Lake Cornelia Capital 的 Judd Arnold 谈 $TOI 及其他诸多话题

Judd, how’s it going?

Great to be here. Happy New Year, and congratulations to you, new father.

Double father. Double father. Yep, double man-to-man defense still.

Oh my God, it’s a lot.

Well, we don’t have to talk about that. Judd, you had the glorious launch of Lake Cornelia, the Substack. The Twitter account was basically dormant for a year, and then you came out with the glorious launch of Lake Cornelia. You’ve been on fire in terms of 4,000-word memos. I mean, my God, people tell me, “Andrew, I don’t know how you do so much. You’re publishing like 3 4,000-word memos a week,” it feels like.

But we’ve got lots to talk about. I wanted to talk Vegas, but I’ll toss it over to you. You’re publishing on everything. Where do you want to start? What do you want to talk about?

Well, you can go wherever you want. I think the Substack is something I’ve been thinking about for a while. I think the transition, big picture, that I made was that I left hedge funds at the start of 2020 with no plan. During COVID, very rapidly, I had a few people reach out saying, “Will you work here?” I didn’t want to work for anybody ever again, so I offered to be a consultant. That was kind of my business for almost 5 years, which had a lot of positives, and I’m still doing that with a few clients.

One of the negatives that always bothered me—and this is something that bothered me throughout my whole hedge fund career, and something a lot of senior analysts talk through—is that it’s really hard to pitch a boss or a client something that’s obvious. You have to be unique. You have to be like, “I have this unique insight.” That’s always better. I think, generally, that’s been nice, but it also put me in a box. A mistake I’ve made historically is doing stuff that’s too cute by half, or just a little bit too fancy, too illiquid, and whatnot.

I just had this moment of, “I write a lot of stuff. People like what I write. What if I do a Substack where I just say, ‘This is interesting’?” I think the other aspect of this is interesting, and this is another thing for senior analysts that was illuminating. I used to have about 10 clients who would pay me for the consulting thing.

What I found—and it was so nuts, because I go back through my career and I worked at 3 of the biggest funds out there—is that, as a senior analyst, you pitch an idea to the PM. What’s the probability of the PM caring? Usually 10% to 20%. The position being put on? Like 5%, less than 5%. It’s really hard. When I had 10 clients, I could see this in real time. At least every idea I pitched, I’d have 1 or 2 people who shared it and would follow through.

But then I sort of had this second-hand epiphany, and this was something a friend of mine who I worked with at 2 funds brought up to me. This just sort of goes to the business of what we do, which is we don’t know which names are going to work. People would pay me for, “I just want to hear what you’re doing, even if it’s not what I’m doing.”

One of my friends said, “I went through all my numbers for the last 3 years. When I was high conviction, only 1 in 4 or 1 in 5 times would the thing actually work. So I think what I should do with my book is have these junior varsity positions, and then I’ll ramp up when it’s starting to play out.” That’s sort of high-conviction inflection investing, and that’s what it is.

All of this has sort of merged together. I’m really liking this Substack thing. I deeply appreciate the response. I’m super thrilled with how it’s impacting me and my investment process, which is that I feel less pressure to be unique, but also to do stuff that’s interesting.

Okay, fuck everything we were going to talk about before. You said a few things that really struck a chord with things I’ve been thinking about. Let’s start with the “too cute by half.”

One thing I’ve really been thinking about is “too cute by half” versus simple, right? And the simple, to me, is this: Historically, I really liked the financial engineering stories. We’ve got this thing, it’s going to be a 2% top-line grower, it trades for 10 times free cash flow, they’re going to buy back all the cash flow, and all the cash flow goes to share buybacks. Then there’s the famous levered buyback story: As they grow, they take on more debt, so that you get even more leverage.

I used to love that story. I don’t think any of them have worked over the past 10 years. I would think that’s an example of simple versus some of the stuff you and I have talked about or done. This is a SPAC that went insane, the private equity sponsor needs to sell all their shares, and the company has 3 of 4 divisions that are absolute garbage and are going to have to shut down. But this one good company, if they can just refi everything, is going to the moon.

That’s a really hairy story. I think 10 years ago I leaned toward the simple stories, and now I lean toward the more complex stories. But then I come back and think, damn, these more complex stories will rip your face off when you get them wrong. So I’d love to hear how you’re thinking about it.

I would add one more continuum to what you said, because the other thing that sort of looped into this—and they go hand in hand—is liquidity.

For every name that’s worked for me, it’s gotten really liquid. I found I kept too many names where I’m fighting the battle of, “Will it get liquid?” I’m buying it illiquid, hoping that I’ll get the mega payoff when it becomes liquid.

You do the inverse of that, and you’re like, “What if I go just to the next level up, where I trade valuation obviousness?” Because that’s really what I was doing. The less liquid names, the valuation was transparently cheap, whereas the more liquid comparable names, the valuation wasn’t as obviously cheap, but it was already liquid.

I’ll give a few examples of this. MiX Telematics, which became Powerfleet, which became AIOT, never really got liquid. It’s still obviously very cheap.

Well, I don't know if it's obviously very cheap because the stock's kind of gone sideways. But on traditional valuation metrics, you're like, “Okay, 7 or 8 times EVA—this thing could trade for 20 times,” blah, blah, blah.

It really crystallized for me with Nebius, or NBIS. You were the one, if I can hop in here, who told me about Nebius, along with a few other people. I can't think of a stock that I have traded more poorly. If I had just bought it when we talked about it—and I did—and never sold a share, I'd be a lot better off than I am talking to you.

With Nebius, you could push back on everything. We started talking about this at 18, and I really had a chance to lean in at 20. That was after the venture-capital funds came in. This was post-tariff.

The number one thing I would say to people when I was talking about it is, I'm 100% certain that people will care. It was trading like 1 to 2 million shares out of the gate. I was like, “This is going to be a QQQ—highly liquid, trading 25 million shares a day.”

It's just like, you have too many pieces with a management team that's known. You have all these pieces, and it became super liquid. So, if something is already liquid, or you're highly convicted that it will be liquid, and the growth story is somewhat tangible and not hard, that's important.

I put ASTS in this bucket too, which is one I'm not involved with, but it's one where I look back over the last 5 years. When they got that first contract and it ripped to 5, then came back to 4 two days later, and the warrants were at 1, I'm like, “People care. This thing can go to 50 to 100.” People would say, “How are you getting there?” I'd go, “I don't know. I just know it's going to 50 or 100, and it's liquid as all get-out.”

T1 Energy is another example. T1 Energy was formerly FREYR. I have 2 notes on it on my Substack. It's a solar name, but it's one where I came to it and thought, “I knew this at de-SPAC.” It was always a liquid one, and it's liquid still now. We have a story. This is worth my time exponentially more than the other stuff.

It sounds to me like, with liquidity, you're really talking about liquid things—things that trade. I like how you framed it: things that people care about, whether you believe the technology or not. I know people are on both sides of believing in the technology and the optionality there, but whether you believe it or not, it's space communications. People are going to care.

NBIS is a data-center cloud business for AI. People are going to care. TOI, which I've done some work on and we might talk about later, probably fits nicely. There are a lot of shares out there. They had a big private-equity sponsor give out shares to LPs, which is obviously a disaster. People probably care if it works, but less so.

I like healthcare services. What I'll say is, when you're right, people are always going to care. I feel convicted enough, and the care there, though, is you have to get more granular. It's a 20% grower in healthcare services that has a fixed capital structure, a multi-year, 5- to 10-year massive TAM, and a high-quality business. If they can execute, people will care. They're going to show up.

Let me ask the liquidity question a different way. When you said ASTS and NBIS, I think what I hear a little bit is, “Hey, I'm looking for the story before it inflects or as it inflects,” right? That's an interesting style of trading, but it doesn't have anything to do with valuation or fundamentals. Are you increasingly divorced from that, where you're just trying to find, “Hey, I want the story before it inflects”?

I know people who want the story before it inflects, and then it inflects and they sell. Maybe it's a zero long term, but they're out before it goes. It went from 40 to 200 to 0, and they're out at 150 or 160. Judd Arnold

One piece to that, which is critically important, is yes, I'm less tethered to traditional valuation metrics. And I'm more tethered toward story and liquidity.

But because I think my advantage is an ability to move quickly and to appreciate things that have a big—I think it is right tail. Is the good one the right tail, or is the left tail the good one?

Right tail is generally the good one. But, hey, the world's left tail is huge—right tail.

But, um, which was—I mean, I still remember you were buying 10-year-old VLCCs for $11 million. It was just like you were buying them at scrap.

But I thought, really getting on my own and leaving in 2020 and leaving hedge funds completely, you lose. This really goes to the nature of what I do. The more people I meet on this journey, it's mostly people who run money just for themselves. It's hard to get other people's money to do this because these are names that are hard to pitch, but you want to pitch them.

Nick and I just keep going back to NBIS. There's no person I pitch that to who says, “This is a terrible idea.” Every single person is like, “It's awesome. Help me with the sum-of-the-parts story,” because that's what I can lean on to pitch my boss.

Yeah.

You know, I can pitch downside. As you know, another one I'd throw into this bucket is WGS, which is this exome and genome company. It bottomed at around 2 and went to—I think it's at 140 right now. It's another one.

No, the way I've thought about this—and you can tell me if I'm wrong; this might be too married to fundamentals—is that you don't want a stock, as I said at the beginning, that's trading at 10 times price-to-earnings, right? Because the quants and the computers are all over that. That's fairly priced.

What you want is somewhere where the financials mean nothing. I'm sure you're at least somewhat familiar with Talen Energy, TLN. To simplify it for viewers, they own a big nuke out in PJM. What you want is a place where the financials mean nothing because all of a sudden the demand for megawatts is going up, up, up, up. It's hit that inflection. You'll never see it in the trailing financials, but the megawatts are like infinite money-printing machines. I don't know if that quite made sense, but that's how I think about it.

And I would go further on that topic. Talen is the spin-off from PPL. PPL was my biggest equity position at my first hedge fund. It made the start of my career. I've been to that plant. It's awesome.

When it came out, I looked at it and was like, “Power hasn't worked.” Power was my original—

You want to know why I didn't feel like it? Say I did distressed debt. I was involved in TXU and a few others. When it came out and I had all these generalists pitching it to me, I was like, “You guys have no effing clue how hard this is.” All of these go bankrupt every 7 years. Maybe you catch a cycle. I never thought you'd catch a cycle.

So that's where I was sort of going. You covered power for a while. How many years did we watch Vistra? “Oh, it's a 20% free-cash-flow yield,” and all this stuff. I literally moved away from power. It was where I started in investment banking. I've been to more power plants in this country than probably anybody on Wall Street. I bought power plants in the Calpine bankruptcy.

I don't know where CEG is trading today. CEG is the nuclear spin-off from Exelon. The last time I looked at that, it was 35 times earnings.

And the point I would make is that was after more than a decade of all these power names—NRG and all this stuff—trading at 15% to 20% free-cash-flow yields. The story being share buybacks, when you get the wave of generalist money coming in and the story changes, you're going to price at something stupid.

In terms of focus, I don't know if it's divorce from reality or valuation reality, but I deal in the liquid world. The first 2 funds I worked at were 15 to 20 billion AUM at the time, single-manager funds. You were really restricted on what you could buy. You can't play this game of, “I think people will care, buy it, have meaningful size, and then exit.” You have to actually be correct when you're at that AUM size.

When you're running 500 million or a billion dollars, as long as there's a little bit at the end, does it matter if you're not intellectually perfectly correct on a 20-year basis that the terminal value was X? Money flows in.

You want to know why I didn't feel like it? Say I did distressed debt. I was involved in TXU and a few others. When it came out and I had all these generalists pitching it to me, I was like, “You guys have no effing clue how hard this is.” All of these go bankrupt every 7 years. Maybe you catch a cycle. I never thought you'd catch a cycle.

So that's where I was sort of going, which is, you covered power for a while. How many years did we watch Vistra? “Oh, it's a 20% free-cash-flow yield,” and all this stuff. I literally moved away from power. It was where I started in investment banking. I've been to more power plants in this country than probably anybody on Wall Street. I bought power plants in the Calpine bankruptcy.

I don't know where CEG is trading today. CEG is the nuclear spin-off from Exelon. The last time I looked at that, it was 35 times earnings.

And the point I would make is that was after more than a decade of all these power names—NRG and all this stuff—trading at 15% to 20% free-cash-flow yields. The story being share buybacks, when you get the wave of generalist money coming in and the story changes, you're going to price at something stupid.

In terms of focus, I don't know if it's divorce from reality or valuation reality, but I deal in the liquid world. The first 2 funds I worked at were 15 to 20 billion AUM at the time, single-manager funds. You were really restricted on what you could buy. You can't play this game of, “I think people will care, buy it, have meaningful size, and then exit.” You have to actually be correct when you're at that AUM size.

When you're running 500 million or a billion dollars, as long as there's a little at the end, does it matter if you're not intellectually perfectly correct on a 20-year basis that the terminal value was X? Money flows in.

So give yourself credit to call what I'll call skill-based sector alpha—which is a factor. When I was at the big pod shop, there's single-company idiosyncratic and the decomp, for people who don't know. Most people get it: The average stock—and there's a wide range—typically gets 40% of its return from the market, 30% from the sector, and 30% from the company.

So if you can call the sector, that's almost worth basically the same thing as calling the individual stock. It's best when you can call both, right? You're looking for stories where the individual company and the sector are both trading in value. When you find that, does valuation matter in the whole checklist of the stock deck? Not really.

One question on what you just said: TOI—and I'm just using TOI; we can talk about it later—is small enough, and there is a sector, right? It's healthcare. There's going to be oncology, payments, Medicare, Medicaid, all this sort of stuff.

But it's small enough and unique enough. When I think healthcare, I think Pfizer. I think Tenet hospitals. When you're dealing with something that's small and a unique-ish business model, do you think sector matters as much? Does it matter just because of the tailwinds?

To me, I would say, hey, if Judd and I were looking at Tenet Healthcare, I'd probably agree with your 40/30/30, and we could split hairs over whether it's 30% for what. But when you look at TOI, I'd kind of be like, yeah, on a day-to-day basis, maybe it's probably 40% market for something that small and 20% sector.

Overall, I think the stock on something a little smaller, a little more illiquid, with a little more inflection—I know you were talking about liquidity—is actually going to be a lot more stock-specific. Do you think I’m wrong, or did I choose one of the things I like about TOI and healthcare services in general? This goes back to why people will always care about a healthcare-services name.

What is the sector for TOI? It’s literally a company unto itself. I’m doing the same process, which is: Do they have a story? I mean, this is where it’s like, is it the company or is it the sector? But certainly, let’s start with the market for a second. Economically, it’s completely insensitive in the short term.

Yes, there are market implications, but on a 2-year basis, 90% of the idiosyncratic move in the stock is the stock itself for something like TOI. That’s something you can size a lot more heavily than a commodity-based company or a consumer-retail, GDP-exposed, or cyclical thing. So that’s one of the reasons why I was able to say to myself, “Hey, I’m going more liquid, but this is really one to take that risk.”

But with TOI, where I really ramped the thing up because I was involved with it, I wrote a lot about it in 2023, and I left the scene in 2024. I round-tripped the thing from $0.35—it went up to $2.20, I got out at $0.65, and the thing ended up bottoming at $0.12 during tax-loss selling last November. It bounced off $4.80 a share a few months ago. It’s back today to about—I think—$4 or $4.20.

We put out a Substack note, and we think it can be somewhere between $15 and $30 in 2 years. It’s still trading under 1× revenue, which is just nuts. But to go from a reasonable position size to a full, massive ramp-up into the thing—and I own somewhere between 2% and 4% of the company—the story of oncology services really is this multiyear penetration.

Basically, with oncology services, they treat cancer and dispense cancer drugs. That’s the business. They buy the drugs from the drug distributor and give them to the cancer patients. You can go into the clinics and get infused chemotherapy, or they’ll give you an oral chemotherapy drug like Keytruda. They make 15% to 20% margins.

Their business model versus the other guys is that about 50% of patients go get the service through a hospital, and that’s the highest-cost thing. About 95% of the ecosystem is fee-for-service. The next level is all the big drug distributors. The big 3 drug distributors—McKesson, Cencora, and Cardinal Health—are starting to roll all these up.

All the private-equity firms are coming in. TPG owns one of the biggest ones, called OneOncology. Their business model—this is the drug distributors rolling these up—is saying, “The fee-for-service benchmark for drugs is ASP, average sales price, plus 6%. That’s what you get paid. Well, we’re the big 3 drug guys. We can procure at ASP minus 20%, and so we can make 20% margins on ASP plus 6% in a fee-for-service model.”

“We’re cheaper than a hospital because we’re going to do it all in an outpatient setting, so we’re going to be lower than the fee-for-service benchmark on patient services when you come in for your infusion chemo. Then, on the drug dispensing, we’re going to be really good because we can procure cheaper.”

What TOI does is one level further. They’re going to look at this menu of 10 drugs and look at what you’re doing. They’re like, “We’re going to service you.” This is sort of a weird way—it’s hard for me to describe this without sounding morbid. They’re doing what’s best for the patient, don’t get me wrong, but if you can get the same patient outcome for a $2,000-a-year therapy versus $20,000 a year, TOI is like, “We’re going to do the $2,000 one.”

The big 3 drug distributors want volume—maximum volume—as much as they can. They go to individual health plans, all the big Medicare Advantage companies, and say, “We will be 20% cheaper than fee-for-service. We’ll take this.”

The typical Medicare Advantage premium payment per month per member is somewhere between $1,000 and $1,500. The oncology piece that TOI touches is somewhere between $30 and $50 PMPM. TOI goes to an insurance company like Humana and says, “We will take that capitated risk on your whole network, all your members. We’re going to lock in $30 PMPMs, and we’re going to save you $20 on top of saving you $20 PMPMs. You’re not going to have to think about this. We’re just going to do the whole thing. We’re specialized in this.”

They’re really ramping up as they expand. The business used to be 95% California. California is sort of a weird island unto itself in terms of healthcare, because you have all these big physician groups that act like mini insurance companies. TOI has really expanded into Florida.

In California, the PMPMs—because the physician groups do most of the administration, delegation, credentialing, and stuff like that—are like $3 PMPM. In Florida, they’re earning $35 PMPM. There are a bunch of states like Florida. It’s a 20% margin business. You’re seeing product-market fit in terms of this new service. They’re the only people doing it.

They got a new CEO about 2.5 years ago, when the company was really struggling, and he really leaned in and pulled this out of nowhere. The company looked like it was going to go under.

But let me jump in. All right, I thought we were still talking about this stuff, but we’re talking about TOI now. I’ve looked at this a few times, and I guess my first thing is: as you said, they’re managing oncology for fee-for-service. Why can they do this so much cheaper?

I understand—nobody’s going to argue with me that fee-for-service has tons of fat to be cut, right? As you said, forget everything else: if you give someone a 6% margin, they’re incentivized to give you the $20,000 drug over the $2,000 drug, just because 6% on $20,000 is worth a heck of a lot more than 6% on $2,000. I totally get that.

But why are these the only guys who can do it? Why can’t someone else do it? Why can’t you and I start this up? Why can’t the drug distributors, as you said, do it? Why can’t the hospitals do it? Why are these the only guys who can do this?

Sure. There are 2 levels to this, right? Most of this care is coming through hospitals. This is a 2% to 3% margin business. You can go through AI. You can call a million GLG or AlphaSense experts, and no one will give me the actual number other than that it’s material, and a big part of hospital earnings is dispensing cancer drugs, oncology drugs, through their pharmacy.

At the highest level, CMS—the Centers for Medicare & Medicaid Services—cracks down on scam durable medical equipment. That’s where they’ve gone. Oxygen has been a big focus. A lot of people you and I know—I don’t know if you own it—the SANUWAVE.

I knew—I thought you were going to mention it. I was like, “Look, I’ve had a lot of people…” Wound care is a really scary area to me, and I’m not a—

—solution to the wound-care scam. But CMS is really going after this wound-care, skin-graft thing.

CMS isn’t going after oncology drugs because, one, it’s oncology—they don’t want to kill people. Two, the hospitals, behind the curtain, are telling CMS, “Hey, man, this is 25% to 30% of the total margin of the hospital, which is a 2% margin business, and that margin comes from dispensing oncology drugs. If you crack down on this, there are going to be real issues in the healthcare system.”

You start at that level. All the big 3 drug distributors were like, “Oh, you’re going to let hospitals charge for the highest-cost drugs and be really expensive for inpatient? Great. We’ll undercut them with outpatient oncology delivery at ASP plus 6%, and we’ll just be way cheaper.”

It doesn’t matter that we’re volume-based, just trying to pump it, because the fee-for-service benchmark is set by this insane hospital spend, which CMS is winking and nodding and letting happen. That’s a huge business, right? So, if you’re anybody but the big 3 or a hospital, you’re like, “What can I do?”

TPG, when they bought OneOncology 3 years ago, partnered with AmerisourceBergen, which at the time had not yet changed its name to Cencora. All the value is getting the drugs really cheap.

TOI doesn’t really have competition in the next layer, which is: What if we look at what the big 3 guys are doing and, instead of being really focused on volume, prescribe a cheaper cocktail of drugs that we know does the same thing? That’s a huge business.

Now, what’s the negative of it? It doesn’t scale as fast because you need to have a great oncologist in every area that you’re in, and you need to get oncologists to cooperate with you to stay on the menu, right? The big oncology groups struggle with this when people try to do it. If you don’t own the oncology clinics yourselves and you do it in an MSO model, these oncologists in your MSO just aren’t going to cut rates. They’re not going to deliver enough value.

This has been the problem for Evolent Health, which is a value-based provider, where they don’t really have any doctors.

Their MLR blew out massively. The stock has gone from $35 to $4 over 2 years because they don't have enough control over the network and clinician behavior.

For TOI, you need to have oncologists. You need to build out all these clinics, and that takes time. Each oncologist you hire takes 6 to 9 months. The aha moment for them was realizing that, in Florida, the market structure of oncologists and patients allowed them to get leverage over the oncologists and execute an MSO model.

So, let me start with where the MSO model doesn't work. TOI is in Nevada, mostly in Las Vegas. Las Vegas is just an island. There are 3 million people in that metropolitan statistical area, and there are maybe 50 oncologists that matter.

If you do an MSO network and you don't like what the oncologist is doing, you call them up and say, "I don't like how you're prescribing drugs to all these people. You're doing it way too expensively." Then an oncologist can say, "Well, I control the members. Where else are you going to go? There aren't that many oncologists here. I don't have to comply."

Whereas in Florida, you have a density of patients, oncologists, and hospitals. Florida, Texas, Ohio, and, I want to say, North Carolina are the markets where you have this convergence. Hospitals disproportionately set the fee-for-service benchmark, and you can get leverage over oncologists because there's enough density and enough people in every area.

So, TOI thinks they can service areas in Florida and 3 or 4 other states with only 20% owned clinics and 80% MSO. That's how you scale, because if you don't have an MSO network, the scaling is going to be so slow.

You're much deeper into this than me, but let me ask you a question. When I was researching it, the thing that popped up was that, again, I looked at this back in 2021 with the SPAC deck. In their SPAC deck, the peers they listed and the business model were not the same because it's oncology, but it had a lot of similarities: "Hey, the golden market is Florida. Hey, we're undercutting hospitals." The peers they listed were Cano, VillageMD, Oak Street, and One Medical.

All the primary-care guys.

They all went—I mean, I got beat up, but all of them either ended in tears for their investors or the investors managed to get the bag off to someone. Oak Street gets bought by CVS, and CVS takes a $6 billion write-down. With Cano, as you're saying this, I knew Barry Sternlicht took Cano public through his SPAC and said the CEO of Cano was the best entrepreneur.

I know. I got hurt on that one. Let's talk about what this thing isn't. There's no MLR risk. Everybody focuses on the MLR variability, but it's not like primary care, where you can have flu season or something like that. The biggest driver is drugs. It's drug costs.

They know that the benchmark is so high. They have this huge gap, and they're just going to prescribe different drugs. The second piece is costs. They've shifted from 4 to 1, oncologist versus nurse practitioner, and now the nurse practitioner-to-oncologist ratio is 1 to 1. They're saving on labor, but it's an iffy business. They screwed up.

Part of the other reason is that this was a recent pivot over the last 18 months, where they shifted to the sort of delegated model in Florida. What they did out of the SPAC was say, "Okay, we're going to land-grab across the country." They took SG&A from $40 million to $120 million in 2 years, built out all this capacity, and the patients didn't show up. They just lit money on fire.

Fast-forward to 2024: They started adding a lot of contracts. They're going to EBITDA break-even in Q4, and now you're going to start getting 20% incremental margins.

The gross margin on this business, I looked at, is around 15% or 17%. How do you get 20% incremental margin with 15% or 20% gross margins?

This is the other piece of how gnarly this thing is. On a capitated business, the margin is 15% to 20%. You also have a big fee-for-service piece in California. The fastest-growing piece of their business this year in Florida is fee-for-service.

Why? They built the clinic and got the oncologist. While they're waiting for these capitated contracts to come in, they're open for fee-for-service business. What stunned them, in a pleasant way, in Florida was the demand for outpatient care, because their service level is so much higher than a hospital's and other oncologists'. They're getting all this fee-for-service business in.

As that fee-for-service capacity transitions—or, I'd say, as the capacity transitions from predominantly fee-for-service in Florida to capitated over the next 2 years—as you start layering in all these contracts, the margin you're going to get on the dispensing side is going to go from ASP plus 6% to fully capitated.

On the patient services side, which is the clinic—chemo, IV treatments, and some hematology—that all goes from fee-for-service to fully capitated. So, you're going to win. You're going to get a margin.

I think that all makes sense, though. I'll be honest: You're deep into this. Let me ask the last question. It's the same question I would ask when people were pitching Molina Healthcare or all the health insurers. I do all the Medicare health insurance, and I worry that even if you're right, CMS is pretty rigid.

If they came and saw this, or the insurance companies said, "Hey, you're all on fixed costs and you're making a good profit. We don't care that you're cheaper than the hospitals. We're just taking your rates down, and you can either go pound sand or take it," I do worry about that ruling from on high, if that makes sense.

I sort of walked through why it's not like durable medical equipment or skin grafts, because it's a huge piece of hospital earnings. The big 3 drug-distributor rollups—CMS loves them. They're saying, "You're saving money. This is great. You're on the right side of this."

The most telling discussion I've had with the company came during a site visit. I got to meet the chief medical officer and their head oncologist in Florida. Their head Florida oncologist was actually the oncologist at ChenMed, which is private but is universally considered the best value-based primary-care provider out there. He switched over to TOI in 2022.

In this whole conversation with him, he was saying, "This is the way it's supposed to go." The most illuminating conversation I had with him was about Keytruda coming off patent in, I think, 2028. I asked them, "What do you think happens? Do you think it's just a one-for-one, immediate step-down as biosimilars and generics step in? Is it going to go down?"

They both immediately looked at me and said, "No way." I said, "So, you're literally—" They said, "Oncologists aren't going to move that quickly, and everyone's going to play the game because it's over 95% fee-for-service. Maybe over 4 or 5 years, it's going to converge to the lowest—not even the lowest-cost Keytruda, but a lower-quartile Keytruda generic. It's going to take years for that to play out."

I said, "So, we could literally be making 20% to 30% of the company's profits just on generic Keytruda versus the benchmark?" They said, "Yeah." I remember a long time ago there was a roll-up in the UK that was basically what you're saying: You buy old brands after they're off patent, and even 10 or 15 years later, there are still some doctors who just prescribe the brand. You've got a little bit of pricing power. It's not the greatest thing in the world, but when you pay a nice multiple, it's actually a royalty stream. That's really interesting.

Let me finish, though. For people listening, let me give everybody the valuation number quickly. This thing is trading at 7 times revenue. It's going to have zero EBITDA in Q4, and it's been negative EBITDA. I think it's going to do $20 million to $40 million of EBITDA next year, and I think you're building up to $75 million of EBITDA by 2027 exit rate.

There are about 130 million shares, and net debt is about $40 million. My target is that this can go to 2 to 3 times sales if you really believe in it, because the TAM is so big as they execute in more states. But this is going to take time as people buy in.

Street numbers are also insanely low. They've beaten and raised all year. I think the Street revenue number for 2026 is about $600 million and change—don't quote me on that—but you're going to get guidance coming up. Mathematically, I think you still have a lot of upside on the annual guide.

I'm going to start thinking about this when it gets closer to 2 times revenue, if I'm so lucky. I think there's a ton of valuation room if this thing starts working. The exit here is likely private equity. One of the private-equity-backed or drug-distributor-backed companies shows up and buys it.

It just worries me because this—I mean, look, this was the pitch for all the value-based guys, and a lot of them did get taken.

I know, but there’s really no MLR risk.

Evolent Health really scared a lot of people because that was the value-based name. It’s worth it for people to go through both. Evolent Health is not really that great of a business—high customer concentration and whatnot.

I’m going back to my notes from the start of the episode, when you were talking about launching Lake Cornelia. You mentioned pitching to a PM as a senior analyst versus pitching as a consultant, or maybe writing something up on Lake Cornelia. How have you seen people work with you differently as a consultant, or when you write something up, versus when you’ve been the senior analyst, head of research, or whatever it is at different places?

I think the biggest thing is when you’re a consultant. I’ve done well enough in my career that, if people don’t like the name, I’m just like, “Okay, great.” I’m not going to take it personally. This isn’t my job, so I don’t need to be upset about it. I have other clients I can pitch it to, and I’ll find another name.

When you work for somebody, how many good ideas are you realistically going to come up with?

That’s the tough thing, right? You’re an analyst, you’re convicted, and that’s why most analysts don’t last forever with the PM. You get one name, you do a lot of work, and you get convicted. Hopefully, the PM likes it, but if they don’t—if you’re a 10 out of 10 and they’re a “meh”—it’s really hard, no matter which way.

Exactly. You have this transmission problem.

Which is, even if they like it, they’re going to buy it when you don’t like it. They’re not going to trade it perfectly, and you wouldn’t size it that way. So you have this transmission problem of the things you would love to put in the book, or the way you would like to put them in, versus how that actually happens. Then your compensation is a function of the latter versus the former.

I think it’s something our buddy from One Main Capital talked about, too. He did a great interview about a month ago, and he made this point about going off on your own: investing is a deeply personal experience. It’s like art, and at some point you reach a point where you say, “I want to do it my way.”

That really resonated with me, because as I’ve gotten older, my willingness to engage with people when they don’t like something I like has decreased. I like having the veto: “You don’t like it? Great. I don’t owe you anything. I’ll move on to the next one.”

What you want to guard against—and one of the other reasons I love the Substack approach versus how I was doing this before, which was getting high conviction, pitching a memo, and then screaming about something on Twitter—is that you want to stay as intellectually open as possible. The more debates you have to vociferously defend yourself in, the more negative it is. It really impacts your ability to think clearly and cogently.

The big decision I made going back into 2025—my decision of the year—was that I had been very prolific and public about my feelings on Sable Offshore. We got back involved a couple of weeks ago, and we’ll see how that plays out from here. You could do 5 podcasts on that from here.

But when you’ve been very public about a name, it can start getting dicey. I sold the whole position in the $20s, and I need to keep the ability to do that.

The way the news flow on Sable works, you could do a podcast on them every day. Probably.

Let me ask you a different question. One other thing you mentioned at the start—and this is something I’ve been thinking a lot about—is having a small position and ramping it up as it inflects. The reason I’ve been thinking about this is that most of the big successes I’ve seen are very similar to what you think.

I’ll give you one. My friend Jeremy Raper, right? On Twitter—

When Twitter was suing Elon Musk to close the deal.

You know, I specifically remember there was something that came out in a court docket. Jeremy had been following it, and he had a small position. I don’t want to give away his trade, and I could be misremembering, but he came in and said, “This is the inflection. Elon is done. He cannot win this case.” And he pulled the trigger.

I had that one in mind because it was Jeremy and it was kind of public. The best guys I see do this. They’re waiting, they’re waiting, and then there’s the inflection and, boom, they hit it.

Now, I do worry on the other side. It’s very easy to see that and pull the trigger. But you know how it works: you’re following a name deeply, you see something, and you’re so convicted. There’s a lot of selection bias. The guys I’ve seen blow up have been, “Hey, I see this. It’s the inflection. Boom, boom, I pulled the trigger,” and they were wrong.

So I guess I want to ask you: how do you weigh, “Oh, this is the inflection,” versus, “Oh my gosh, am I just confirming my own biases?” It’s very easy: “I love this, I love this. All the news is positive. Oh, I was wrong, and it’s a zero.” How do you weigh that as you’re looking for the inflection?

Well, look, the biggest defense against all of this is keeping a liquid portfolio and retaining the ability, when you’re wrong, to hit the eject button as soon as possible. We’re not burdened by a $15–$20 billion fund.

I’ll go back to the first guy I worked for. There were 2 of them who ran the fund. One guy did everything but run the money, and one guy ran the money. People would say, “Oh, you see the returns,” and they’d be like, “Whoa, that’s not that great.” I’m like, “You try to run $20 billion at 14% net with 4% vol. Tell me how that works for you.”

Every dollar of AUM gets exponentially more difficult. For us, we’re hanging out in super-liquid things, and I think that’s important. I mentioned keeping myself as mentally open as possible with Twitter and Substack. I think that’s a big piece.

The second piece is that, at this stage, I kind of know—I’ve felt what it feels like to be right enough times. Holding on and saying, “Okay, I’m going to get conviction. It’s going to come,” is different. If I don’t have it, all right. If I do that and I’m wrong, we can exit because it’s liquid.

Can I push back on one thing you said? You’ve said a few times, “If I’m wrong, I can exit because it’s liquid.” I would guess that you’re a much better trader and changer of your mind than I am.

Intel is a company that I looked at deeply. I saw clear red flags, and a lot of people looked at it. A lot of them got out. They bought in at $20 or $30, it ran up to $200, and a lot of them got out there. A lot of them got out at $100.

But one day you woke up and, hey, the company was a fraud, the stock had been halted, and it was going to open up probably at zero. That’s not investing advice; I don’t know. I guess that’s an extreme example, but you keep saying, “If I’m wrong, it’s liquid. I can exit.”

I mean, sometimes you wake up to these things and you’re wrong, and it’s not liquid.

If you have something with downside jump risk in it—Sable is a situation where there’s actually downside jump-to-default risk, as I would say in a CDX context, or a risk-arb trade that has break risk, however you want to describe it—you have to account for that.

When you’re in names, I’m going to remove fraud unless fraud is clearly on the table.

I just use Intel because it’s a good example. The stock literally got halted, so no matter how—

Good. Certainly, it was out there that maybe this guy was sketchy. It was a SPAC, so you had to ascertain at least a 5% chance of that, right?

Or take something with huge quarterly earnings volatility, like a consumer retail name, where you really get in there and you’re like, “Wow, most of the stock falls in a quarter.” I think I would wrap that all into this: the ability to size something is directly related to your ability to ascertain and feel conviction about your downside outcome.

And that’s it. Think about it this way: let’s start with Sable because it’s top of mind versus, you know, TE or TOI. You have to respect the fact that Sable might have a California Coastal Commission or California governor scenario. Here’s the menu of scenarios where you could wake up down 25% or down 40%. That’s realistic, and it has to impact sizing.

You’re trying to think about it three-dimensionally. There are a few aspects to making a lot of money. There are the stocks that you like, and then there’s the total potential P&L in a year—from the price at which you like it to the price to which it goes. You’re trying to capture as much of that bar as possible, whether you think about it horizontally or vertically.

Bernard Baruch, or some other trader, said, “You miss the first 20, you miss the last 20, and you capture the middle 60.” That’s like Steve Cohen, too. If you’re doing it perfectly, you never try to call the bottom. The market will tell you, and you wait for a breakout.

Okay, the second-order aspect to that debate is sizing as well. So it’s really not vertical or horizontal bars; it’s really this cube, right? The other side of that cube is the potential P&L from $10 a share to $20—say maybe that’s the range of the year. How big were you at $10, $11, $12, $13, $14, all the way up to $20? Then there’s the downside—the below-the-iceberg, if you will, risk that you’re taking—which really bleeds into portfolio management as well.

So it’s not just, “Are you right?” Take the name for 3 years. What’s my probability of being able to ascertain and understand it correctly? Is it in my circle of competence? Well, it probably wasn’t at the beginning, 3 years ago, when I put the trade on. When I really ramped the thing up, I thought, “I think I’m right. My conviction level of missing something is very high. Here’s the bounded downside. I can really size this thing up.” And that all wraps up in one thing.

I would agree with you—I’ve been guilty of this as well. There are certain names you just can’t size.

Let me ask: we mentioned Nebius at the beginning, and for those who don’t know, Nebius is the old Yandex. Russia invades Ukraine, Yandex is forced to separate, and they emerge—in rough numbers—with the stock at, let’s say, $18. They’ve got $15 per share of cash, plus 3 startups that are worth anywhere from, depending on who you ask, $0 to $100 per share, and they move into the data-center business, right? That’s 1. So I just said $18 per share with $15 of cash. Now, they will burn cash on the data-center business.

Let’s just compare it to the other extreme: Sable has a billion dollars of debt, long-lived assets, and a California government that does not want them producing. If the California government wins, they’ll never produce. A billion dollars of debt—it’s a zero, right? One of the things I’ve struggled with—and I use those because we’ve mentioned them on the podcast, and they’re so diametrically opposed—is how do you have those 2 in the same book?

For me, I looked at a lot of net-cash biotechs this year, trading below net cash. How can I have those in a book with literally anything else? On those, it’s just kind of a liquidation-to-cash play, with very little downside unless management lights it on fire—which 1 or 2 of them did light on fire. But it’s tough to do that. How do you have something with such diametrically different risks in the same book?

We don’t have to have only Sable and Nebius. It’s a key point, though: Sable in 2023—I bought the warrants at $1 pre-close, and for most of 2024 as well—I would argue the duration of the investment, whether you were right or wrong, was sufficiently long that your downside was protected by volatility.

These are 100-bagger scenarios, and this is where I was going back to right-tail potential. If you are correct that people will care, and the right tail—potentially, the perception of the right tail—is big enough, right? So Sable people are like, “Okay, if you’re right, it’s $100 a share.” The stock was trading at $15 before Santa Barbara settled with them over the valves in the summer of 2024.

It’s just an option theory, right? They had enough cash. The Coastal Commission hadn’t woken up yet. You didn’t think you could die, but if you were going to die, you were going to die 2 to 3 years from now. When you’re sitting there in 2024, what is that worth? It’s worth a ton.

Nebius is too. That’s what got me there, which was, okay, the sum of the parts is kind of helpful, but if we play this out, these are the best data-center guys out there. They’re execution Jedis. They have the best relationship with Nvidia outside of any company in the United States, and they’re vetted by institutional investors and whatnot. I was like, “Okay, this is probably $75.” How wrong I was. So it’s sort of a $10-by-$75 scenario, but with 2 to 3 years of duration.

Even if I’m wrong on the sum-of-the-parts math on day 1, is it really going to get worse than $15? But this is where time becomes the enemy for you, and you have to be disciplined as well. This is where liquidity matters. Sable struggled. It bounced off $30 3 times in 2025, but by late summer, after the GEC ruling, nothing was happening with the fire marshal. All of a sudden, what you were leaning on in 2024 was an option with massive duration.

Options really start to get compressed. So if you’re not going to lean on valuation traditionally, you have to at least lean on option theory, which comes down to right-tail potential and then vol.

That makes total sense. It’s really interesting, and I don’t know if you’re following the Warner Brothers, Paramount, Netflix deal. It’s probably a little bit larger than the stuff you normally have. Yeah.

Yeah. I get a lot of people who are like, “How can you own Warner Brothers right now?” Disclosure: I own Warner Brothers, I guess. But how can you own Warner Brothers right now? There are antitrust risks. I’m like, look, there is going to be an antitrust discussion, and we can have that, but in the next 30 days, you’re not going to get a DOJ or FTC ruling. Actually, the only one you could probably get is positive if they cleared Paramount. That’s a conversation for a different day.

I would go further with this point for people, and I bring this up: after I left Distressed Debt, I went to a big pod shop. They relaunched their event-driven group. I was the non-risk-arb guy on the risk-arb team, so I got to see a $4 billion risk-arb book. This was right during AbbVie-Shire and the inversions and whatnot.

What was the inversion apocalypse? There were 2 of them. One of them was—

Pfizer. I’m not going to remember the other one; if you said it, it’s another drug company.

It was Pfizer-Allergan, I think.

AbbVie.

Then my team watched—I had fun at Neuberger Berman. They’re still there. It’s not my team anymore; I left in 2020. But I got to see 7 or 8 years of risk arb. I think it should be mandatory for any event-driven investor to study the Akorn broken merger.

Oh, I know this one. Yeah.

Fresenius, this German company, was buying Akorn, this generic drug company. They sued over a MAC, or material adverse change. It was going to Delaware court, and nobody wins on a MAC.

Akorn did kind of look like they had lied; their facility didn’t work. The stock was around $10. I think the takeout was at $32 or $33, all cash. The stock was $12 going into the trial, and you had this 25-day period before the trial started. The stock went from $11 to $12 to $18 during those 3 weeks.

And you talk to people, and by the way, Fresenius won, Akorn was zero; it went bankrupt. But to your point, and this is sort of on downside risk and narrative, right? This is everything. Everybody thought it was the most genius trade watching that thing unfold, because the people who were wrong on it—you had a lot of people who were short, and all the long guys were like, “Everybody’s short. You can’t die before the trial. There’s a trial. I know.”

The thing peaked a day before the trial started. The stock peaked, and it went from $18 all the way down. But they were like, “At $30, do your probability weight: $33 versus $0.”

I also don’t think people thought it was a zero at the time, right? I don’t think people thought that. And look, until the trial, the company is always going to—especially a company that bad—they’re always going to risk it for the trial, right? They’re always going to take that risk.

Heading into the trial, you mentioned it peaked the day before because, guess what, there might be a settlement. Every other MAC case has basically settled, because even the buyer knows, “Hey, we wanted this strategically,” right? So if we can get a price cut, the seller generally knows things aren’t great if they’re getting sued, so there’s generally a lot of room for a cut.

Akorn’s the unique case because they were supposed to be running clean rooms, and there were cockroaches running around the clean rooms. That’s the one unique case.

It’s just a freak case.

So timeline and monetizing, and thinking through an option—all this stuff comes together. One of the other lessons is that you can explore. Why do I write, going back to the Substack? Why do I write a collective? I wrote about eBay and PayPal. People are like, “Why are you doing that?” I go, “Because in everything, there’s a lesson, and you don’t know where the lessons are going to come together.”

Being free to explore is just so fun. Going back to Warner Bros., I looked at it; I put on some Netflix. I took it off because the older stuff was more exciting. But Netflix is super interesting. Where’s it now? I haven’t even checked.

Let’s pull up the handy Bloomberg. It’s $90.

I mean, Netflix is super interesting to me. You know, free cash flow finally inflected. The thing’s pooping cash. You had a negative guide into it on a single-stock basis before the deal happened, right? You had a broken bull story that looks very secular to me.

Then you have this, where even people in the long-only community throw up their hands: Is Netflix going to raise its price? If they don't raise their price, the stock is going to be in merger purgatory for 2 or 3 years. The best I could hope for is T-Mobile during the Sprint merger, which I thought was going to be in purgatory for 18 months. T-Mobile's stock actually did very well.

But that is an outlier example of something facing regulatory risk that works. So, if you're long Netflix, what you're really rooting for is that it completes on your existing terms. That resolution would be helpful.

Yeah. If you're long Netflix, you're hoping Paramount comes in at $34 and doesn't raise. The second-best thing is that you complete the deal on your terms. The worst thing that could happen is that they bomb and then take on a bunch of debt.

From a Netflix perspective, I think it's interesting because they did have soft earnings, but I do think they bought Warner Bros.—and they've never bought anything, right? When this came together, I kept telling people, “Netflix has bought 1 thing in its entire history.” They bought that small comic-book publisher. These guys don't buy anything. They're not going to buy Warner Bros. for roughly $100 billion. No effing chance.

They bought them. They won. It might not have been the best bid, but it was very close to the Paramount bid. Warner Bros. thought it was superior, whatever. I think Netflix is going to have a little bit of taint on them for the next couple of years, just in terms of investors wondering, “What was Netflix seeing in its business that it needed to go buy Warner Bros.?” Netflix is going to say, “Great synergies. We can run it the best,” all this sort of stuff. They're probably right, but I think investors, whether they buy Warner Bros. or not, for the next 2 years are going to be saying, “What was Netflix so worried about that it needed to do one of the largest M&A deals of all time, from a company that does everything organically?”

Fair. I mean, I think that's kind of the thesis from $110 to $90 on the way down, because you had the earnings miss into this. And I think that's what I was getting at: If they don't screw up—or if they lose and walk—that would be great. I think the other side of it is, who knows? The synergy number looks really juicy to me on the low side from Netflix. Only $2 billion of synergies seems exponentially low. So I think Netflix has really been hurt more by the potential merger purgatory.

My problem with merger arbitrage since I've left hedge funds, though, is that once every other year, I feel like I find myself in a merger-arb situation. I'll put on a position, and then 2 days later I'll remind myself I don't need to do this anymore. I know.

Yeah.

You do it and you're like, “Oh, cool. I'm paying for 10% upside, 15% upside.” And then you're like, “Oh, if I get hit, everyone says, ‘You have to look at 100 of these situations to find something that falls out.’” And it's usually the weird thing that falls out. That's why I'm sort of fixated on Netflix: In these 3-way mergers, they start pricing weird things. I think most people get that.

Netflix—the other thing is, I've got, again, a big position in Warner Bros., but they're so big. I do like that everybody says, when something really big happens, “Hey, there's not enough risk capital in the world to close the spread.” I've never seen that. For Warner Bros., when I was putting this on, I had never seen a stock get a $30-per-share topping offer, a hostile bid, and, on the day it came out, trade at $27. I've never seen a company in a bidding war trade below one of the cash bids, which is what I thought was so interesting there. I totally agreed, and then they went hostile. I mean, it's just the political aspect. It's fascinating.

Well, we had Pfizer—Pfizer, who was bidding for them in M&A. I mean, literally just 2 months ago, we saw that. Was it Novo that bid against them? I can't remember, but before that it had been a while—maybe Disney–Fox–Comcast. Okay, we're way off topic. This has been great. We're going to have to schedule a follow-up because I have 15 questions and notes on Caesars in Vegas that we're not going to be able to get to today, but this has been awesome.

Judd Arnold, Lake Cornelia, thanks for hopping on, buddy. I hope to see you soon. Have a good one.

Thanks so much.