Is NetEase (NTES) a Buy Before Earnings? The Q2 Profit Drag Has Already Reversed

TL;DR

• Yes, on the shares, and the reason is a timing quirk rather than a turnaround. NTES closed Friday at $125.12, up 2.01%, on roughly 16x trailing earnings with a 2.41% dividend yield.
• Nomura models RMB 9.4 billion of Q2 non-GAAP profit against a Street near RMB 10.2 billion, and attributes the gap to mark-to-market losses on NetEase's Alibaba and Pinduoduo holdings.
• Those marks were struck on June 30, when BABA closed at $95.98 and PDD at $76.28. On Friday they closed at $123.81 and $84.79, up 29.0% and 11.2% since the measurement date.
• The rebound does not reach the income statement until the Q3 report in November. Thursday's number is a photograph of the worst moment for two assets NetEase does not operate.
• The bear case has nothing to do with any of that. Games grew 6.9% against Tencent's 17%, and Ananta moved to 1H 2027, so the operating catalyst is more than a year out.

Is NetEase a Buy Before the August 20 Print?

Yes, though I want to be precise about what I am buying. This is not a call that Thursday's number beats. It is a call that Thursday's number is worse than the business, in a way that is knowable in advance and reverses on a schedule.

NetEase reports Q2 2026 before the US open on Thursday, August 20, with the release at or around 6:00pm Beijing time and the call at 8:00am ET, per the company's August 6 filing. At $125.12 the stock sits 21.6% below its 52-week high of $159.55 and about 18% above the $106.06 low.

The Two Stocks Inside the Profit Line

NetEase's non-GAAP definition strips out share-based compensation and very little else. An equity stake that falls in value lands in the headline profit number carrying the same weight as an operating miss, which is the mechanic that makes this quarter mispriced going in.

Nomura's July 27 preview models revenue of RMB 29.3 billion and non-GAAP net income of RMB 9.4 billion, roughly 8% below the Street, and puts the shortfall in the Alibaba and Pinduoduo positions rather than in games. The bank kept a Buy and a $155 twelve-month target on the ADR.

Here is the part I think the market has not done the arithmetic on. NetEase's June quarter closed on June 30, 2026. That is the date those holdings get marked. BABA closed that session at $95.98, near the bottom of a 52-week range that runs to $192.67. PDD closed at $76.28, against a 52-week low of $71.94. Both were marked at close to the worst prices either had seen in a year.

Since then: Alibaba closed Friday at $123.81 and Pinduoduo at $84.79. That is +29.0% and +11.2% off the mark dates, most of it on the Qwen-driven China AI re-rating that has run through the summer.

So the drag Nomura is modelling is real, and it is also stale. Anyone reading Thursday's profit line as information about NetEase is reading a June 30 snapshot of two other companies' share prices, both of which have since gone up a lot.

There is a wrinkle worth naming, because it changes the horizon rather than the direction. Alibaba reports its own quarter the same morning. NetEase's Q2 print will carry a stale mark on a company reporting live results a few hours later.

What I Could Not Size

I could not find a current, reliable disclosure of how many BABA and PDD shares NetEase holds. Without that, I cannot put a number on either the Q2 drag or the Q3 reversal, and I am not going to derive one from Nomura's residual and present it as a fact.

What that leaves is a sign without a magnitude. The direction of the Q3 mark is close to settled, since both stocks are up double digits from the June 30 close and the quarter is six weeks from over. The size is unknown. I think the sign alone is enough here, because the whole argument is that the August number understates and the November number overstates, and the shares are cheap in both.

The timing matters more than usual. Q3 runs to September 30 and gets reported in November. A reader buying on Thursday's dip is waiting roughly three months for the mechanism to show up in print, and collecting a 2.41% yield in the meantime while a $5.0 billion buyback runs underneath, of which about $2.1 billion had been executed as of March 2026.

Games Grew 6.9%. That Is the Real Argument Against.

I do not want the mark-to-market story to smuggle in a growth story that is not there.

Games and related value-added services grew 6.9% last quarter. Tencent's domestic games line reaccelerated to 17% when it reported on August 12. In Q1, NetEase's total revenue was RMB 30.6 billion, up about 6%, with games at RMB 25.7 billion, per the company's results release. One competitor accelerating while the other decelerates is the kind of divergence that usually means share is moving.

Margin has been doing the work instead. Games gross margin ran 74.8% against 70.2% a year earlier, which is how group gross profit grew faster than revenue. Margin expansion of that shape is finite. You cannot run it a third year without the revenue line joining in.

And the thing that would make the revenue line join in got pushed. JPMorgan moved Ananta from 2H 2026 to 1H 2027 in a June 29 note. Where Winds Meet and Marvel Rivals are both performing, and neither is new enough to reaccelerate a RMB 25 billion quarterly base on its own.

So the sober version: I am buying a company growing mid-single digits, at 16x, with a decelerating core, whose next real product catalyst is more than a year away. The multiple already says all of that. What the multiple does not yet say is that the August profit print is artificially depressed by an input that has already turned.

What Would Make Me Wrong

Three things, in the order I think they are likely.

Games decelerating again, below the 6.9%, would break the argument outright. At that point the cheap multiple is correct and the equity marks are a distraction from a business losing ground to Tencent.

A second leg down in Chinese equities would reverse the reversal. The Q3 mark is struck on September 30. BABA giving back its 29% before then puts the November print right back where the August one is.

The third is the one I would find hardest to argue with: this is a value call on a China ADR, and China ADRs have spent years being cheap for reasons that never resolve. A 16x multiple that stays 16x for another two years is not a loss, but it is not a thesis working either.

The sell-side average target compiled by stockanalysis.com sits at $161.72, about 29% above Friday's close, from a book with 31 buys and no sells. That is the sell-side's number on a twelve-month horizon, not mine, and a rating distribution with zero sells on a Chinese ADR is worth about as much as it sounds.

I have no live NTES option chain to work from with the market shut, so I am not logging a structure here. The preview already logged a pass on the straddle against a 3% realised-move threshold, and nothing I found this weekend changes that view. This one is a call on the shares.

Trade log

#	Stance	Structure	Strikes and expiry	Cost or credit	Spot at writing	Implied move	Conviction	Breakeven
1	Bullish	Long common stock	No expiry, 12-month horizon	$125.12	$125.12, Aug 14 close	not sourced	6/10	$125.12

The One-Line Read

NetEase's August profit line marks two other companies at their June lows, and both have since rallied hard. I am buying the quarter that looks worst on paper.