Amazon Could Be a Surprise Winner as We Close 2026
Could the political fight over data centers make Amazon even more attractive? Here’s my take and strategy for this tech/retail icon.
You've reached your free article limit
You've read 0 of 1 free Pro articles.

The market is under pressure again on Monday as AI and technology sector in general weigh heavily. The Invesco QQQ Trust (QQQ) is down around 2%, Roundhill Magnificent Seven ETF (MAGS) is down more than 2% and the VanEck Semiconductor ETF (SMH) is down 4%. Small caps have a little relative strength with a loss of 1%, but the only areas of strength are defensive names in pharmaceuticals and retail. Breadth is running 34% positive with 350 new lows to 70 new highs.
It is unattractive action, and the dip buyers aren’t showing up. For weeks there have been a lot of retail buyers trying to catch a low in data-center names like Nebius Group (NBIS) and CoreWeave (CRWV), and they appear to be giving up.
We have earnings from Nvidia (NVDA) on Tuesday and the Anthropic IPO later in the week to shake things up. I believe those will help move us closer to some lows that will hold and set up a rally to end the year. The question is which stocks I will want to own. It will depend on chart development, but in the Magnificent Seven group, I believe that Amazon (AMZN) has the potential for a strong finish to 2026.
What the Earnings Reaction Told Me
Amazon reported third-quarter earnings on Oct. 29 and came in slightly short on both earnings and revenue. The stock dropped to around $229 the next morning, but by the close that day it had recovered most of the loss, and less than two weeks later it was trading above $270.
When a stock comes back that fast after a disappointing headline, it tells me that big buyers were waiting for a chance to get in and used the drop to do it. What they were buying is Amazon Web Services, the cloud business that rents computing power to other companies so they don’t have to build and run their own data centers.
AWS revenue grew 39% in the quarter, which is the fastest pace in nearly five years. For a business that already brings in more than $180 billion a year, that kind of acceleration is unusual. Management then did something it rarely does and gave a forecast two years out, saying AWS growth should top 40% in both 2027 and 2028. A forecast is only a prediction, but this one is backed by $578 billion in signed contracts that haven’t shown up in revenue yet, which is more than double the level of a year ago. Management spent most of the conference call on the cloud business rather than on retail.
Why AWS Is the Advantage
Everyone in AI needs the same thing right now, which is data-center capacity with power already attached to it. It doesn’t matter how good your AI model is if you have nowhere to run it. AWS is the largest cloud provider in the world, and according to The Wall Street Journal, Amazon has more data-center capacity under construction than anyone else.
Amazon pays for most of that construction with cash from its own businesses, and it designs some of its own chips. When it wants outside money, it can get it on good terms. This morning it agreed to sell a 75% interest in 13 newly built data centers in Indiana and Mississippi to Blackstone Digital Infrastructure Trust (BXDC) in a deal valued at about $5 billion, and it will lease all of them back for at least 12 years.
Amazon gets cash out of buildings it has already finished and can put it into the next ones, and it keeps full use of the capacity. With the 10-year Treasury yield above 5% and investors asking hard questions about how Nvidia finances its customers, I want to own the company that big investors are lining up to finance.
Amazon will put about $220 billion into capital projects this year, and that has pushed free cash flow negative. That is the main reason the stock sold off on the report, and if the demand doesn’t show up to justify it, the stock will suffer.
The Political Map
Data centers were the big losers in the midterm elections. Voters approved restrictions nearly everywhere they were on the ballot, and candidates who promised to pause new construction won the governor’s office in Texas, Ohio and Georgia. About 40% of the data-center capacity planned in the U.S. sits in states where the governor wants a pause.
The WSJ looked at which companies are most exposed, and all of the capacity OpenAI has planned is in those states. Less than half of Amazon’s planned capacity is there, and the pauses are aimed at new approvals rather than at data centers that are already running or being built. That is where Amazon is ahead of everyone.
When it gets harder to build something new, what already exists becomes more valuable. Amazon jumped 6% the day after the election while the chip stocks and the smaller data-center builders were sold. Investors figured out quickly that restrictions on new supply help the company that already has the most of it. A pause that reached projects already under construction would change my view, and I’ll be watching how the new governors write their rules.
The Chart and the Plan
Amazon is down about 2.6% this morning with the rest of the group, which puts it under $260 and about 9% below the closing high it hit in early August. It stalled just under $275 last week after the run off the post-earnings low, and a move through that level would likely bring in momentum buyers and set up a test of the August high.
Last week I pointed to support around $265 for a trade. That level gave way this morning with the rest of the group, so I’m now looking to the 50-day for my core position.
I own Amazon as a core position, and I’m looking to add to it on weakness. The 50-day simple moving average at $255 is the first level of support, and the low after earnings around $230 is the level that tells me I’m wrong.
Amazon owns a stake in Anthropic, and between that IPO and the Nvidia report, there is likely to be plenty of volatility in AI names this week. I’d rather buy Amazon on a pullback caused by somebody else’s news than chase it through resistance.
You can be right about AI in many ways and still lose money on the stock you picked. I’d rather own the company the others have to rent from.
At the time of publication, DePorre was long AMZN, NVDA.
