TMTB Weekly: A tactical bounce coming? What Tech Stocks to Buy

Mar 08, 2025

Another wild week in the books with QQQs -3.2% (dn 11% from highs at one point).

Here’s some of the carnage over the last 3 weeks in each sector:

I always enjoy sitting down to write these pieces on the weekend as I try to synthesize all the disparate pieces of information that have come out during the week. On weekdays, many times it feels like a firehose trying to just keep with everything that’s coming out. With the market closed and the mind rested, it’s nice to go through everything and see where the chips fall. I usually don’t know what I want to write or where the charts, datapoints, tweets, and recap of the fundamentals will take me; the fun part is just “getting out of the way” and seeing where the puzzle pieces fall.

Today, I didn’t think I’d end up feeling more positive — back to “choppy and upwards trending” and near-term tactically bullish — but after going through everything, but that’s how I feel and here’s why:

Let’s recap quickly. The market has sold off on a combination of a growth scare + DOGE fears + gov’t/policy uncertainty around tariffs + seasonality. It was the main reason why we shifted to a neutral “protect pnl” stance vs. our previous view of a “choppy but upwards trending market.” From our Feb 23rd weekly:

Top down the market is clearly saying to be more careful as the macro narrative has shifted - to us that means not being aggressive and protecting pnl/live to fight another day…the market is taking more seriously a growth scare given mixed data, gov’t shutdown (3/13) + tariffs (3/1) closer on the calendar and DOGE cut concerns; Trump’s shine as a biz friendly/confidence enhancing president has taken a bit of a hit; the narrative around retail bid disappearing through tax day has entered the market; we think China [TMTB: Add Europe to that] is actually a more neutral event for US Tech as could mean a shift in flows; and as we mentioned above: bottoms-up price action has gotten worse

While we were right in our “protect pnl” stance, we missed getting more negative especially as the growth/policy picture + investor psychology/risk appetite/de-grossing began to shift decidedly more negative. The unexpected violent moves in many tech stocks caught us off-guard and we didn’t move our feet fast enough - we were more focused on finding positive tea leaves instead of noticing the negative tea leaves in the macro narrative pile up.

Just take it straight from Treasury Secretary Scott Bessent on why the market will likely remain choppy:

Bloomberg

Treasury Secretary Scott Bessent warned that the US economy may see some disruption as the Trump administration shifts the basis for growth away from the government and toward the private sector.
“Could we be seeing this economy that we inherited starting to roll a bit? Sure,” Bessent said in an interview Friday on CNBC. “Look, there’s going to be a natural adjustment as we move away from public spending…Bessent argues that the expansion under the Biden administration was artificially supported by government spending.
“The market and the economy have just become hooked, and we’ve become addicted to this government spending,” Bessent said. “There’s going to be a detox period.”

While choppy remains the name of the game, we think stabilization and upwards trending are more likely going forward with and think there’s plenty of evidence to suggest a likely tactical near-term bounce is in the cards:

It’s funny how certain datapoints take hold of the narrative zeitgeist and others don’t. This time it was the Atlanta GDP Nowcast released on Feb 28th showing a huge drop that took the hold of investor’s attention, catalyzing growth fears and the next leg down in the market:

Well, yesterday Atlanta Fed was out in a Linked-in post saying the majority of the drop was driven by gold imports:

That is, as we now know from the March 6 full international trade report—but could only strongly suspect based on anecdotal and non-US government data until then—much of the widening of the trade deficit in January was due to an increase in nonmonetary gold imports from $13.2 billion in December to $32.6 billion in January. This accounted for nearly 60 percent of the widening of the goods trade deficit.
Although GDPNow does distinguish gold from other imports, the Bureau of Economic Analysis does, in tallying up the total of the net exports, subaggregate within GDP. Removing gold from imports and exports leads to an increase in both GDPNow’s topline growth forecast and the contribution of net exports to that forecast, of about 2 percentage points. The topline growth forecasts also increased today—standard model -2.4 percent to -1.6 percent, “gold adjusted” model -0.4 percent to 0.4 percent—as data from today’s labor market report came in stronger than the model was expecting based on the limited February data the model received prior to that release.

So ex-gold, growth is still positive at 0.4. And they are saying their next future update (they haven’t had an official release yet) will show an uptick to -1.6% from -2.4% currently. We think this release will be especially important only because it was important in grabbing the market’s attention earlier. I don’t make the rules: that’s the game we play.

Have we reached Peak DOGE fears?

Despite Musk and Trump telegraphing Federal cuts, the market was caught off guard by the speed and chainsaw-like approach Musk was taking to federal agencies, eroding confidence quickly. However, several developments over the last few days makes it seem like Musk might have to put his chainsaw down:

Have we reached peak Tariff Tantrum?

Bears had a chance to take this market lower on Friday and missed:

Despite all noise, the market is actually doing exactly what it has done over the last 20 years from a seasonal perspective and we are now entering a more seasonally friendly period.

Also aligns with post-election seasonality:

Market at one of the most extreme oversold signals in history:

We’ve seen extreme capitulaory flows. From BAML yesterday:

CTA supply will largely be finished next week:

Renmac: In Dec we warned of the beta extreme and it's vulnerability. Today, Beta is in the 1st percentile historically and narratives are developing around a recession. If beta is leading the narratives as sentiment surveys suggest, then investors are being too pessimistic.

We’re at extreme fear:

The dollar is down 7% from its highs. It wasn’t too long ago when fears of fx headwinds dominated the Tech narrative in January ahead of Q4 earnings. Will this be flipped 180 in April when investor focus shifts to Q1 earnings?

Oil below 70 and potentially going lower, tailwind to the consumer:

The 10-year is down 50bps from highs:

Putting the mosaic together: we think a near-term tactical bounce is in the cards and our base case has shifted from “choppy” to “choppy but upwards trending.” We got more long on Friday. The path forward remains uncertain: DOGE/Trump/washington policy (pot’l gov’t shutdown on 3/13) are wildcards, the employment picture is ok but tenuous, more de-grossing flows could overwhelm, macro releases will continue to have an outsized impact on the market, and we aren’t out of the woods on the growth scare yet and things take a turn for the worse. As always, never a sure thing in the market: we like the intraday low (-2% from here) on Friday as a risk mgmt spot where we would reduce our length and re-evaluate.

Ok enough macro, let’s go over some what we like on the long side in Tech:

Let’s have some fun and bucket the different longs:

For those that love risk/reward:

MU: Every week we like this stock more and more as we are getting increasing datapoints that the cycle is turning. Last week, Citi told us that DDR5 spot price has risen 4% mom, largest increase since June 2024, due a combo of lower supply growth and strong demand from DC and handset markets. (DDR is 70% of overall DRAM market). Yesterday we got news that SNDK is raising NAND pricing 10% on all products, which on the margin is good for MU as NAND is 15-20% of MU’s GP. In addition, we like MU as it’s more exposed to int’l consumer demand vs. just domestic demand, where the growth picture is more tenuous. Samsung has yet to qualify at NVDA for HBM and unlikely to do so until late 2025. So we have the cycle turning for conventional DRAM/NAND which we think adds lots of support to the downside and think by itself likely makes the stock work. However, if AI semi sentiment turns more positive, this can be absolute HR. After digging more into the impact of a shift to inference/reasoning models, we’ve become more positive on MU:

We like the equity. Any time the stock dips to $90, we love selling the Apr 80Ps and using that to fund the June 110Cs. We’d be very happy to get long the stock at $80 and we think AI semi sentiment has a good chance of potentially turning by June as it will encompass the NVDA May print, which will be hopefully be the big BW print. With conventional DRAM pricing going up, we struggle to see stock trading far below 2x BV and 8x P/E which it’s trading at now and think it should hold the $85 lows on the chart. Our upside case is 12 x $14 of EPS in ‘26 (Street <$11) which is close to $170. ~90% upside. If DRAM reverses or we get some new developments on HBM side, then would have to re-evaluate the thesis, but right now, we like it.

META:

META remains our favorite compounder and checks continue to point to upside for the q. Yipit is tracking at 15%+ vs street at 14%. Clev was out Thursday saying Q1 spending tracking ahead of partner expects on strong ROI, recovery in brand, and success with Adv+ although tailwinds from TikTok likely moderating. Clev models growth 15%+.

Risk/reward math: We think META can do $31+ in FY ‘26. Upside case is $31 x 28 = $850+, 35%+ upside. Base case = $31 x 25 = $775, ~25% upside…downside case —> 20-22x $29.50 = ~$590, 6% downside. Sounds pretty good. We like the equity and we also like selling the 590Ps and using them to fund the 660Cs.

NVDA

From Bernstein:

Got news this week Musk is ramping his plan to build out the 1M Memphis cluster. 1M NVDA GPUs at $30k a pop = $30B (they just did $32B DC revs this past q)…Need we say more?

AVGO:

50% share at 3 customers and conservative assumptions for other 4, prob gets you $11+ eps in '27 at low end. High end making some less conservative assumptions for new customer can get you close to $14+ (street still sub $10). So at $200, stock trading sub 20x low end ‘27 EPS - sounds pretty good to us for a co that will likely CAGR EPS 25-30% through CY ‘27. Upside is $14 x 25x = $350...Downside of $11 x 16x = $175, or 10% down from here. Base case somewhere in the middle —> $12.50 x 20 = $250, ~30% upside

PINS:

2% down is the earnings gap from when they beat and raised and we would be very surprised to see stock go below that barring a big miss/macro picture getting worse. Stock’s currently at 15x ‘26 street EPS of $2.15. Bull case is something like $2.3-$2.4 in EPS in FY26 and giving a 20x multiple to that is close to 35%+ upside. This week, got some good news. First Clev was out positive saying they think the co is working towards the launch of new 3p DSP partnerships. Although not launched yet, Clev’s work suggests “connectivity through SSPs could provide access to major US demand platforms. So a nice little catalyst if more 3p partnerships can launch. In addition, we thought CEO sounded really good at MS:

"We’re at record highs on MAUs, record highs on weekly active to monthly active… Even as we’re bringing record new users onto the platform, we’re seeing deepening engagement… Our most recent user cohorts are approximately twice as engaged as prior user cohorts… We’ve got our best product market fit ever… We grew 90% in clicks in Q4, even as we lapped 100% growth in clicks to advertisers the Q4 before that… The flywheel is really spinning on that."
66% of our weekly Gen Z users think of Pinterest as a first place to go shop… 85%-plus of our users come to us via our mobile app… That’s up 10 percentage points since I joined the business 10 quarters ago
We launched Performance+, our AI automation suite… It cut campaign creation time by 50%… We’ve seen 20% improvement in CPAs in beta testing… We just launched item-level bidding… ROAS bidding is coming out before the end of this quarter…
There’s still a lot more to go, both in terms of getting more share of wallet from the largest and getting to that $1 billion to $30 billion group."
We’re using AI to make significant improvements in our product that are leading to revenue now, leading to good profitable growth now… It’s the asset-light nature of our business."

ORCL:

We still think co can do $8 in EPS in CY26 and think 22-23x is the right multiple, which is at the very high end of where stock traded at before OCI took hold and would mean $175-$185. A more bullish 25x multiple is $200+ and possible but would require ORCL begin to show OCI upside/acceleration instead of just talking about it. Our dnside is $135 which is 21x CY25 street EPS of $6.60. So at $150, our base case $30 up and $15 down, which is 2:1. If stock got down to $145, then r/r begins to get more juicy at $10 down (<10% downside), $40 up (30% upside). That’s also good resistance from previous breakout back in Sept. We don’t have a great view of the q, but checks have been good. Here’s a few from Friday:

Clev positive saying OCI accel & Cerner better in Q3 while outlooks remain upbeat going forward helped by AI and DC multi-cloud.
Cowen’s channel checks indicate 1) expectations for modest SaaS growth acceleration (Fusion, Netsuite) in CY25, and 2) DB@hyperscaler initiatives beginning to generate stronger migration demand for ORCL databases to Cloud, Autonomous Database upgrades, and workload offloading to OCI (including stronger new customer acquisition). Additionally, Cowen’s hiring tracker shows positive sales hiring in Q3 following 6 declining quarters, suggesting ORCL positioning for strengthening demand.
Jefferies checks indicate 1) Partner practice growth is anticipated to increase modestly in 2025 vs 2024, supported by infra & app growth 2) 70% of partners achieved F3Q targets, with 50% exceeding plan. 3) 50% of partners reported q/q pipeline improvements during Q3 4) 35% of partners identified OCI capacity limitations as a challenge (last q only 65%)

CRM:

We’ve discussed this one at length - still think upside of $400+ is possible. Really have trouble seeing this one below previous trough multiples ahead of Agentforce cycle and with cRPO still growing 10%+:

MDB:

While we don’t love the story and think it likely makes sense to be patient and wait for a better set up later in the year (Ai likely to be immaterial this year and they are investing more for decelerating growth), risk/reward is v intg. From @Techfundamentals (honestly, i can’t believe MDB trading near 5x revs)

MDB has a history of guiding conservative with F24 initial guide at 16.4% —> actual = 31%….F25 initial guide at 13.8% —> actual = 19%. Mgmt had extra incentive to guide conservative given the ongoing CFO transition. So say with initial guide at 12-13%, we still get high teens growth. With Atlas 70%+ growth and growing mid 20s while non-Atlas declines, in a few years Atlast will be close to 85-90% of revs. 1)Ramping usage on solid new workloads 2) Waning impact from weaker Q1FY25 consumption headwinds 3) Renewed focus on deep-pocketed strategic account expansion deals both improving in the q” (Barclays) = a pathway for Atlas accelerating in Fy26.

For those that love a good catalyst path:

APP: Stock down 45% from highs and not added to SP500 on Friday. Our $400C flier we took last week is going to $0. That’s ok. Despite the stock being down 5% in the post-market on Friday, we wouldn’t be surprised to see stock end up in the green on Monday as we think expectations for an add had come down significantly. Mgmt is on the road over the next two weeks, they have an accelerated share buyback plan in place, and mgmt is bent on getting stock px up.

U: Recently announced the migration of the Unity Ad network to their new AI platform – Unity Vector (their version of APP Axon) and at MS further elaborated on the product’s ability to drive better customer outcomes in digital ads. This will get released at the end of Q1/beginning of Q2. One of our favorite high risk/high reward plays as numbers have been significantly de-risked and we really only need 5% of the outcome of APP to make this a decent stock doing forward.

NVDA: GTC on march 17th + bigger BW beat coming in MayFor those that love good 3p data:

For those that love good 3p data:

META: see above

RBLX: DAUs/Hours Engaged tracking 6-8 ppts above street while the Jan bookings #s at Yipit/M-sci were close to 40% vs street at 23%.

CVNA: Units/Revs tracking to high 40s vs street at 34%

Sidenote: AMZN 3p NA retail downticked Friday, now only tracking. 1ppt above street. While we like the r/r here at $200 and have confidence stock likely higher before end of the year as op margin thesis is intact, but the 3p dataset remains mixed (retail 1ppts above, int’l/AWS inline/below). A turn in the dataset would make us very excited, especially at this px as stock getting close to trough EBITDA multiple:

For those that love some beta and a good narrative/secular story: HOOD, RBLX, SHOP, NET, TEAM, APP, TWLO, NOW, SNOW, PLTR

For those that think best looking charts are the ones to buy: SNOW, SAP, BABA, Tencent (we still love our China longs)

“Has TMTB lost it?” ideas:

Quantum/IONQ/RGTI: Execs sharing stage with Jensen at GTC on March 20th. Could stocks rally into it?

TSLA: Peak DOGE/TSLA hate: Re-posting what I did above re: Musk’s influence:

We think Musk has done huge damage to his reputation and int’l sales look to be in freefall. However, Q1 deliveries miss already expected then we likely get some good news around lower-cost model/FSD in Q2. Stock -50% and just filled the election gap and feels like it’s at peak hate. A nice stop at the election gap fill/Friday low (-4%) looks like good risk for a nice potential pop/potential that was the ultimate low.