AI Stocks are Crashing. Top 5 Stocks To Buy Now.

It's a crazy day in the stock market with the entire AI trade selling off. You have countless AI hardware related companies selling off more than 10% including memory makers like SanDisk and Micron, chip companies like AMD, equipment companies like Lamb Research. You have Neoclouds like Nebius and Coreweave selling off and some of the giants like Nvidia down 8.49% and Broadcom down 8.79%. This is causing a broader sell-off on the overall indexes. The Nasdaq 100 has sold off 5.81% today. That's 8.86% over the past 5 days. And the NASDAQ 100 is now in extreme fear territory. But as Warren Buffett says, you should be greedy when others are fearful. And this market selloff could be a great opportunity if you're looking to buy into the AI hardware side of the trade. Anything going into AI data centers, this could look quite attractive longterm. The NASDAQ 100 is down 13.77% over the past month. And year to date, the index is only up 6.31% when the earnings of the companies which make up this index have rapidly exploded throughout the year. So today, I'm going to go through my watch list of stocks and I'm going to rank the top five companies that are selling off big in this AI sell-off, which I think are still good buys long term. And as you can see, we have a sea of red going on today. Now, first I was asked to address this from some of our community members. What is causing this AI sell-off? And I don't want to get into the politics of everything because it is somewhat political. It has to do with the recent election in the United States. As many of you likely already know, there was a Democratic sweep. They won a majority in the Senate, in the House, and they won a lot of top governorships, including in Texas, which was somewhat surprising for some people. So, what does this have to do with AI? Well, one of the big topics in this election was the idea of banning AI or banning AI data centers. And today you just had a Democrat who got elected who tweeted top of the Democratic agenda on day one needs to be a pause on advanced AI development, regulation of this outofc control industry and a ban on artificial super intelligence. Now this lady isn't that influential, but there definitely is a large portion of the party who agrees with this. You have Bernie Sanders who is very influential. He wants a national moratorium on AI data centers, a pause on advanced AI development, and a ban on super intelligence. So, this is what's causing the selloff. The Texas governor who just won, one of her policies was a pause on AI data centers in the state. However, an important detail to focus on is that Greg Abbott, who she was running against, also for the past 6 months, has had a pause on the buildout of AI data centers in the state. And in this recent election, you had many localities around the country ban in their region construction of new data centers. And really for the past year to year and a half, this has been an ongoing problem related to the data center buildout. And this election, it just put it on the center stage. So am I worried about a national data center moratorum? Am I concerned as an investor in the AI infrastructure trade? I'll give you a few brief reasons as to why I am not concerned. Now, yes, it's not ideal if your goal is to build as many data centers as fast as possible in as many places as possible, but there still are many areas that want to have data centers, and there are many projects actively under construction, and it's continually ramping up. And over the past 2 weeks, we just saw all of the major hyperscalers give their guidance for capex in the coming years. And what do you know? It's going up. The buildout is continuing. And yes, these elections just happened in all of these different areas. And data centers were a big issue. And it's clear they're not popular. But the reality is this has been known for more than a year at this point. And the projects being planned going forward are taking this into account and they are not targeting communities where they don't want them. And there are plenty of places to build AI data centers. It might not be in your community or in your state. And if the country gets crazy enough, it might not even be in this country, but people are going to build them. It is frankly too important. This is not going to stop. It is going to continue. And I'm just talking about on land. There is also the long-term strategy of putting AI data center satellites in space, which SpaceX is working on starting in the next year and a half pretty much. And after the election, Elon posted data centers in space looking pretty appealing right now. But before the election, that was already going to happen. Still going to happen. And AI data centers on land are still going to happen. The reality of the situation is that not even all Democrats support a national moratorum on building data centers. And even if they did and they passed it in the House and the Senate, it's not how politics works in this country. The president, President Trump right now, he views AI data centers and the AI buildout as a national security very important issue. He has said he will veto anything related to that. And in the US, in order to overcome a veto, you need 2/3 of the Senate to approve that. Zero chance that happens. So the risk of a national data center moratorum is 0%. This issue has been known for more than a year at this point. We know that data centers are incredibly unpopular and there are regional risks of where you want to build them. And during that time, we've seen an incredible ramp in capex spending at Amazon, Alphabet, Meta, Oracle, SpaceX. The top AI spenders have spent more than $767 billion in capex over the trailing 12 months. All of these companies, they just reported guidance and they said that 2027 will be higher capex than the trailing 12 months. This was literally in the past 2 weeks. Companies are not going to shift on a dime like this. All of these projects are in active development and long-term planning. Are Amazon, Alphabet Microsoft Meta Oracle SpaceX going to stop their plans? No, they're not. In Amazon's latest quarter, they said they had $220 billion of capex in 2026, mostly for Amazon Web Services, and they're expecting meaningfully more in 2027. They had $75.9 billion in Q3. Alphabet is hitting their 195 to $25 billion in 2026. They just said they're expecting it to significantly increase in 2027. Microsoft just spent more than $50 billion of capex in their latest quarter. They are the most conservative of all the hyperscalers and even them they're saying they're going to have $170 billion for the full year and they're not expecting this to go down. Meta ramping up capex significantly and if you know they just launched Muse surprise surprise agents are incredibly token hungry so they are going to need to build out massive data centers. Oracle's capex is continuing. SpaceX just announced the most absurd scaling and spend related to the capex of their AI data center buildout. not going to stop and they are going to be a hyperscaler. The AI buildout is not over because an election happened. This is a long-term thing that will be going on for years and years and years. And AI itself as a technology is a very real thing that is going to be here for decades and decades and decades to come. We are still in the first inning of AI development. It is still incredibly early in its diffusion throughout society. So with that said, what an opportunity the market is presenting for us today. When stocks sell off and are cheaper, that's a good thing. You're paying less for the same share of a company. You're buying in at a cheaper valuation. So, as long as it's a good business that compounds and becomes more valuable in the future, the intrinsic value of the company is going up. Then, as an investor, you will do well buying the stock at a cheaper price. So, today I'm going to share five companies that are selling off today and I think are still great long-term investments at the current valuation. And at a minimum, they look interesting. With that said, let's roll the intro and get into today's stock analysis video. My name is Zach. This is dividend dating and you should leave a like and subscribe to the channel if you enjoy the video. Throughout I will be using the best stock research tool which is available at dividenda.com link in the description and pin comment of the video if you want to follow along or you can scan the QR code in the bottom right corner of the screen. Stock number one I'm going to share today is Nvidia ticker symbol NVDA. This is the main AI semiconductor company and it sold off 8.74% today bringing the stock down 3.35% year to date. This is during a year in which its earnings per share went up 73%. And as I will explain today, the stock price has been extremely lagging the fundamental growth of the business, which is a crazy thing to say when that stock price is up 513% over the past 5 years. But it's slowed down a lot. It's down 6 and a.5% over the past year. So in those five years, earnings per share at Nvidia have gone up 1,697%. Free cash flow at the company has gone up 1,461%. Growing at an 84.57% compound annual growth rate. Nvidia is a free cash flow machine. They are raking in money. That is up 76.34% year-over-year. The company will be reporting earnings on November 17th. I'll be following it on this channel. So, make sure to leave a like and subscribe if you want to see that future video. But this growth not expected to slow down. Nvidia already gave guidance for next fiscal year based on the supply that they're increasing and the amount of chips they can sell. So, the market had to up their fiscal 2028 guidance to 70.1%. That's growth year-over-year. And why is this happening? Nvidia's entire business at this point is data center related. 92.5% of their revenue in the latest quarter. And you can see that's responsible for all of their recent growth. Their data center business over the past 5 years is up 2,92% growing at a 105% compound annual growth rate. And in the latest quarter they had $89 billion of data center revenue up 116% year-over-year. And this is not just topline revenue growth. This is bottomline profits. Nvidia has very high margins. 75% gross margins, 62% net margins. And they have always had a great free cash flow efficient business. That's because they have a capital light business model. Relative to their operating cash flow, they have very little capex. Although it started ramping up as they build their own AI data centers for training and also just testing and everything. But within the AI trade, the semiconductor companies like Nvidia, they have just been raking in profits. That's why Nvidia has a $4.41 trillion valuation and was the most valuable company in the world. Now second place again. So Nvidia is in a very profitable position of the AI infrastructure trade and they have the most profitable companies in the world all buying their product. Microsoft, Amazon, Oracle, Alphabet, Meta. All of these companies are building AI data centers. And the question you want to ask yourself today is Microsoft, Amazon, Oracle, Alphabet, Meta. You can throw in SpaceX, you can throw in Open AI, you can throw in Enthropic. Are any of these companies done? Is this going to reverse? In my opinion, no. I think it's fairly obvious and I'll give you a little hint hint as into the rest of this list where some of the names might be coming from. It's people in the supply chain of the AI data center buildout TSMC who manufactures the chips that Nvidia designs. That's why Nvidia has such a capital-like business model. They don't have to build the chip fabs themselves. TSMC takes on that investment and this allows Nvidia to generate tons of free cash flow. However, one of the big new costs in Nvidia's business is now memory. And it's due to the memory shortage and the massive price hikes over the past year. Companies like SKH Highex, Micron, Samsung, they design and manufacture memory. So, I'll give you a little hint hint. We're going to be talking about some of these other companies today that are selling off. Now, I know some of you look at the 5-year chart for Nvidia's stock price and you say, "Hasn't the stock already gone up a ton? It doesn't look cheap." Everything with investing is relative to the fundamentals of the business. Nvidia's earnings per share have gone up way more than its stock price has. Its free cash flow has gone up way more. And now with today's 8% sell-off, Nvidia is trading at a 19.68p ratio for fiscal year 2027. They already did Q1 and Q2 2027. They are about to report Q3 2027. So this is the current year. For next year, which is fiscal 2028, more like calendar year 2027. They are expecting 70% earnings per share growth again, which would give you a forward PE of 11.6 based on today's stock price. And if we look at the PE analyzer here on dividend data.com, we can look at it based on Ford or even trailing 12 months. So this is what they had over the past 12 months. Nvidia's PE trailing 12 months is 23.11. That's the cheapest it's been in the past 5 years. Today is the cheapest Nvidia stock has been in the past 5 years. In that time, Nvidia stock went up more than 500%. And it's now the cheapest it has ever been based on earnings. And the forward PE is in the fifth percentile. By the way, this is also true on the 10-year view. It's in the second percentile on the forward PE ratio. The median has been 37.6. And historically, Nvidia's traded at high multiple because it's not a commodity. They have a highly differentiated product. It's an incredibly high margin business. It's a capital-like business model with great free cash flow generation. We have this tool called the fair value graph on dividend.com and it takes the median multiple that a stock has traded at over a selected time period. So right now we're looking at earnings per share. So that 37.37P ratio I was talking about over the past 5 years. If you implied that Nvidia is trading at that same multiple today based on its current earnings, it would give an implied fair value share price of $347.19. that is 88.5% above today's price. This puts it in deep value category. And like I've been saying frankly all year, Nvidia stock based on its earnings looks quite cheap right now. If we switch it to trailing 12 months view with adjusted earnings per share, the median over the past 10 years has been 51.07. And you can see the fair value calculation continues to climb as the earnings per share have increased over time. And right now, Nvidia is trading 49% below implied fair value based on trailing 12 months earnings. And it looks cheap on everything, whether it's free cash flow, operating cash flow, it even looks cheap based on revenue. And the reality is, if Nvidia continues growing its earnings per share, as it's expected to in the coming years, it doesn't even need to get a multiple rerating back upwards for it to be attractive. If we take this same 19.7P ratio that it is right now and we just hit these earnings per share targets, which for next year, which we can't look too far ahead, you know, anything can change in the future, but for fiscal year 2028 looks incredibly likely. If Nvidia stock at the end of fiscal year 2028 trades at a 19.7p ratio, that would give it an implied share price of $311. That would be a 70% increase from today's price. And if it continues hitting these earnings per share targets in the years after based on fiscal year 2031's earnings per share target, it would give an implied share price of $523, which is 187% upside from here. 28.4% growth annualized. Now, as I say this, I want to let you know that I am biased. I have been building up a stake in Nvidia during 2026. I am buying more of this stock today, and you should always do your research before making any investing decisions. I just like to be transparent on this channel and show what I actually think and what I'm actually doing. And I'm including Nvidia number one on this list because I think in a risk adjusted sense, it is the least risky of the AI trade and the biggest sure bet on AI infrastructure. Stock number two that I'm going to share today is Broadcom, ticker symbol AVGO. This stock is also down 7.84% on the day. It's down 13% over the past 5 days. It's down 18.35% over the past month and it's down 14.86% 86% year-to date. Over the past year, Broadcom stock is down 17.55%. Similar to Nvidia, this has been a great performer over the past 5 years, 429%. It's a 39.57% compound annual growth rate. Earnings per share for Broadcom stock are up 248% over the past 5 years. However, this growth has accelerated significantly recently. It's up 199% over the past 3 years. In the latest quarter, earnings per share is up 94.45% year-over-year. Similar to Nvidia, this is a capital light business model. They get their chips manufactured by TSMC. Lay's quarterfree cash flow is 13.67 billion, up 94.55% year-over-year, and they just had a blowout quarter in the last quarter. Everything is pointing to this continuing. Now, the reason why Broadcom's growth hasn't been as significant compared to Nvidia over the past few years is because they were not as concentrated in the data center business previously or even AI semiconductors, but it's now becoming an incredibly significant part of the business. Semiconductor solutions in the latest quarter was 70.4% of their revenue. And within that segment, AI related semiconductors is becoming the largest business segment. If you don't know, Broadcom, they partner with all of these other companies to make their own custom AI chips. Their longest term partner has been Google with their Google TPUs. This is the second largest atscale AI data center related chip behind Nvidia. Broadcom partners with the entire industry basically to design their chips. Amazon with their tranium chips. OpenAI's new chip if you've heard about it. Jalapeno, they work with Broadcom on that. Microsoft has started designing their own chips. They work with Broadcom. I believe Meta is also working on their own chips. They work with Broadcom. And if we look at the semiconductor business for them in the latest quarter, $20.84 billion, up 127% year-over-year. And this is expected to continue growing. And really when I think about Broadcom, I'm mainly thinking about that Google TPU business, which if you've saw Google's latest quarter with their Google Cloud growth, a large part of that, they're starting to now sell Google Tpus to external customers. So Google within them, they have a little bit of that Nvidia business. Google rapidly increasing capex. They're trying to build as many TPUs as they possibly can. That's only going to help Broadcom. Broadcom stock is starting to look quite cheap. In fiscal year 2027, they have $19.32 of earnings per share expected. That's 65% growth from this year. That would be a 15.5 forward-looking P ratio. Based on trailing 12 months earnings per share for Broadcom, it's currently in the midpoint 42nd percentile of P ratio. But remember, trailing 12 months is backwards looking. And all of that rapid growth has been in the past few quarters. So, in my opinion, forward-looking is what you want to look at for these kind of companies that are growing rapidly, especially where the expectation is continued high growth. So, a 15.39 forward-looking P ratio. So, if you look at the fair value graph for Broadcom, it's trading right around fair value. If you look at it based on trailing 12 months earnings per share past 5 years, median multiple has been 31.2. This company used to trade a very cheap multiple. It's now growing way faster than it used to. Broadcom back in the day, it used to be the dividend growth fan, the big dividend growth king, and it used to trade super cheap. I was buying the stock then. I frankly sold too early because I didn't expect this kind of massive growth in their data center business. Broadcom's business growth has really only been the past 3 years where it's had this rapid increase. So, if we look at it, trailing 12 months, multiple has been 41.36. If you're growing fast, you deserve to trade at higher multiple. 25% below fair value, 403 implied share value. That's 35.1% above today's price. And if we look at based on forward earnings per share, that's where it gets even crazier because Broadcom is not stopping growing. It's going to accelerate. It's going to continue to get bigger, at least in the next year or two. It's trading well below fair value there. And here's analyst expectations for earnings per share growth for Broadcom. Again, it gets nutty if they hit these 2029 2030 targets. We're talking about six times earnings based on today's price. Broadcom's customers are some of the biggest in the world. Alphabet, Meta. Also, Enthropic has a large part of their tech stack built on Google TPUs thanks to their partnership with Google and Google Cloud. Enthropic is the leading AI company in the world at the moment. And Broadcom mainly gets their chips manufactured by TSMC. Stock number three I'm going to share today is Taiwan Semiconductor Manufacturing Company, ticker symbol TSM. This stock has sold off 8.72% today. That is down 16.34% over the past 5 days, 26.41% over the past month. However, it is up 11.86 86% year to date. This is arguably the least replaceable company in the entire world. They manufacture all of the high-end AI semiconductor chips as well as a broad base of lower-end chips as well. They make a bunch of the Apple chips that go in iPhones and other products. They are a diversified business, but AI is becoming their largest business segment, and this has been a great company in the long run. And as you can see, when we look at their earnings per share, they just hit an all-time high in quarterly earnings per share. Their EPS is up 325% over the past 5 years and the latest quarter is up 57.5% year-over-year. They are currently trading at a 24.15 trailing 12 months P ratio. And on a forward-looking basis, it's 21.73. So on the past 5 years, this is historically cheap for Broadcom. But if we shorten that to 3 years, it looks even cheaper. And this is relevant because over the past 3 years, that's really when this AI chip buildout has been rapidly increasing. And TSMC is one of the most important companies to be following if you're following AI and the AI data centers because they're viewed as one of the constraint mechanisms on the buildout because these are the companies actually manufacturing the chips and they haven't been rapidly expanding as much as companies like Nvidia and Broadcom 1. Nvidia is pestering TSMC to please build more. Please build more because right now Nvidia knows if they had more chips they could sell way more chips. They say this every single quarter. Now, if we look at TSMC, their operating cash flow has grown massively. It's up 45.63% year-over-year. 95.94 billion of operating cash flow. Now, this is interesting. This is kind of how supply chains work. TSMC, they make a bunch of money making chips for companies like Nvidia and Apple and Broadcom who make even more money based on what they do with those chips. However, TSMC has far more capital expenditures because they have to build out these chip fabs. So this is one of the mechanisms for why the buildout isn't even larger. It's because they don't want to potentially risk an overbuild and building too much capacity. So it's taken a lot for TSMC to be convinced to increase their capex 95% over the past 5 years. They've only increased it 95%. And even with this more capital-intensive business model, TSMC still generates tons of free cash flow, 45.73 billion over the trailing 12 months, up 69% year-over-year. TSMC makes chips for all of the biggest companies in the world. It's not just for Nvidia, Microsoft, Alphabet, Tesla, Nvidia, Intel even, which is crazy because Intel's a fab itself. Qualcomm, Broadcom, AMD, Analog Devices, Marbell Technology, smaller companies like Sabrias. Not going to include them in this video, but they have sold off a ton where it's pretty interesting. I got to do more of a deep dive into that company. So, when we sit here on a day like today, and TSMC stock is down 8.91% and it's sold off a bunch in the past week, in my opinion, this is a pretty attractive opportunity for this one. An investment in TSMC is just an investment in chip manufacturing in general. And if you're following AI, you know we're going to need a hell of a lot more of those in the coming years. And TSMC, they haven't ramped up their margins and trying to extort companies making tons of money like Nvidia as an example. They've just slightly raised margins. They have great margins already, 66.4% gross margins, 51.1% net margins. But that's not true of all semiconductor manufacturers. And speaking of that, we have stock number four on the list today, Micron, ticker symbol MU. This is one of the major companies that makes memory. As we know, there is a memory shortage going on right now. And Micron, a company which has been a commodity historically, big boom and bus cycles, they are taking advantage of the opportunity. The whole memory sector is Micron, SKH High, Samsung. They are charging out the ass right now because they know that Nvidia, one of the largest customers, is willing to pay more for it cuz they're making a ton of money. In the latest quarter, Micron now has 86.8% 8% gross margins, 80.7% operating margins, 69.5% net margins. Micron now has higher margins than Nvidia, and they are making an absolute killing as I will go into. And you might say, "Oh my gosh, isn't this stock overvalued? It's gone up 825% over the past 5 years." Well, if we look at the sell-off today, it's down 15% today. It's down 27.65% over the past 5 days. It's down 36.6% over the past month. wall still somehow being up 113% year-to- date. And believe it or not, that is trailing the actual earnings growth significantly. This year, Micron's earnings are up 599%. And that is because they are charging way more. So, they have had an explosion in their profitability. Look at this chart historically. It's up, it's down, it's up and down, it's up and down. In the latest quarter, they're making what is this? Eight times their previous high back in 2018. That's in one quarter. And this is now going down in the next year or two. There's still a severe memory shortage. Micron just made $ 32.86 billion of free cash flow in the latest quarter. That's up 45,000% year-over-year. And they don't even have as much of a capital-l like business model compared to Nvidia, AMD, Broadcom, the companies I brought up earlier, because they actually manufacture this stuff themselves. And that's why they're starting to ramp up capex because they have been told, "Guys, you need to make more." And if you listen to basically every single CEO of all of these companies, they are saying they are not building nearly enough because all these other people, they would love to see memory become a commodity again. They want it to be cheaper. They do not want a memory shortage. But Micron, SKH Highex, and all these guys, they're like, "Oh my god, we're making more money than ever." And it's going to keep building on that base from this recent quarter into next year. So what does that mean? In fiscal year 2027, the current expectation is that earnings per share for the company are going to be $17,561. Based on today's stock price after falling again, that gives it a 3.8 forward-looking P ratio. That is not something you typically see in the stock market. This is so cheap where it makes you even question what is even going on. And basically if you want to know like the reason why it's being priced like this right now is because people think the company is over earning and 2 3 4 years from now perhaps earnings per share drop again significantly. They have to lower margins again and this is just a brief period of overearning and in the long run probably 10 to 20 years that's definitely going to be right at some point. It's not ideal for the rest of the economy for there to be a memory shortage. But the people who are giving the bull case for companies like Micron, they are saying that there has been a secular change in the market because now AI data centers, companies like Nvidia, they are going to need this continued high demand for memory and that might very well be true and it definitely looks true for next year, the year after that. And just to go into more of what's going on here with Micron, this is not a company that is a AI innovator, a leader. They are making the exact same product they were years ago. Pretty much slight improvements. They're not leading this revolution by any means. They are taking advantage of the situation and charging more for the products they already made, DRAM most notably. And they are charging more across the board. The data center is becoming a much larger part of their business and cloud more generally. But these are all existing business segments that they are now just charging out the ass for mobile. This is why your iPhone's more expensive. This is why if you buy a laptop, it's more expensive. It's all because of RAM increases. And that's because this data center business much smaller a few years ago. It's much larger. So they need it and they make the most money using it. So they are willing to pay more than anyone else for it. So that's why despite Micron stock going up 825% over the past 5 years, it looks like it might actually be a crazy value right now. This 15.29% sell off, it's starting to look very attractive. Again, just to put some numbers around this, the whole company, the market cap, what it's being valued at based on today's share price, $762 billion. In the latest quarter, they made $32.86 billion of free cash flow. If they just do that over the span of a year and don't grow at all from that base, that would be $131 billion of free cash flow, that means the company is trading at $5.8 times free cash flow. It looks stupid cheap, but the risk is that if the memory shortage somehow gets solved and they can't charge as much as they are right now, then this business is going to be making way less. Potential risks about it. As someone who follows AI very closely, there would need to be large algorithmic advances so that inference training AI agents need way less memory. The smartest companies in the world, all of the AI labs who are building super intelligence, all of the big tech companies, they are all pretty focused on solving this memory issue because it's now becoming a pretty big expense for them. So there perhaps can be innovations all the way through the supply chain. Maybe Nvidia has some innovations where they have some new design in the next generation that uses less memory. Or Enthropic and OpenAI, they point their super intelligence the same way they're solving Navier Stokes and all these math problems and they're saying solve memory. These are all things that could happen. They might happen. And the crazy thing is there might be just so much demand that it could happen and Micron still might be fine. Who knows? That's a risk if you're investing in Micron. And then stock number five that I'm going to share today, the final stock. It's not an AI semiconductor company. They're not designing chips. They are not manufacturing chips. They are a company which has benefited from the AI data center buildout significantly and they are selling off likewise related to that trade. That is comfort systems USA ticker symbol fix. One of their big businesses is HVAC and they are heavily focused on supporting data center buildouts because that's where a lot of the money is. and they were lucky in that they had this large network of talent and they have been serving this market for years and then all of a sudden there is more the demand for it than ever. So in the latest quarter for them 76.2% of it is related to new construction and I'll dive in to explain all of that but I brought that up just so you understand what's going on here. Over the past 5 years the stock price for fix is up 890%. That's a 58.21% compound annual growth rate. Today it's down 11.54%. Over the past 5 days it's down 27.21%. Over the past month it's down 42.72%. And over the past 6 months it's down 50.12%. Year to date it's down 2.23%. Now this is another company that is just seeing insane growth. Latest quarter adjusted earnings per share were up 68.61% year-over-year. But over the past 5 years earnings per share is up 1,237%. That's a 74.46%. 46% compound annual growth rate for a company which historically has been pretty boring and a lot of this is due to this new construction revenue and that's supporting the AI data center buildout. It's up 630% over the past 5 years and they are now supporting companies with seemingly unlimited money building these massive projects. So a company that was you know 17 18% gross margins it's now up to 24.8% gross margins. Net margins were 5% now up to 13%. That does a lot for the bottom line combined with the revenue growth in the top line. Here you can see Fix's rapid growth in free cash flow generation, $2.44 billion over the trailing 12 months, up $1,446% over the past 5 years. And this is one of my favorite charts to look at with this company. The dividend is up significantly. They increase every single quarter. Now, this is a small yield, small payout ratio as well. The payout ratio is 4.32%. It's a 0.41% forward-looking dividend yield. So, this dividend is incredibly safe. It has an A dividend safety score on dividend.com. But this kind of dividend growth, it's kind of unheard of. Three-year compound annual growth rate of 58.74%. Every quarter, they're doing a double-digit hike. And again, using the fair value graph or dividend data.com, it's the cheapest it's been in the past 3 years. This is based on trailing 12 months adjusted earnings per share. Currently 40% below the implied fair value. It was about that cheap in March 10th, 2025. I believe that was around the first time I started covering this stock on the channel. And I make this mistake a lot of times. Sometimes I give you guys great research and then I don't even act on it at all myself. I didn't buy this stock back then. This was in the April tariff selloff. I made a video similar to this one where it's like top five stocks to buy in the April tariff selloff. This was one of them. I didn't buy it. I probably should have because up 179% since then, even with this recent sell-off, but that's the game. And again, you should always do your research before you make any investing decisions. Do not listen to me blindly. I'm just doing this for educational purposes. It also helps me think through my research, preparing for these videos, recording these videos. And if you want to use the stock research tool we have throughout, it's all available at dividenddata.com. We also have a community of investors on there. You can see my buys, what I'm buying, talk to other people investing with similar strategies. I was chatting with people before I was recording this video. They were giving some feedback about what to talk about. If you enjoyed, make sure to leave a like, comment, and subscribe to the channel.