NCI: Hedge Fund 13F Deep Dive

NCI Hedge Fund 13F Deep Dive

[00:00:00] Hello, good afternoon, and welcome to our next episode of Lumina Non-Consensus Investing Live.

‍

As always, this is our quarterly discussion where we review the 13F filings of top-performing hedge funds. We’re going to go through that today. We’ll also use the Lumida app to automatically download hedge fund filings and analyze them.

‍

A few things first.

Hedge funds update their portfolios 45 days after quarter-end, so this is not real-time positioning. That said, most hedge funds do not turn over their entire portfolios. They usually have thematic and fundamental ideas that last one to two years or more. Warren Buffett has owned Coca-Cola since 1986, and you can see position changes quarter by quarter. Apple shows that too. The point is that the signal still has validity.

‍

The second thing is that because this data comes 45 days after quarter-end, there may already have been shifts and changes. A lot has happened in the last 45 days around geopolitics and other factors, so that is something to consider.

‍

The third point is that February options expiration is tomorrow. Quarterly options expiration usually brings heightened downside volatility during this period of the month. So if you find good ideas, it is probably better to note them and wait for a good entry, which could come in a few weeks.

‍

Usually, energy does pretty well this time of year, while technology, consumer discretionary, and financials do not. You can see that playing out in the market today and through most of this week. So if you find a great idea, make a note of it. You do not have to act on everything. As we like to say, you make your money on the buy, so wait for a good entry.

‍

With that said, let’s get started. I’m going to shift gears and jump to the app here.

‍

[00:03:00] What we’re showing here on the homepage is the app. You can see the basic market indicators. There are different functions here. I’m not going to walk through all of them now. I’m going to get right to the hedge fund ideas and discuss those.

You can see Smart Money, which is where we’ll focus. There is also a live newsfeed. It works like a Bloomberg terminal in your pocket and covers macro developments, thematic trends, earnings, OpenAI, private credit, and more.

‍

For example, you can see that private credit stocks fell after Blue Owl halted fund redemptions, echoing what we saw with Blackstone and BREIT a few years ago.

There is always something happening. We also have a special feature called Trump Watch, because Trump has been moving markets frequently. It is useful to see what he is saying and doing. There is also a dedicated macro newsfeed, where you can see discussions on mortgage rates, crude oil, Fannie Mae, Freddie Mac, and more.

Let’s start with the hedge fund portfolios.

‍

What you want to do is scroll down to Explore Portfolios, or click the Ideas tab or Smart Money tab. There, you can see a couple dozen hedge funds that we’ve curated. Most of these funds are highly skilled, though not all of them. We have quant funds and specialist funds focused on areas like energy and financials. So you get both quantitative perspectives, which have historically outperformed, and deep fundamental ideas.

We also have value managers and contrarian thinkers here as well. There are quite a few compelling hedge funds to look at, including Druckenmiller.

Our team has already reviewed a number of 13Fs over the past few days, so we’re going to highlight a few ideas and key themes.

From time to time, I may tap my analyst team for help pulling research notes.

‍

[00:06:00] Let’s jump right into the quant funds.

Let’s start with Renaissance Technologies. This is the late Jim Simons’ quantitative hedge fund with an incredible track record.

A few things stand out. A few quarters ago, this fund was long what I call “animal spirits” stocks, names like Reddit and Palantir. They no longer are, and they’ve been selling those names. That is a big change.

‍

I can quickly see the AUM of this strategy, their top holdings, and what they’ve been buying. You can see that they like momentum, which is typical for quant funds. They’ve been adding to Micron. Again, this is as of year-end last year. Costco is another momentum name, especially recently, with the bid in staples, even though it is quite pricey. They also bought Tesla and Netflix.

‍

Not only did those positions work, but I would not be surprised if they have since sold Tesla and Netflix, since both are in downtrends right now.

You can also see what they’ve been selling. One common theme we’ve noticed is that Mag 7 names have generally been sold by hedge funds. You can see that here with Nvidia. We did the same, and I talked about this on Twitter before these 13Fs came out.

‍

Renaissance also trimmed Google. Google had a great run last year as it moved from being an unfavored AI name to a favored, consensus AI name. They also sold other high-beta names like AppLovin and Meta entirely. This is a theme we are seeing across the quant funds we are looking at.

‍

You also see them selling animal spirits names like Hims. The bars show their exposure over time. The most recent bar shows lower exposure versus prior quarters.

They were also long quantum stocks. They owned IonQ in Q3 and exited it completely. IonQ later raised a billion dollars.

‍

You also see reduced exposure to capital markets themes, like CME Group, as well as SaaS names such as Snowflake. Renaissance did a pretty good job avoiding some of the weakness in SaaS.

‍

Interestingly, they have also been selling late-cycle momentum names like FedEx and UPS. Those stocks had a strong run. FedEx, for example, went parabolic. That really reflects the broader market theme: the Dow Jones has been outrunning the S&P, which has been outrunning QQQ, which has been outrunning Bitcoin.

‍

It is the old-economy, value-oriented names that have been doing well: logistics companies, industrials, energy. These are themes that had been out of favor and overlooked.

They also recently exited FedEx and UPS. Renaissance is very skillful. They know how to get into momentum and how to exit it at the right time. They were also in Walmart, which had an incredible run. Walmart is now trading at a 45x P/E. They de-risked Walmart at the end of December, around here on the chart. That was pretty good timing, even though the trend continued.

‍

They also held GameStop, another animal spirits name. That is fascinating, because it shows that quant funds have started to appreciate the power of social media and incorporate it into their investing.

‍

But the broader theme remains the same: quant funds have been exiting animal spirits names more generally and also reducing exposure to Mag 7 names.

‍

[00:11:00] Now let’s shift gears to Druckenmiller, specifically Duquesne Family Office.

There are some legendary investors out there. Buffett is obviously one. Druckenmiller is another. Point72 would be another.

‍

We are not going to focus on Buffett today. There were not many great insights in his latest filing. He bought New York Times, which looks expensive, and sold some Amazon. I think Buffett has been making a lot of mistakes lately. More importantly, Druckenmiller is a better investor than Buffett, with stronger performance. I would also argue that Point72 is better than Druckenmiller.

‍

So we’re going to spend more time on Point72 and the quants. But we would be remiss if we did not talk about what Druckenmiller is doing.

Druckenmiller added to financials through XLF. Again, this is as of year-end last year. Financials have not done that well recently, especially with discussion around a 10% cap on credit card rates. He also bought Brazil through EWZ, which we own as well, and added equal-weight exposure through RSP. This ties back to the theme of being underweight the Mag 7. We saw that with Renaissance, and we see it again here.

He continues to own Google and has actually added to both Google and Amazon, but his large position in equal weight is notable.

‍

What did he sell? He sold Teva, which is interesting. In one of my first quarterly 13F reviews, we discussed Teva based on Druckenmiller’s filing. It has since had an incredible run. If you had bought it when we flagged it, you would have done very well. Now he is selling, which in my view is exactly the right time. You can see that massive reversal wick on the chart, which suggests peak euphoria. On the monthly candles, it also looks overbought.

This is a great feature of the app: I can move from daily candles to monthly candles. The larger the timeframe, the less noise. This is a strong sign of exhaustion.

‍

Druckenmiller loves to sell parabolic moves. Teva was in a parabola, so he sold it and moved on.

He also trimmed a couple of semiconductor names, including Taiwan Semiconductor.

When you go to View All, you’ll notice that the AI has its own independent assessment. The three dots represent fundamentals, management/guidance, and stock/technical setup.

The first is the AI doing independent buy-side analyst work on the name. Green is good, red is bad.

‍

The second reflects transcript analysis: is management saying the right things about backlog, pricing power, market position, and guidance? If management is talking about price cuts and market share loss, that shows up as red. Of course, management can still misrepresent things, but this is designed to read tone and direction.

‍

The last is the charting AI, which tries to identify whether you should buy or stalk the stock. It is still in development, but it is solid. It has a good trend-following framework and has learned some of the basics of O’Neil-style analysis. It needs more work, but it is already better than what the average retail trader has access to.

‍

Sometimes there are no dots. That just means the AI has not yet been trained on that stock. This process is expensive, so we prioritize S&P 500 names and about 500 more from our broader universe.

‍

Druckenmiller has also been selling Meta. Again, that fits the broader theme of hedge funds trimming Mag 7 exposure. Meta is spending heavily on CapEx to support its AI ambitions.

He also sold GE Vernova, which has been an incredible industrials name, but is now very expensive. The app highlights that it is more than 2.5 standard deviations above normal, which is another way of saying it is overbought.

‍

When you click on a stock, the AI gives you a bull and bear case. It is important to have a point of view. In GE Vernova’s case, the bear case is that wind segment Q4 orders plunged 41% amid regulatory and execution setbacks, and the stock is trading at a very high multiple, around 56x earnings. To put that in perspective, Nvidia’s forward P/E is around 25x. This is extremely expensive and being driven by momentum. Eventually momentum breaks, and larger players tend to get out before that happens.

‍

You see that in Vistra Energy too. Druckenmiller got out before nuclear-related independent power producers corrected. Very good timing.

‍

Another important point is that he sold both CapEx payers, like Meta, and CapEx receivers. He trimmed both legs of that trade, which I think makes sense. We have been talking about that ourselves. On average, companies spending a lot on CapEx tend to lag because those are expenses. CapEx receivers, such as Nvidia, Taiwan Semiconductor, KLAC, and Lam Research, have done very well, but they are now priced for a lot of future expectations that may not be met.

‍

One other idea from Druckenmiller is airlines. Personally, I do not think it is a great idea at this point in time. Maybe it was a late-quarter idea, or maybe he has changed his view since then.

‍