AAA Games on Korea Invest Insights

Crimson Desert 5M: The Sell-Side Consensus Gap

Sat, 18 Apr 2026 09:00:00 +0900

This post is part of the Pearl Abyss Crimson Desert Thesis series. For the franchise IP re-rating thesis, see the earlier post Pearl Abyss: Crimson Desert Conquers China — 86% Steam Rating & 4M Copies Sold. That post examined the China market breakthrough and production-level success confirmation. This post focuses on a different question: why does sell-side consensus still lag, and what does that gap mean analytically?

The Central Question: 5M Copies Confirmed — So Why Is the Stock Still Slow?

Pearl Abyss (263750.KQ) officially confirmed the 5 million copies sold milestone for Crimson Desert on April 15, 2026. This is not a rumor, projection, or beta metric — it is a direct company disclosure. The question that production success raises is: “Is the game viable?” That question is now answered, definitively, in the affirmative.

Yet the stock’s price action since the milestone has been notably subdued. As of the April 17 close, shares traded at 54,100 KRW, with volume at approximately 1,081,588 shares — roughly 40% of the trailing average of 2,691,280. Relative performance versus KOSPI on the same day was approximately -5 to -6 percentage points in a session where KOSPI gained over 2%.

One analytical framework for understanding this divergence is what can be called the sell-side consensus gap thesis: the lag is not caused by information asymmetry — the market knows about the 5M milestone — but by an interpretation delay driven by how sell-side models update and how institutional capital is positioned relative to quarterly earnings confirmation cycles.

This working thesis does not constitute a buy or sell recommendation. It is an examination of market mechanics around a specific information-processing dynamic.

The Consensus Snapshot: Where Sell-Side Currently Stands

The following table summarizes publicly available full-year 2026 unit sales estimates from Korean sell-side research houses, as of the date of this analysis. These are the figures that currently anchor institutional modeling.

Brokerage	2026 Full-Year Sales Estimate
SK Securities	4.95M copies
Mirae Asset Securities	5.00M copies
NH Investment Securities	5.26M copies
Samsung Securities	6.00M copies
DS Investment Securities	8.00M copies

Source: Individual brokerage research reports. Figures reflect estimates as of the time of publication and may have been updated subsequently.

The structural pattern is clear. Four of the five tracked brokerages have full-year estimates at or below 6 million copies. Three of those four sit at 5.26 million or below — meaning their full-year estimates are functionally already achieved or exceeded as of April 15, with approximately eight and a half months remaining in the calendar year.

Only DS Investment Securities carries an 8 million estimate, which better reflects a post-5M normalization trajectory. The consensus, by any weighted reading, remains anchored to a pre-milestone world.

This is the gap. Not a gap in disclosed information, but a gap between the fact on the ground and the model universe most sell-side analysts are still operating inside.

Why Sell-Side Models Lag: Three Structural Mechanisms

Understanding why this gap exists requires understanding how sell-side research actually operates — not as a continuous real-time feed, but as a periodic, model-driven system with specific update triggers.

Mechanism 1: The Quarterly Earnings Reset

The most important single mechanism is the earnings confirmation cycle. Institutional capital — particularly long-only funds and active managers with quarterly attribution accountability — does not re-rate stocks primarily on press releases or milestone announcements. It re-rates on audited, line-item-level P&L confirmation.

Knowing that 5M copies were sold is useful. But knowing how those 5M copies translate into Q1 2026 revenue recognition, gross margin, and operating income requires a published income statement. Until that income statement exists, the institutional base tends to hold rather than expand.

Q1 2026 earnings are scheduled for release on May 12, 2026. Under this lens, that date — not the April 15 sales announcement — is the true forced-recalibration point for sell-side models.

Mechanism 2: Model Inertia and Rebuild Cost

Updating a sell-side financial model is not trivial. Changing the top-line unit sales assumption requires propagating the change through ASP assumptions, platform mix (Steam vs. console vs. regional), royalty cost structures, and quarterly phasing. A Korean brokerage analyst running a Korean gaming company model has additional complexity: Crimson Desert is a packaged AAA title in a market where Korean analyst teams have historically modeled live-service mobile games, not one-time purchase PC/console titles with long-tail decay curves.

The practical result: even when a milestone announcement arrives, model updates tend to wait for the next regular report cycle, not the day of the announcement. This is not analyst negligence — it is the rational allocation of research bandwidth under time and validation constraints.

Mechanism 3: The Analytical Frame Has Shifted — But Not Everywhere

The most sophisticated reframe after the 5M milestone is this: the debate is no longer “will Crimson Desert succeed?” That is a resolved question. The active debate is now about deceleration rate and terminal unit count.

A stock price driven by uncertainty about binary success will respond dramatically to success confirmation. A stock price where the market has moved on to debating the slope of the decay curve will respond more slowly and more episodically. Part of what makes post-5M stock behavior feel “slow” is that the most information-rich participants are already pricing a different distribution of outcomes — one anchored to deceleration assumptions rather than launch-risk uncertainty.

Velocity Analysis: Reverse-Engineering the Sales Trajectory

One of the cleaner analytical tools available here is a simple velocity calculation from the two official public datapoints.

Date	Confirmed Copies Sold	Incremental	Days Elapsed	Daily Run Rate
April 1, 2026	4,000,000	—	—	—
April 15, 2026	5,000,000	1,000,000	14	~71,400 / day

The April 1 to April 15 interval produced approximately 71,400 units per day — the most recent official velocity observation. This is an important baseline because it captures the game’s trajectory after initial launch enthusiasm had partially normalized, incorporating any Qingming Festival tailwinds.

From this baseline, one can construct scenario-conditioned estimates for the remainder of the year. The critical insight: significant deceleration is already embedded in every scenario below Bull. The scenarios are not extrapolations of peak velocity, but modeled decelerations.

Scenario Framework: Three Normalization Paths

The following table presents three independent scenario analyses for Crimson Desert’s full-year 2026 unit trajectory. These are not Pearl Abyss guidance figures, not sell-side consensus estimates, and not predictions. They are analytical scenarios constructed under explicit assumptions about deceleration rate.

Scenario	Full-Year 2026 Units	Incremental Needed	Required Daily Avg (≈260 days)	Key Assumption
Bear	7.5M	2.5M	~9,600 / day	Heavy deceleration; limited content; no new platforms
Base	8.5M	3.5M	~13,500 / day	Moderate deceleration (~81% below April velocity); regular updates
Bull	10.0M	5.0M	~19,200 / day	Mild deceleration; content expansion; possible new region or platform

A critical observation: the Bear scenario requires just ~9,600 units per day — roughly 87% below the observed April 1–15 velocity of 71,400. Even the most pessimistic modeled path produces a full-year outcome (7.5M) well above the majority sell-side position (≤5.26M for three of five brokerages).

The Base scenario at 8.5M implies ~13,500 units per day — an 81% deceleration from the recent observed rate. This is a conservative, not heroic, assumption.

This is the structural reason the consensus gap exists: if even a Bear scenario produces outcomes above most current full-year estimates, the current consensus is almost certainly anchored to outdated assumptions.

What the Market Is (Still) Getting Wrong

Under this analytical framework, the primary mispricing is a framing lag: a meaningful portion of the sell-side is still running models calibrated to answer “will the game succeed?” rather than “how slowly will it decelerate?”

These are fundamentally different questions. The first is a binary/bimodal distribution problem. The second is a continuous parameter estimation problem. Markets price these very differently.

When the key question is binary, a success confirmation triggers sharp repricing. When the question has already pivoted to deceleration slope, the market needs additional data points — specifically, the May 12 Q1 results and subsequent sales cadence disclosures — before it can anchor its deceleration estimate with statistical confidence.

The genuine observable opportunity in the consensus gap is the delta between where sell-side models currently sit (5M–6M majority) and where they will be forced to move after Q1 earnings require a model rebuild. That delta is quantifiable. The timing of its resolution — May 12 — is visible on the calendar.

Near-Term Catalysts

Three near-term events are identifiable under this framework as price-relevant:

1. May 12, 2026 — Q1 2026 Earnings Release The single highest-impact near-term event. If Q1 revenue recognition materially exceeds embedded consensus assumptions — which the velocity analysis suggests is plausible — sell-side models will update, price targets will revise upward, and institutional capital waiting for hard data confirmation will have the basis to act. This is the structural analog to Post-Earnings Announcement Drift (PEAD): stocks that beat consensus tend to continue drifting positively for weeks after the announcement as institutional repositioning completes.

2. 6 Million Copy Milestone Announcement The timing and velocity of the 6M announcement serves as a real-time update to the deceleration slope estimate. An announcement arriving quickly (before end of April) would strengthen the Base/Bull narrative. A longer wait shifts the probability distribution toward Bear.

3. Content Update Cadence Pearl Abyss’s ability to moderate deceleration depends on the content calendar — major patches, DLC, and potential new platform or regional availability. Each confirmed content event is a quantifiable deceleration-moderator with direct model implications.

Bull and Bear Case

Bull Case: Q1 earnings on May 12 reveal revenue materially above consensus; sell-side models rebuild upward; 6M announcement arrives quickly, validating Base/Bull velocity assumptions; deceleration proves shallower than Bear scenario; DS’s 8M estimate proves conservative; multiple expansion accompanies upward earnings revision.

Bear Case: Q1 revenue recognition disappoints on accounting phasing or cost surprises; deceleration accelerates faster than Bear scenario assumptions; content update calendar stalls; 6M milestone takes longer than expected, confirming a steeper decay curve; broader KOSDAQ multiple compression limits re-rating scope.

Frequently Asked Questions

Q: Is this a buy or sell recommendation? No. This is analytical commentary on market information asymmetry — specifically, the mechanism by which sell-side consensus models lag confirmed factual developments. Nothing in this post constitutes investment advice. All investment decisions involve risks that each investor must assess independently.

Q: What does the 5M milestone actually confirm? The April 15, 2026 official disclosure confirms production-level commercial success for Crimson Desert on a global multi-platform basis. It validates that the title has crossed the threshold where its economic contribution to Pearl Abyss’s 2026 fiscal year is material and largely de-risked from a binary success/failure standpoint. It also confirms that the company’s first major AAA packaged title — a significant strategic pivot from its Black Desert Online live-service model — has achieved industry-relevant scale in its launch window.

Q: Why does sell-side lag buy-side in situations like this? Sell-side research operates on a periodic model update cycle anchored to earnings releases, not continuous news flow. Updating a financial model requires propagating a changed assumption through revenue, cost, and earnings estimates — a process requiring time, validation, and formal publication. Buy-side analysts, operating without publication obligations, can update internal models in real-time. The structural result: sell-side consensus reflects the world as it was before the last major data point, while sophisticated buy-side positioning may already reflect the updated view. This creates the observable “slow stock” phenomenon — the price has not been pulled to a new equilibrium because the consensus anchor has not yet moved.

Q: What is PEAD, and is this related? Post-Earnings Announcement Drift (PEAD) is a well-documented market anomaly in which stocks reporting earnings surprises — particularly positive surprises — continue to drift in the direction of the surprise for weeks to months after announcement. The leading explanation: institutional capital repositioning takes time, and sell-side model updates and target price revisions are episodic rather than instantaneous. Under this framework, if Pearl Abyss’s Q1 2026 earnings on May 12 materially beat sell-side embedded assumptions, the PEAD dynamic could create a prolonged drift as consensus catch-up extends across the following weeks.

Q: Are the 7.5M / 8.5M / 10M scenarios official Pearl Abyss guidance? No. These are independent analytical scenarios constructed for this post, based solely on publicly disclosed unit sales data and observed velocity calculations. They do not represent Pearl Abyss company guidance, brokerage price targets, or any official projection. Pearl Abyss has not issued explicit full-year unit guidance as of the date of this analysis. These scenarios are a structured framework for thinking about the range of plausible outcomes under different deceleration assumptions — not forecasts.

Conclusion: The Gap Is in the Interpretation, Not the Information

The central finding of this analytical framework is simple: Pearl Abyss’s stock is not slow because the market lacks information. It is slow because the market is transitioning its interpretive frame from a binary success/failure question to a continuous deceleration-rate question — and that transition requires confirmed quarterly earnings data, not just milestone announcements, to complete.

The consensus gap is quantifiable. The majority of sell-side full-year 2026 estimates sit at or below 5.26M copies. The confirmed fact as of April 15 is that 5M has already been reached. Even a conservatively pessimistic deceleration scenario (7.5M Bear) produces outcomes well above the current sell-side majority. The forced-recalibration event — Q1 earnings on May 12 — is visible on the calendar.

Alpha, in this type of situation, tends to emerge not from better information but from a more accurate mental model of where the consensus is anchored and where it will be forced to move.

Related reading in this series: Pearl Abyss: Crimson Desert Conquers China — 86% Steam Rating & 4M Copies Sold — franchise IP thesis and China market re-rating.

Disclaimer

This post is market interpretation commentary and does not constitute investment advice, a solicitation, or a recommendation to buy or sell any security. Pearl Abyss (263750.KQ) is referenced solely for analytical and informational purposes. Sell-side estimates cited in this post reflect figures from individual brokerage research reports as of their respective publication dates; these figures may have been updated or revised since publication. Past sales velocity is not indicative of future performance. All investment decisions involve risk, including the risk of total loss. Readers should conduct their own independent research and consult qualified financial advisors before making any investment decision.

Pearl Abyss After 5 Million Copies: Why 7.5M Is Now the Floor, Not the Ceiling

Wed, 15 Apr 2026 18:00:00 +0900

Pearl Abyss (263750.KQ): 5 Million Copies Sold — Why the Real Question Is No Longer “How Many?” but “How Long?”

Crimson Desert crossed the 5 million copies milestone on April 15, confirmed via Pearl Abyss’s official social channels. That’s 100 million additional copies in just 14 days since the company formally announced 4 million on April 1. The game continues to sell on PS5, Xbox Series X|S, Steam, Mac, and Epic Games Store — a truly multi-platform footprint built on Pearl Abyss’s proprietary BlackSpace Engine.

But if you’re still framing the Pearl Abyss thesis around “will it hit 7.5 million?”, you’re asking the wrong question. The math has moved on.

TL;DR

7.5 million copies is no longer the upside — it’s closer to the floor. With 5M confirmed as of April 15 and a recent run-rate of ~71K copies/day over the last official tracking window, reaching 7.5M requires only ~9,600/day for the remainder of the year. The gap between current velocity and required pace is enormous.
The real unlock is 9–10 million copies + a formal DLC/multiplayer roadmap. Free updates matter not for their direct revenue but for their ability to flatten the decay curve and extend the revenue tail into 2027.
BlackSpace Engine is proven. Dokabi is still an option. The engine has graduated from tech demo to commercial AAA infrastructure — but the bulk of any market-cap re-rating still comes from Crimson Desert’s post-2027 cash flows. Dokabi is option value at this stage, nothing more.

1. Fact Check & Structural Breakdown

Confidence level: High.

Sales milestones, the update roadmap, engine/R&D disclosures, and management commentary are well-documented. What remains unconfirmed: DLC pricing, scale, and attach rates; Dokabi’s launch window; and Crimson Desert’s 2027 decay rate.

Here’s the factual foundation:

Crimson Desert launched globally on March 20, 2026 across PS5, Xbox Series X|S, Steam, Mac, and Epic Games Store. Pearl Abyss led with its BlackSpace Engine’s high-fidelity graphics, dynamic combat, and environmental interactivity. The company officially announced 4 million copies sold on April 1, then confirmed 5 million via official social media on April 15. Between April and June, the company has committed to sequential free updates including boss re-challenges, reconquest systems, difficulty settings, new skills, new costumes, dedicated storage, new pets/mounts, and control/UI/draw-distance improvements — with the caveat that details may change as features are still in development.

Translated into capital flow, the structure is straightforward:

Base game sales → installed base expansion → free updates to sustain retention/reviews → DLC/multiplayer attach → lower discount rate on follow-up IP.

The two choke points: First, can BlackSpace Engine actually support large-scale open-world AAA in live production? Second, can the live-ops team absorb weekly user feedback fast enough to bend the decay curve? So far, both have passed their initial validation.

Breaking it down into P × Q × C: P = base game ASP by region plus future DLC pricing. Q = cumulative sales, monthly exit-rate, and DLC attach rate on the installed base. C = platform commissions, marketing, and post-launch live-ops costs. The current upside is driven more by Q than P — not “how much per copy” but “how long before the curve breaks.” This is the fundamental difference from a mobile/MMO valuation framework.

2. Multi-Dimensional Analysis

Thesis 1: 7.5 Million Is Now the Floor, Not the Ceiling

Classification: Idiosyncratic Alpha

As of April 15, we’re at 5 million. To reach 7.5M by year-end requires selling 2.5M more over ~260 days — roughly 9,600 copies per day. Meanwhile, the most recent officially trackable window (April 1 at 4M → April 15 at 5M) implies about 71,000 copies per day. That pace won’t hold indefinitely, but the gap between required and observed velocity is striking.

In our assessment, 7.5 million is no longer “how high can it go?” — it’s “what’s the minimum it can hold?”

Translating to share price: our internal FY26 earnings sensitivity model puts 7.8M copies as a base case at approximately KRW 74,500, with sensitivity of roughly KRW 10,000 per 1 million copies. So 7.5M alone doesn’t trigger a dramatic multiple re-rating — it justifies the low-to-mid KRW 70,000 range. The real upside begins at 9 million: that’s +1.2M above base, implying the high KRW 80,000s on sensitivity alone. At 10 million, we’re looking at KRW 100,000 territory.

The market’s persistent mispricing: the “launch spike → immediate decay” framework that Korean investors reflexively apply to packaged games. The math currently says the opposite. 7.5M is too low a bar. The debate should center on 900K–1M and whether cash flows survive into 2027.

Red Team: If the May–June exit-rate collapses and daily sales converge to ~10K, the “7.5M is the floor” thesis weakens significantly. The slope of deceleration in May–June matters more than the absolute sales total.

Thesis 2: Free Updates Are Not About Free Content — They’re About Decay-Rate Defense

Classification: Quality Compounder validation stage

The April–June update roadmap isn’t a simple QoL bundle. Boss re-challenges, reconquest systems, and difficulty settings restart the endgame loop. New skills, pets, mounts, dedicated storage, and UI/control improvements target session length and return rates. Pearl Abyss explicitly stated these updates reflect player feedback and will roll out sequentially from April through June.

The effects are already observable. Over a 48-hour tracking window following recent updates, we recorded: peak concurrent users up 35.9% (146K → 198K), trough concurrent up 72.5% (75.5K → 130.2K), US sales rank improving from #5 to #2, China from #8 to #4, and new review positivity at 94%. This is evidence — not proof, but evidence — that update cadence and feedback integration are meaningfully influencing short-term sell-through and retention.

From an investor’s perspective, the value of free updates is not direct revenue but incremental base-game sales. With sensitivity at ~KRW 10,000 per million copies, if the update program pushes an additional 300K–600K cumulative sales in 2026, that’s KRW 3,000–6,000 in share-price support. That’s the first-order effect.

The second-order effect is larger. When the market starts believing “this isn’t a one-and-done package — they’re building a live franchise through 2027,” the multiple itself changes. Our model leaves room for PER +2–3x upon formal DLC/multiplayer roadmap announcement.

Paid DLC direct cash flows aren’t negligible either. Under speculative assumptions — installed base of 9–12M, DLC attach rate of 15–30%, net unit price of KRW 20,000–30,000 — direct DLC revenue ranges from KRW 27 billion to KRW 108 billion. But we view the more important signal as franchise validation rather than the DLC revenue itself. Recent Reddit-based sentiment research also shows net-positive sentiment with strong demand for DLC/expansion content.

Red Team: The company itself noted these features are still in development and subject to change. Schedule delays, below-expectation quality, or weak attach rates would break the “free updates = franchise building” narrative.

Thesis 3: The Engine Is Proven — Dokabi Is Still an Option

Classification: Idiosyncratic Alpha

BlackSpace Engine has graduated beyond “pretty demo engine.” In its launch press release, Pearl Abyss highlighted the engine’s photorealistic graphics and environmental interactivity. At GDC 2025, the company presented seamless open-world loading, detailed physics/environment interaction, and dynamic combat as core technical achievements. The company invested KRW 61.2 billion in R&D (37.5% of revenue) in H1 2025 and KRW 132.9 billion in the prior year on engine development — specifically on ray tracing, rendering quality, physics implementation, and large-scale open-world optimization. This is no longer a tech demo. It’s commercial AAA production infrastructure.

But a common mistake follows: equating “proven engine” with “Unreal-like platform business” is overvaluation. The engine’s direct value is modest. A more conservative IP SOTP puts Dokabi at KRW 0.2 trillion and engine/other options at KRW 0.1 trillion as a base case. Combined, that’s still far smaller than the base game’s long-tail value. The upside is still Crimson Desert.

Dokabi expectations follow the same logic. At the March 27 AGM, management said they’re “preparing to showcase Dokabi quickly after Crimson Desert” and will “reveal development progress at the appropriate time.” However, a separate February conference call report estimated roughly two years from Crimson Desert’s launch to Dokabi’s release. The conclusion: a reveal may be close, but actual earnings contribution is distant.

The implication for valuation is clear: Dokabi is not the main thesis that replaces the base game — it’s option value worth several hundred billion won at most. A playable demo and rough timeline could compress the discount rate in the short term. But pricing in more than KRW 1 trillion of Dokabi value before proof is sentiment trading, not analysis. If the base game’s long-tail weakens, Dokabi cannot compensate. This is the market’s key mispricing variable.

3. Actionable Framework

Theme: Global AAA package → franchise conversion re-rating Ticker: Pearl Abyss / 263750 / KOSDAQ One-line thesis: Crimson Desert’s hit status is confirmed. The remaining upside is determined not by “how many more copies” but by “does this become a franchise that generates cash into 2027 and beyond?”

Verdict: Wait

The meaning is straightforward. Existing positions are holdable, but fresh capital should wait for 9M/10M path confirmation and roadmap clarity. As of April 15 at 15:30 KST, the stock trades at KRW 56,000.

Valuation framework operates on two layers:

Base game upside uses the FY26 earnings sensitivity model: 7.8M copies base = KRW 74,500; ~KRW 10,000 per additional 1M copies; PER +2–3x on DLC announcement.
Engine/Dokabi options use a conservative IP SOTP: Dokabi base KRW 0.2T / bull KRW 0.4T; engine and other options KRW 0.1–0.2T.

Scenario matrix:

Copies Sold	Price Target Range	What It Means
7.5–8.0M	Low-to-mid KRW 70,000s	Minimum floor validation
9M	High KRW 80,000s to mid-90,000s	Re-rating begins in earnest
10M + H2 DLC/multiplayer roadmap	KRW 100,000–110,000	Realistic range
11–12M + DLC attach + Dokabi playable reveal	KRW 115,000–130,000	Upper bound opens

Above KRW 100,000 requires sales volume + roadmap together — volume alone isn’t sufficient. Above KRW 120,000 additionally requires Dokabi reveal momentum.

Entry conditions:

April–June updates execute on schedule
Q1 earnings confirm Crimson Desert revenue recognition and commission structure in line with expectations
May–June data supports “900M+ path” rather than “7.5M ceiling”

Catalysts:

5M milestone official confirmation (already occurred)
Q1 earnings disclosure
April–June update execution
H2 DLC/multiplayer roadmap formalization
Dokabi development status reveal (secondary)

Invalidation triggers:

May–June exit-rate collapse turning 7.5M into a ceiling
Update schedule delays or below-expectation quality
H2 passes with no DLC/multiplayer roadmap and Dokabi still vague
No visible bridge to 2027 cash flows

Final note: The question from here is not “was Crimson Desert a hit?” — that’s settled. The question is whether this hit is a one-time event or the beginning of a franchise. 7.5M is too low a bar. The variables that truly move the stock are 9–10M copies + formal DLC/multiplayer roadmap. BlackSpace Engine and Dokabi are clear positives, but trying to explain KRW 100,000+ with those alone produces a weak argument. Above KRW 100,000: base game long-tail first, Dokabi second.

4. Evidence Classification (Appendix)

[Fact]

Crimson Desert launched globally on March 20, 2026
4M copies confirmed April 1; 5M confirmed April 15
April–June free updates announced: boss re-challenges, reconquest, difficulty settings, new skills/costumes/pets/mounts, storage, UI/control improvements
H1 2025 R&D: KRW 61.2B (37.5% of revenue); BlackSpace Engine focus on ray tracing, rendering, physics, open-world optimization
Management: Dokabi reveal “at appropriate time” after Crimson Desert; separate report estimates ~2 years to launch

[Inference]

7.5M is now a floor rather than an upside target
Free updates primarily defend long-tail decay rate rather than generating direct revenue
KRW 100,000+ requires sales volume AND roadmap formalization
Engine is commercially proven, but direct value is smaller than base game long-tail

[Speculation]

DLC revenue of KRW 27–108B possible at 9–12M installed base, 15–30% attach, KRW 20–30K net price
9M → high KRW 80Ks–mid 90Ks; 10M+roadmap → KRW 100–110K; 11–12M+Dokabi → KRW 115–130K
Dokabi is additive option value, not a replacement thesis

[Blocked]

Actual DLC pricing, scope, and distribution model
Platform-level sales mix and net ASP
2027 Crimson Desert decay rate
Dokabi confirmed business model and launch window

The next decision point is singular: do May–June data support the 9-million-copy path? If confirmed, the conversation shifts entirely from “how much downside?” to “how high is the ceiling?”

Disclaimer: This analysis is for informational purposes only and does not constitute investment advice. All investment decisions should be based on your own research and risk assessment.