying the leader."
→ If macro doesn't clear, candidates all give back.

7.3 Stock work < macro gate

The core message of this post:
"Cycle before the stock."

A great company bought at a high price at the wrong time
ends poorly.
A mediocre name bought as macro recovers
ends fine.

→ Spend even 30% of your stock-research energy
→ on the 4 macro gates every day
→ and the quality of your entry timing improves disproportionately.

8. How this links to other posts

The macro cycle synthesis piece:
→ "Iran / Hormuz → oil → US inflation → long rates → global discount rate"
→ Today's KOSPI -6% is the "settlement day" of that cycle.

The Japan PPI piece:
→ Mechanism for the Japan axis — weaker marginal bid for USTs
→ Connected to today's foreign selling.

The US-China summit piece:
→ "Find the least-priced pocket"
→ Today's RS names are exactly those candidates.

The Samsung Electro-Mechanics piece:
→ "At KRW 1.02M, chasing is inefficient"
→ Today -1.4%; the 1,024,000 level broke
→ Wait for a reclaim before buying.

The consumer-rotation piece:
→ "Capital broadened to earnings-backed consumer staples"
→ Today Samyang Foods +0.35% (defensive)
→ Still a valid rotation candidate.

9. The one-line bottom line

May 15’s KOSPI -6.12% is not simple profit-taking. It is the day the global macro cycle settled in the Korean market. US PPI +6%, Japan PPI +4.9%, US 10-yr 4.46%, Brent 108, USD/KRW near 1,500 — every one of those variables pressed Korean risk assets in the same direction.

On days like this, there are two temptations. One is to hunt the names that didn’t fall in search of the next leader. The other is to chase, saying “everything fell, so it’s cheap.” Both are inefficient if the macro gate doesn’t clear.

The first thing to check Monday is not Hana Micron, Samsung E-M, or Jeju Semiconductor. It is the US 10-year, Brent, USD/KRW, and the VIX. Until at least 2–3 of those stabilize, chasing even a great stock is inefficient.

Today’s RS screen is not a “buy signal” — it is a “watchlist-generation signal.” The next leadership stocks are not discovered AFTER the macro recovers. They are the names that did NOT fall while the macro was bad, recorded, and bought once the macro recovers.

Cycle before the stock. Macro gate before stock work.

This article is research and commentary only and is not investment advice. May 15 KOSPI / KOSDAQ moves are per the exchange’s official close. Stock-level returns and turnover figures are per Naver Stock and Korea Economic Daily market data. Close-location uses (close − low) / (high − low) and does not guarantee a perfect read on demand strength. US 10-year, Brent, USD/KRW, and VIX figures reflect May 14–15 market data. Gate thresholds (4.45%, 105 USD, 1,480 KRW, 18) are the author’s judgment and are not absolute. Company financials and target prices are per broker reports and DART filings and may change. Scenarios and entry conditions are the author’s hypotheses. The analysis may be wrong. Data cut-off: May 15, 2026 KST.

Disclaimer: For research and information purposes only. Not investment advice. Names cited are for analytical illustration; readers should perform their own due diligence and consult licensed advisors before any investment decision.

Relative Strength on Korea Invest Insights