August Commentary: Preseason Hype Meets the Playing Season
Every sports fan has either uttered or heard the phrase, “the team looks great this year.” Star players and coaches have press conferences talking about how they feel good about the upcoming season, and that they are putting in the work to make the playoffs. But pre-season talk is a lot easier than delivering results in the games.
We have talked about several issues in the market, and there were some big themes that emerged early in 2026 – the pre-season hype – that are now playing out, and reality is setting. Let’s run through them, somewhat in order of importance.
- Artificial Intelligence Build Out: Huge pre-season hype, tough for execution to meet the expectations of the market.
- Market Earnings: Again, huge pre-season hype, but only a few players really deliver. Concerns arise that the whole season could crumble if we have a few main player injuries.
- The Federal Reserve New Fed Chair (Kevin Warsh): pre-season had mixed feelings, and the first two Fed meetings and press conferences have created more confusion than clarity.
So how has the market held up as the season plays on?
Market data as of July 31st (source: ycharts.com)
| As of 31-Jul-26 | July | Year to Date | 12 Month |
| S&P 500 | 0.2% | 10.1% | 21.5% |
| Nasdaq Composite | -2.5% | 9.5% | 23.6% |
| MSCI All Country World Index | 0.7% | 14.4% | 30.0% |
| Bloomberg Aggregate Bond Index | -1.1% | -0.7% | 1.9% |
| PHLX Semiconductor Index | -15.3% | 59.7% | 104.6% |
| GSCI Gold Index | 0.6% | -5.4% | 20.8% |
Source: ycharts.com
There is a lot to unpack in the table above, but the 12-month numbers reveal a bit of the pre-season excitement, the year-to-date numbers give you a little about how the season is going overall (pretty good, playoffs still in reach), and the July numbers tell you how the market feels about the last few games (not the best).
Warsh and the New Federal Reserve:
We have talked about Warsh in previous commentaries, and the story kind of goes like this:
- Kevin Warsh was a surprise pick by Trump. He has been known for being critical of the zero-interest rate policy after the financial crisis. Notably, Warsh’s hawkish stance on interest rates contradicts Trump’s 2025 push for rate cuts through Powell.
- Warsh’s first Fed meeting and press conference on June 17th showed a fiery Fed Chair who was dead bent on “delivering price stability.” He called five years of above target inflation “unacceptable,” removed forward guidance, announced five task forces and refused to submit his own dot plot. There was consensus on the Fed Board to keep interest rates unchanged with a 12-0 vote, but the market took the aggressive talk to mean that there were possible rate hikes in the future. He was dead bent on showing the market he would not be Trump’s sock puppet. This is like a coach calling for big change on the team so they can finally start winning games!
- Warsh’s second Fed meeting and press conference on July 29th was a very different show. Instead of big action talk similar to the first press conference, he used phrases like “watchful thinking, not watchful waiting.” Imagine the coach of your favorite team going from talking about “unacceptable” to then saying things like “watchful thinking.” Not exactly the great reform language from a month and a half ago. To add insult to the lack of conviction in Warsh’s tone, the vote to keep interest rates steady went from unanimous in June to 9-3. This departure of three Fed members voting to increase rates is a significant shift and also shows that some members are choosing to make a statement early in Warsh’s tenure that they are not going to sit quietly and follow his lead.
Source: https://www.nytimes.com/2026/07/31/business/federal-reserve-warsh-meetings.html
Overall, Wall Street and the bond market seem to be listening more to the inflation concerns than the philosophical ponderings in the July press conference. Wall Street believed the aggressive Warsh in June and pushed interest rates higher after the press conference and Wall Street analysts started predicting rate hikes in 2026. After the July press conference, rates went up again, choosing to look at the dissenting three Fed members as a sign of pressure from inside the Fed to raise rates. It’s not great when the coach sends mixed messages month to month, and some players start arguing separately for a different strategy. This is not a unified team!
The Earnings Scorecard: How the Draft Picks Are Playing
Since we are in the thick of Q2 earnings season, let’s stick with my analogy and offer a sports-style update, because I think the earnings data tells the real story better than any narrative.
Back in January, FactSet estimated that S&P 500 earnings would grow 14.9 percent for 2026. The Mag 7 were projected to grow 22.7 percent, and the other 493 companies were projected to grow 12.5 percent. Think of that as the draft board coming out of the pre-season. The Mag 7 were the consensus first-round picks. IT was the consensus first-round position group. Energy was the afterthought nobody was drafting in the first round.
Q1 was the opening kickoff, and the coach’s pre-season speech looked prophetic. The final blended growth rate came in at 28.6 percent, nearly double the projected 12.6 percent. Eighty-five percent of companies beat their estimates, the highest rate since Q2 2021. The Mag 7 grew earnings 63.2 percent with 100 percent of them beating expectations. IT led the sectors at 54.3 percent growth, Communication Services at 48.9 percent, Materials at 42.5 percent, Consumer Discretionary at 40.8 percent. Even the bench, the other 493, grew 17.4 percent. The first-round picks played like first-round picks.
Q2 is halftime, and the score is absurd, but there is an asterisk. As of the FactSet update on July 31, the blended growth rate is tracking at 47.4 percent with 86 percent of companies beating estimates and a surprise magnitude of 31.4 percent. That 31.4 percent is the highest FactSet has ever recorded. But here is the asterisk: strip out Alphabet and Amazon, and the growth rate falls to 28.8 percent. Alphabet’s quarter included a $98 billion gain, and Amazon’s included $53.4 billion from its Anthropic stake. Those are real gains on paper, but they are not operating revenue from AI compute. It is like a team winning on a pass interference penalty in the end zone on the last play. You take the win, but you do not confuse it with dominance.
All FactSet Data Source: https://insight.factset.com/sp-500-earnings-season-update-july-31-2026
The sector MVPs so far in Q2: Energy is leading the league, which nobody saw coming when the year started; Communication Services, Consumer Discretionary, IT, and Materials are the top five in earnings growth. Health Care is the one draft pick that is not panning out, the only sector in decline. Among the Mag 7, Alphabet and Amazon are the Q2 stars (asterisk and all), Nvidia played like a first-round pick in Q1, and Meta’s numbers included an $8 billion tax benefit that is the kind of thing that shows up in garbage time.
The Street has revised full-year estimates up to 29.1 percent from the original 14.9 percent, and Q3 is now projected at 27.4 percent with Q4 at 25.2 percent. So, the earnings story is not falling apart. But here is the connection to the AI section above: the Mag 7 were drafted first overall to carry the AI capex into revenue, and at the earnings line, they have; but the earnings are arriving with an asterisk. Mark-to-market gains, investment revaluations, tax benefits are inflating headline numbers without improving underlying cash generation. The cash flow story is different from the earnings story, and that is the gap the market is starting to worry about.
The AI Build-Out Hits a Pothole
For two years, the AI capital expenditure story has been the engine driving this market. I have written about it in nearly every commentary for the past 18 months, and the story has been remarkably consistent: the hyperscalers are spending hundreds of billions, the semiconductors are the picks and shovels, and the productivity gains are coming. The earnings numbers back it up. Q1 grew at 28.6%. The Magnificent 7 grew earnings over 60%.
But something shifted in July, and I think it is worth paying attention to.
The shift is not a crash. It is more like the moment in a road trip when the road gets bumpy, and everyone in the car stops chatting and starts looking at the GPS. The confidence has not disappeared, but it has been replaced by something more cautious. Let me try to simplify the three big reasons why.
Reason 1: The Politics of Power (and I do not mean the political kind)
The easiest part of the AI build-out to understand was always the digital part. You buy land, you pour concrete, you stack servers. The hard part, which we flagged back in January, is the physical infrastructure. Data centers can get built quickly; grid connections cannot.
In July, New York became the first state to pass a statewide moratorium on new data center permits, citing grid capacity concerns. Twenty-seven other states have some form of data center legislation in the works. The build cycle is hitting a major speed bump in the form of social push back and real-world constraints.
Here is the analogy: imagine you built a beautiful highway, but you forgot to build the on ramps. The highway is impressive, but no one can access it. That is where we are with AI infrastructure. The digital infrastructure was the easy part. The physical infrastructure – the power, the water, the grid connections – is where the friction lives.
Reason 2: Cancelled Orders and Supply Constraints
This one is messier because the signals are genuinely mixed. On one hand, the supply constraints are real. TSMC’s advanced packaging capacity is fully allocated through mid-2027. GPU lead times are running 36 to 52 weeks. On the other hand, there are reports that 30 to 50 percent of planned 2026 data center projects are being delayed or cancelled.
The sharpest signal came from Meta, which reportedly shopped around $10 billion of compute capacity to Anthropic in July. Meta is spending roughly $135 billion on infrastructure this year. The fact that they are quietly trying to lease out capacity suggests they may have built more than they currently need. And Microsoft, which had been the biggest spender of the group, guided capex down sequentially in its July earnings call. The market actually rewarded them for it. That is the tell. When the market rewards a company for spending less, the definition of good news has changed.
Reason 3: Too Much, Too Fast
This is the one that gives me the most pause, because it rhymes with history in a way that is uncomfortable.
In the 19th century, the railroad boom transformed the American economy. The railroads connected markets, opened new territories, and created enormous productivity gains. But a lot of the first investors went broke, because the build-out happened faster than the demand could absorb. Transformation was real, but the timing was wrong, and the capital structure could not survive the wait.
The AI build-out could be similar. Companies are expected to spend $7.6 trillion through 2031 according to Goldman Sachs, and that spending needs to produce real returns for the spenders. Source: https://www.goldmansachs.com/insights/articles/tracking-trillions-the-assumptions-shaping-scale-of-the-ai-build-out. With a significant decline in free cash flow from Google (first negative free cash flow quarter since going public in 2004) and a steep decline in Meta’s free cash flow last quarter, the market is getting antsy. We know where this technology is going, but the path is uncertain. And with uncertainty comes volatility. Investing in the future can be risky, and we need to be thoughtful on how to navigate it.
Conclusion: Talk is Cheap, We Came for Results!
Here is what ties all three of these stories together. The AI build-out, the Warsh Fed, and the earnings season are all stories about confidence running into the friction of real-world execution.
In the spring, AI was the pre-season speech. Boundless confidence, infinite capex, productivity just around the corner. Warsh in June was the same. Defiance, reform, price stability guaranteed. The earnings season started with the first-round picks playing like first-round picks.
By August, all three are in the “it is a long season” phase. The doubts in each case are not necessarily fatal. They are the doubts you get when high conviction meets messy reality. The AI build-out is hitting the equivalent of a slump. The Warsh Fed is learning the players on his team, the reality of their individual strengths, the squad’s dynamic and his coaching staff. The earnings are historic, but not unlike some of the most accomplishment athletes, some of the numbers have an asterisk.
August is not the month to make irreversible calls on thin information. It is the month to pay attention, keep your position, and make sure you are not the weak hands getting shaken out in the quiet summer traffic.
As always, if you want to talk through any of this, give us a call. WE LOVE THIS STUFF!
Chief Investment Officer
Green Ridge Wealth Planning
Disclosure:
Green Ridge Wealth Planning, LLC is a registered investment adviser. The information provided here is for general informational purposes only and should not be considered an individualized recommendation or personalized investment/tax advice. The investment/tax strategies mentioned here may not be suitable for everyone. Each investor needs to review an investment/tax strategy for his or her own particular situation before making any investment decision(s). You are responsible for consulting your own investment and/or tax advisor as to the consequences associated with any investment.
The opinions referenced are as of the date of publication and are subject to change due to changes in the market or economic conditions and may not necessarily come to pass. Any opinions, projections, or forward-looking statements expressed herein are solely those of the AUTHOR, may differ from the views or opinions expressed by other areas of Green Ridge Wealth Planning.