On The Brink with Castle Island - Weekly Roundup 05/22/26 (AI and power bills, SpaceX IPO, Prime Trust clawbacks, USG invests in quantum) (EP.721)

Episode Date: May 22, 2026

Matt and Nic are back with another week of news and deals. In this episode:  Are AI datacenters better or worse for the environment than Bitcoin miners? AI is not responsible for your power bills Why AI could actually drive down residential power prices Is AI less popular than crypto ever was? The negative societal effects of tech companies staying private for longer We analyze the SpaceX S1 OpenAI solved one of the Erdos problems The US government is taking equity stakes in quantum computers Jane Street is still dealing with the Terra fallout Prime Trust is trying to claw back assets from Swan The SEC has questions about tokenized equities Content mentioned in this episode: Nic on Substack, AI is not hiking your electricity bill - yet

Matt Walsh and Nick Carter are partners at Castle Island Ventures. All of these expressed by them or the guests on this podcast are solely their opinions and do not reflect the opinions of Castle Island Ventures. Guests and host may maintain positions in the assets discussed in this podcast. You should not treat any opinion expressed by anyone on this podcast as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of their personal opinion. This podcast is for informational purposes only. Brought down by bad mortgage investments, Lehman, which has 25,000 employees, will be liquidated.

The federal government loans American International Group, AI, $85 billion. This is a different kind of market, and the Fed is asleep. The federal government is stepping it to stabilize Fannie Mae and Freddie Mac, the two mortgage giants that have been threatened by the housing crisis. The Bank of England has pumped 75 billion pounds more into Britain's ailing economy with a new round of quantitative easing. You print a couple trillion dollars, and all of a sudden, people start to worry.

So out of this worry, we have something called a Bitcoin. Welcome to On the Brinkum. Matt Walsh. And I'm Nick Carter. I'm recording this one with a swollen gland. You ever have one of those? It's like painful to talk. Yeah, I can see it.

It's tough. You can see my gland? It's looking large. Oh, that might just be my neck. Busy week. You came to the rescue of the AI industry with a very long substack here around energy usage. Yeah, you know, I feel like I just warmed up with the whole Bitcoin energy usage thing. And now it's the same exact fight.

It really is, isn't it? It's like, it's funny how they just reuse the same arguments. but I'm not as deep in the AI data center world. What's the verdict here? Is AI just drowning the world of electricity? You know, AI is better and worse than Bitcoin mining. It's better in some ways, worse than other ways. Bitcoin miners are actually more benign than AI data centers

because Bitcoin miners are interruptible. So they can turn off and on without too much issue, right? So that makes them more suitable for like demand response programs. So that's why in Texas, they're actually quite, useful to have. AI data centers so far have not figured out how to do that. So that's the one thing. But yeah, I had a very long piece.

I made all the charts with Claude. Claude is very good now. Clod's really good. Very good. It's my personal AGI. We've reached it or post AGI. That's what Mark Andreessen said this week, said we've reached AGI. Claude is better at making charts than me.

And making charts is the one thing that I thought I was pretty good at. Yeah, I mean, I don't know if you'd be employable anymore if you're 22. Yeah. So I'm very sad about that, but I get to make a lot of charts now, which is great. So a lot of people are saying the AI is driving up there, energy bills. That's basically not true. It's sort of true a little bit, but it's basically not true. So there's just one chart that I think is really important,

where if you divide the states into quintiles according to how, much energy data centers use within that state. So divide them all up into, so the most energy intents from a data center perspective states are, Virginia, Texas, Nevada, North Dakota, et cetera, right? Divide all the, and then you have a bunch of states with basically no data centers. So then you track their real, not nominal, real residential electricity rates

over the last decade. What you'd expect to see would be a massive move up in power prices for the highest quintile, right? What you see instead is the most data in 10 states have the lowest electricity prices throughout the whole period, right? Which that's not that surprising, right? Because if you're deciding where to put a data center, you probably put it somewhere with cheap power, right?

Right, right. But also the lowest rate of growth in residential electricity prices. of all the quintiles. That's interesting. And what states are those? Those are Virginia, Nevada, North Dakota, Nebraska, Iowa, Oregon, Texas, Illinois, Arizona, and Wyoming. So those are the top 10 data center states on a relative basis.

And they have the lowest increase in residential prices. And then I look at the data of like 10 different ways. But that's the main thing for me. Yeah. But the cost of data centers has not begun to bite yet. The one exception is PJM, which is the power authority in Virginia, Maryland, Ohio, Pennsylvania, a few places. They had a capacity auction this year. Virginia, 25% of Virginia energy is data centers, believe it or not.

That's actually really good for Virginia because those companies pay tons of taxes. So if you live in Loudoun County, your property taxes are actually much lower than the others wise would be. So kind of a total opposite of, you know, everybody's complaining about data. Data Center is going to be very good if they come into your neighborhood. Those capacity auctions did raise prices in PGM. So now you get Maryland complaining about it because they pay the higher prices. Virginia gets the benefit. So it's a little unfair.

but we're only talking about a 5% increase. So it's not overwhelming. So I'd say based on my analysis, data centers have not driven up the price of power so far. The states where power has increased a lot, it's either due to inflation. That's across the board. Inflation, not an AI thing. And it's actually mostly blue states. So the three worst are Maine, California and New York and Massachusetts.

Massachusetts. But that's just because those states have pursued renewable energy transitions. That's expensive. They haven't built pipelines, especially New England. California has this whole separate problem with wildfires. Those are not AI things. Yeah. So in my opinion, AI is mostly blameless. And there's another thing, which is when you get a large load on the grid, they are buying a defined amount of power and they're drawing it very predictably over a long period of time, that can actually lower power prices, ironically, believe it or not, because the cost on the grid is mostly a function of fixed costs, right? So the cost of infrastructure, transmission, transformers, literal power lines, things like that. And that cost has to be,

you know, divided across all the people that are buying power. Right. So if there's a certain amount of capacity available and a new buyer comes in to scoop up a lot of that capacity in a relatively benign way, not involving lots of spikes, that can actually reduce what residents pay because now there's a larger pool of capital, you know, effectively paying for all the grid improvements. So in a situation where the grid is not being strained, a new buyer of power can lower residential prices. Very counterintuitive finding. That's interesting. The AI people need you in their corner, man. You got to get out there and defend them like you did the Bitcoin miners. Yeah, I mean, I don't feel as motivated to do it, but I think the facts are actually very good for the AI people. Do you think that

AI is just less popular than even crypto ever was in this country? It just feels like the tides have really shifted. You have people booing AI. related speakers at commencement addresses the past couple weeks. It feels like there's a vibe shift here. Yeah, it's really weird because AI is more widely utilized than crypto for sure. Yeah. So even the people that hate on AI use AI. I mean, basically everyone uses AI. You can't not use AI. Right. And everyone benefits from it. So why do people not like it? I think there's anxieties around job loss. You know, people feel like their profession might just be blown away. way in the next few years. Creatives don't like it. Some people don't like it for IP. But I think

that's mainly that people acknowledge that they use AI, but they have serious anxieties around their own employability. I think the job loss thing is a big part of it. I also think just the fact that people have not been able to get financial exposure to some of these trends is not great for the economy. So Open AI Anthropic, if they were public like two years ago, I think that would have been better for society. Yeah, it was just thinking about that. I mean, it used to be the case. that tech companies went public when they were much smaller. Yeah. And that prosperity could be shared.

And now you have to buy these things at a trillion dollar plus valuation. It's very unfair. It is. Sarbanes Oxley just had really perverse side effects there. It's too bad. I mean, you used to be able to ride a, you know, it wouldn't happen a lot, but you used to be able to have public companies deliver 50x returns, kind of like venture style returns in the public markets.

those are so much harder to find. Yeah, and you're going to see all these stories in the coming weeks about the SpaceX IPO. I mean, there's venture funds taking down decadillions from the SpaceX IPO. And it's just prosperity is not being shared. It's not. It's not great for the country. So the SpaceX IPO, I guess before we get into the deals, we can talk about that. There's a lot of interesting things in this S-1, but one of the more interesting things in our industry is that they have 1.45 billion.

dollars worth of Bitcoin on that balance sheet. So that's more than we thought, right? Yes, definitely more than we thought. And, you know, obviously Tesla has a lot of Bitcoin as well. So Elon has, you know, he's been on and off with Bitcoin, but you got to just look at the scoreboard, I guess. It holds a lot of it. Yeah, this IPO, I read the S1, I'm very troubled by it. Elon, obviously you never bet against Elon, never ever.

but I think the bulk of the value is future expectation of growth, which is not clearly evident from the filing itself today. Yeah, I guess that's how SpaceX always was, though, too, right? I mean, you're just looking at the total addressable market for things that you could do in outer space is huge. I mean, you're talking about mining asteroids, you're talking about just space travel. There's a lot of places this could go.

Yeah, I mean, the telco aspect of the business is really, they have the best kind of monopoly ever, which is a fleet of satellites. I totally buy that story. I mean, everyone on Earth is now connected. They're going to have a cellular product. The tam there is enormous. You'll have hundreds of millions of people using it.

That's great. The data centers in space thing, I don't know. I'm a skeptic. I hate to say it. I'm a skeptic. You know, I don't know enough about data centers to give an opinion on this, but it just seems like there's a lot of maintenance that could not happen in outer space. So just how these things are even built in the first place, you know, there's no one to,

like, go turn it off and on. Yeah. And venting heat in space is hard. It's harder because there is less matter in space to radiate the heat away. Now, I guess ultimately, if there's one entrepreneur that could figure it out, it's probably Elon Musk, right? Yeah, that's right. So I'm not, taking a position long or short on this one. Just going to let Elon figure it out. Anthropic and opening eye I feel a little better about, but I

actually still worry about those. Well, they're just highly levered business models, not even in the traditional sense of the debt load. It's more of just the kind of the forward contracts that they've agreed to.

And so if you miss an upgrade cycle there on a model, it probably could get pretty hairy on the churn. The thing I worry about is will these Chinese open source models that are distilled versions of their frontier models be as good and much cheaper?

And how do you stop those? Gavin Baker was on Invest Like the Best, Patrick O'Shaunce's podcast this week, talking about the game theory around releasing some of these frontier models and trying to prevent them from being distilled. So saying that, you know, is there a world where anthropic and open

AI don't release API access for their latest models because of that reason. But then of course, if someone does, then it forces the other one to release it. I mean, meanwhile, though, opening I just announced their model, just a general model solved one of the Erdos's problems yesterday. And what was the, so that's, you know, that's a math problem, but what math problem is that? It's, so. So this is a conjecture in discrete geometry. This is the planar unit distance problem first proposed by Paul Ordos in 1946. It's one of the best known questions in combinatorial geometry.

Ah. So I don't know if that means anything to you, but. I'm pretty sure Goodwill Hunting solved that, actually. Will hunting solve that on Blackboard, if I recall correctly. The question is, this is the question. You ready? if you place end points in a plane, how many pairs of points can be exactly one distance, one apart?

Is that the whole question? That's the question. Okay. That would have, yeah, that would probably take me a couple hours to figure out. Yeah, so I couldn't tell you, I don't actually know what the answer is, but apparently chat GPT has the answer. That's pretty incredible that chat GPT has the answer to that.

I wonder, I would love to understand the method of solving that problem. are there like Reddit posts that are kind of close and it just fixed it or fascinating? So this is a big deal because yes, AI has solved various lemmas and conjectures, but this is the first time it's solved a problem that mathematicians have heard about. Yeah. This is the most high profile math proof ever developed by a non-specialized model. It's not a math model. It's a general model.

Crazy. It does make me think about quantum. We have some quantum stuff to talk about today, but people that are betting that technology will not advance quickly, it's a bad bet. It's a bad bet. Acceleration has begun. All right, let's hop over to the deals of the week.

First one up is a crypto ATM company, Bitcoin Depot. They have filed for bankruptcy protection. They have cited an unsustainable business model. That's kind of how I always felt about the whole category, to be honest with you. Yeah, that's sort of. an anti-deal there. Next up, Deloitte has acquired the team at BlockNative, a blockchain infrastructure company, focus on Menpool monitoring. Congratulations to them. Then it's Checker, which is a stable coin infrastructure company. There is $8 million from Galaxy

Digital. Katana Labs is an AI Native financial services platform. There is 30 million from A16Z, accrue general catalyst in Coinbase. Then you have Town Square, which is a cross-chain yield infrastructure platform. There raised 16.3 million from World Liberty Financial, Amber Group, and OKX ventures. Decent XYZ is a cross-chain liquidity platform. They're required by Moonpay. Fourth acquisition this year by Moonpay. Those guys are really rolling up the industry. Then last one is cycles, which is a blockchain clearing protocol. They raised 6.4 million from blockchain ventures and Coinbase. Those are the deals of the week. We are a little light on the news. The news item in the week for me was the U.S. government is now buying equity in quantum companies.

Crazy. Crazy story. So it came out in the journal this morning. The U.S. government is awarding $2 billion in grants tied to nine quantum computing companies. These are deals that will include equity stakes in some cases. I'd be fascinated to see the structure of these. Companies that were listed in the journal article are IBM, global foundries, de-wave quantum, reggae. computing, atom computing, cy quantum, quantum, quantium, and inflection. So what do you make of this?

Yeah, it's pretty fascinating. So the government's doing industrial policy. I guess they own a lot of Intel as well. Is Intel one of the nine here? Because I just listed eight. No, that's obviously to do with chips. So it's not a quantum thing. But pretty interesting.

The journal didn't really specify, the purpose of the deal, but quantum obviously is of immense strategic value to the government. So it makes sense that the government would work very closely with the private sector. And so this formalizes that regardless of what the first applications end up being, what we know what quantum will be good for, code breaking and physics simulations, whether it's recovering Bitcoin or stealing Chinese state secrets, the government will want to be the first user of a scaled quantum computer. So I don't know if this is necessarily a Bitcoin thing,

but I think it's very indicative of where this is going, which is the government's going to want to be involved. Yeah, in one level you're like, okay, we don't need the government taking equity positions in companies. That's not what the government should do. on the other hand, it's like, yeah, but this is a complete arms race, and this is a national security issue around quantum. You don't want to live in a world where the U.S. doesn't have access to this technology and China does. Yeah, so I've hypothesized that the solution to the Satoshi coins matter could well be one of these private companies,

recovers of the coins at the behest of the government. There's a legal framework to do it. And then Satoshi can go claim the coins, whatever. That seems to be getting more likely by the deal. day. A hundred percent. I mean, if you're the government, again, back to how these deals are structured, first of all, can I see the investment memo? I'd like to maybe take a look at those. What was your diligence process? How many experts did you call? Let's get a little bit more detail on this. But let's make sure that the term sheet says, to the extent that part of your business model is going

after Satoshi's coins, which by the way, several companies on this list are doing that. That's part of their strategy, that the U.S. government gets the coins and you get a take rate. Yeah. So this isn't science fiction anymore. Yeah. And I think if you are any of these companies, you would much prefer a deal like that where you're not just stealing the coins, you're lawfully salvaging the coins and you get to keep a percentage. Yeah. So really interesting deal. I mean, there's a lot going on in Washington. Last week, we talked about the Clarity Act and where that stands. John Thune just sent Senate home basically for the rest of May. And so there's a lot of, there's a lot of kind of pushback on that fund that Trump proposed around making restitution

to folks that were prosecuted by the Biden administration unlawfully. It seems like that's slowing down the reconciliation bill. It's going to delay the Clarity Act from getting brought in front of the Senate as well. So just a lot of moving pieces here. There is an interesting development in this Jane Street case. So they were accused of insider trading Terra back in 2022. There was a new court filing in Manhattan alleging they used a private telegram back channel with

Terraform Labs insiders to dump 192 million of TUSD just before it collapsed. This group was called Bryce's Secret. Who's Bryce? And what was his secret? It was apparently a former Terraform intern Bryce Pratt, who later joined Jane Street. And was his secret that Tara was a Ponzi? Yeah, I mean, that wasn't a secret.

Yeah. We all knew that. Yeah, that's kind of a known thing. Yeah, I mean, if I'm Jane Street, I'm selling UST as fast as humanly possible, no matter what. I think kind of everyone was. It would be interesting to see where that goes. That kind of reminds me prime trust. We hadn't talked about that.

So I feel like we hadn't talked about prime trust in like two or three years. This was the custodian. Where are they based? Nevada, something like that. Yeah. You know, they claimed to be a qualified custodian and they went bankrupt and a bunch of the Bitcoin is missing. And over the years, we've talked about them in the context of they had a failed like

fireblocks upgrade at one point where they had coins that were sent to an old address that they didn't control. But it looks like they were just running a deficit on Bitcoin. So I don't know where those Bitcoin. went, but they're in bankruptcy now, and the estate manager is going after swan Bitcoin, strike, and a number of other Bitcoin-related front-end companies, brokerages that use prime trust on the back end. So it's, I mean, you kind of feel for the companies that were the customers here that withdrew funds because they're, you know, they're being asserted to have

fraudulently conveyed on these transactions. But when you look at that, you're like, well, these guys said they were qualified custodian, you know, like, what do you want us to do? We're actually bifurcating between brokerage and custody in traditional markets. That's a good thing. We obviously picked the wrong custodian, but they told us there were a QC. Yeah, I'm hesitant to side with Swan on anything, but I'm a little sympathetic to them, frankly. Although Blockspace reports that the complaint alleges Swan had access to inside information about Prime's financial distress and use it to withdraw assets from Prime. And so they're now seeking to claw back those assets.

This is a lot of Bitcoin, by the way. How much Bitcoin? Are we talking 12,000, 12,000 Bitcoin. So this is worth about a billion dollars. But good luck getting that back. I mean, because Swan's customers have already withdrawn that strikes, presumably the same thing, right? So this is going to be a, I wonder if this is going to go down to the end users of Swan.

Yeah. I mean, my takeaway. is just because someone is regulated doesn't mean they are solvent. So due diligence, demand real proofs of reserve. And then for the end clients, if you're dealing with a company that's just a thin interface on top of one of these custodians, be mindful of that. Look at these companies that are full stack and have a proof of reserve. Again, that's just really, that's hard. I mean, these companies were operating in a space where Gary Gansler is still the head of the SEC. see he was coming out on a monthly basis and saying that there needs to be this bifurcation

between custody and brokerage. That's the model they did. You know, you feel for them. Now, if they had inside information, I think that's a different story and that's a whole different consideration set where, okay,

maybe you still should be pulling funds off, but then you probably shouldn't allow your customers to pull those funds back. He should just wait for the court process to play out. But talk about a mess. Yep. elsewhere blockchain.com has confidentially filed for their US IPO. Another confidential filing that is fully public. I fail to understand what the point is there, but that'll be interesting. You know, does any IPO matter after SpaceX, Anthropic, and Open AI?

Yeah, I'm feeling a little bit of ennui about it because basically you just have to be in one of these giga AI companies, and that's the whole game. I was talking to a, pretty legendary enterprise software investor outside of crypto yesterday. And that's kind of what he said. He's like, it feels like my very best companies are just not getting any attention right now because everything is just Open AI, Anthropic, and SpaceX. Yeah, I was reading an informational article and they said that over the last year, Anthropic and Open AI have actually increased their share of AI revenue in the whole AI space. So not only have they 10x or whatever, 50X their revenue. They've also increased their relative market share throughout that time.

Yeah. Because they're gradually cannibalizing all of these, you know, applications that are built on top of them. It's a very strange economy we're living in. I mean, you have publicly traded enterprise software companies right now trading at a 1x revenue multiple. And all these CTOs of major tech companies are just leaving to become ordinary employees at Anthropic. Carpathie did the same thing. This week. Carbothy, yeah. In other news, South Carolina, they've passed a law this week. It is banning central bank digital currencies in the state of South Carolina.

So if you want your CBDC, you want to pay with your CBDC down there, can't do it. The law also protects Bitcoin miners and cryptocurrency users. So I get it. I mean, North Carolina tried to pass a similar law. It seems like a lot of these southern states are going this direction. They don't like CBDCs. Yeah, I mean, very low. risk of a CBDC existing anyway federally, but yeah, it's nice to see them doing this.

Yeah, I mean, but if it does exist, it's not going to exist in South Carolina. All right, I want to talk a little bit about this. Bloomberg had a piece. It said that the SEC will soon be coming out with a innovation exemption relating to tokenized equities. And it's a little bit light on the details, but in that Bloomberg article, they talk about giving the ability to tokenize an asset without the sponsor's permission. So I would interpret that if it's true, if that's actually going to be part of this text from

the SEC, that a third party could take control of like Tesla shares and issue a security token on top of those, which I find fascinating. And if that's the way this market structure evolves, I think there's going to be just a lot of replumbing and re-architecting that needs to happen in financial services. because what you're going to see is this balkanized liquidity landscape. And it's almost like a rewind the clock before reg MMS in a lot of ways. We're just going to have like a really hard time, I think, determining best execution on these things. You're going to have tokenized versions of things that sit with highly creditworthy counterparties.

You're going to have others that are not highly credit worthy. Maybe they're all bankruptcy remote, but I don't think they'll trade at parity. And I think you'll also see just massive price dislocations on these various venues, depending on just how liquid they are, how easy they are to use. I suspect that this is actually going to be really good for proprietary trading firms. You know, on one hand, you could look at this and you could say, hey, that's not great for Citadel because Citadel has a phenomenal market positioning in the existing Tradfine market, and flow is going to go away to other venues.

But that kind of assumes that Citadel does nothing. Like if Citadel goes out and says, hey, we're going to connect Citadel Jane Street, Hudson River, like any of these guys, we're going to just go connect to every venue. We're going to connect to D5 protocols. I think you're going to make a ton of money because I think you're going to have shares of Tesla trading at three cent gaps depending on, you know, the venue that they're trading on. So I think we're in for a really interesting time here for financial plumbing. So we could have tokenized equities of the same firm trading at different prices based on who's issuing them? That's what I, again, the language is not out by the SEC. The report was that there is some guidance coming and it alluded to the fact that it would have to do with tokenized securities and the ability to issue a representation of a token without asking permission of the underlying company.

but if that is the case, then for sure you're going to have, you know, it will not be a fully efficient market, I think, because you're going to have, you know, Cracken will be in this business, Coinbase will be in this business, Robin Hood, you'll have stuff trading and D5 protocols. It's really going to be a brave new world, I think. And a lot of these firms that are lobbying against us, Citadel included, are just saying, like, hey, we can't manage that. Like, that's a huge upgrade to our technology stack. But you're going to have other firms that step in.

I could actually see like a crypto native firm like a winter mute deciding to get into this and saying like, hey, we're going to become super efficient on touching all these venues. We're going to be connecting these people. Companies like Talos that we're an investor in, you're going to have these companies that see it early and start connecting the pipes. And I think it's going to be the underlying prop shops that do really well. And do you think this will be just for public equities? No, I think this is going to be for private. I think this could be for anything. I think you're going to have tokenized private shares.

I think you're going to have tokenized funds. I think that's a huge category. I think credit funds, real estate, like REITs will be a big category there. I think you're going to have a bunch of esoteric assets that get tokenized this way, like art and whiskey barrels and things like that. Yeah, we're going to have to iron out the kinks in the tokenized private equity market because you saw what happened with the tokenized anthropic tokens just recently. Massive dislocation, unclear the relation to the underlying.

I don't see how that can exist, the status quo there. No, and I think they're going to have to differentiate between public equities and private equities, to your point. So I think doing this with a private equity, you can just write into the operating agreements as the private company that we don't allow transfers, things like that. So there are legal ways that you can prevent this. So I interpreted the Bloomberg piece to be referring more to the public equity side of this, is that you could immobilize something.

I think of it almost like an American depository receipt. We'd have the Japanese or Chinese companies, but they trade in the U.S. because they get immobilized. The shares get immobilized with Bank of New York Mellon, and then there's representation of those shares. Like new units are created. just representing the underlying holdings of the ADR. So I think this is like ADRs for crypto.

The National Hockey League and the CFTC signed an MOU aimed at ensuring the integrity of prediction markets tied to professional hockey. Yeah, memo of understanding. And I looked into this a little bit more and it's like an agreement to have meetings. So I have memos of understanding with so many people right now. I'm trying to reduce my memos of understanding. Yeah, I think the issue is in match fixing. Yeah, I mean, all professional sports, I think, have this issue.

Yeah, that was kind of an issue before we had prediction markets. Boston College men's basketball team, legendary fixers of games. Is that right? I think that was in the 90s. Oh, yeah, the BC men's basketball team got hooked up with the mafia, and they started fixing games. You don't really hear about match fixing it. It's not really a modern problem. No, but it seems like there must have been a glory days of the, you know, you call it match fixing.

Yeah, why would you call it? Yeah, fixing the game or something, but. Oh, I guess that's like a British term. Yeah, is it? What do Americans call it? Yeah, like fixing the game. He fixed the game. Okay.

Yeah, you don't hear about it as much anymore. But didn't you, you had the NBA referee a few years ago, Donahey. He was kind of doing this, right? He was betting on his own games and he was calling more fouls on certain players. I just feel like it's very easy to catch. Like, if the lines move suspiciously, you know, you can kind of just see that. Yeah, some of these, too, it's like over anders on player minutes. The player could just pretend he's injured.

Yeah, it's like, you know, when people inside a trade on public markets, they just buy short-dated options and then, you know, an acquisition gets announced. It's like very easy to catch that. Yeah. You'd think it's kind of the same. in a soccer game or something. You would. Yeah, you would. Well, in soccer games, I mean, no one scores any goals anyway.

You just run around. People pretend that they're hurt. They start crying in the middle of the field. So I take you, you're not going to the World Cup games. And I think there is some in Massachusetts even. Yeah, there's some down in Foxborough. No, I won't be there. I think it's a...

I don't think the U.S. is playing there anyway, though. So I've looked at the prices, and I'm scandalized by it. Super expensive, right? In Miami for the group stage games, the floor price is like $2,000 for a nosebleed. Wow. Who is buying these? Yeah.

I think it's like $100 to get to the game on the train as well. That's pretty expensive. Yeah, and the team, they expanded the teams in the World Cup. So now you have, I'm exaggerating, but like Kurosau versus Turkmenistan group stage game, and $600. Why? Who's watching that? I don't know. Who's playing on the U.S. team? Do we have anyone good?

Yeah, we got Pulisich, others. I'm actually an England fan as far as soccer is concerned. Oh, come on. You're... What? You got to root for the U.S. England's better. No. England is a better chance at eventually winning it. I mean, I will promise you this. If the U.S. makes it to the World Cup finals, I will watch my first soccer game they probably won't do that no but they could save it they could make it to the quarters of the summies that is within the realm of possibility oh really okay yeah it could watch it i'll check the updates on my phone okay just say zero zero the whole time

then there'll be a penalty kick at the end we have like a columbia portugal game here in miami and that will be chaos i can assure you the yeah the fans will be breaking in to the stadium, you know, that kind of thing. All right. I think that is it for the week. Busy week. Everybody have a safe and healthy weekend and we will see on Monday.