国会失灵为何利好加密货币

My takeaway is that we’re still just incredibly incompetent as an industry. Why does he have his signing keys on the same laptop he does all his work on? For $6 million, right? Come on, guys.

The incumbent always loses the midterms. This is what happens.

Bubbles don’t pop when sentiment looks like this.

Pretty good time to be long.

There’s always an elevator down. It’s always an elevator down. I’ve never doubted that. Buddy, I’m too old for self-custody at this point, man.

You make one bad political move, and now everybody fucking hates you.

What the fuck do we need an L2 for? What do we need another chain for?

I’m kind of happy there’s a bit of gridlock and he can’t do as much crazy shit.

All right, fellas. We’re recording this the day after—or the night after—the midterm elections, which is probably the biggest topic on everyone’s mind right now. To set the stage, we came in with a Republican trifecta across both the presidency and Congress. There was what many have referred to as the Democratic sweep, the blue sweep, which I’d like to think was expected by a lot of people you talk to, even if it wasn’t necessarily priced in.

Last night’s results not only confirmed that, but I think probably went even further than some people expected. So my question to you guys is: as you wake up this morning, how do we feel about the results? More importantly, what are the big implications going forward that you’re watching, specifically as they relate to markets, bags, holdings, positions, and theses? What are the takeaways?

Yeah, non-factor. I either said this 2 weeks ago or a week ago: chop until midterms, and then after midterms we go higher. Verbatim. This is what happens every midterm election cycle. Clinton got smoked in the midterms, lost the House and the Senate. Bush got smoked in the midterms. The incumbent always loses the midterms. This is what happens. It happened to Clinton, happened to Bush, happened to Obama twice. It happened during Trump’s first term, and it happened this time.

This is just generally how it works. The pendulum swings one way, then it swings the other. People are pissed about one thing, and it galvanizes the base. They turn out, and then it swings the other way. That’s how it’s happened historically. Markets like clarity, and we have clarity on this front. You just end up in gridlock now. Very few things will happen over the next 2 years.

It’s bad for the AI trade for sure, because you’re going to see data center build-out slow down. In this case, I think it’s bullish crypto. One, there’s less of a capital vacuum going there, and attention kind of shifts back here. I’m not saying we go to all-time highs, but I think this is a very constructive environment for it.

I don’t really think there’s an insane amount to read into it, because this is just exactly what happens during every election cycle. It played out as it always has historically, where incumbents lose the House and the Senate, and you end up in a scenario where there are constant blockages on both ends. The president still has veto power, but the House and the Senate can prevent anything from really taking place, including appointments or anything along those lines. Both sides are pretty defanged, and you end up in limbo for some time until the next election.

I think it makes me more bullish on my store-of-value thesis as a good corner for crypto over the next 2 years, because I agree the AI trade probably slows down a bit. But I’m not sure that’s necessarily a bad thing. The slowdown could temper the risk of a big bubble blowup, and maybe things are more moderate. That’s healthier.

I think speculative capital will come back to crypto, and store-of-value assets are the easiest to bid because they’re the safest from any potential harsh upcoming regulation and pushback in a couple of years. I’ve been getting more and more bullish on Zcash recently. One thing I hadn’t really thought of, and then I saw this LBO guy who works for MetaDef tweet it, is that privacy is actually really good for cyber defense.

If you think about the Binance hack, if their hot wallets were all private, it’d be much harder to identify and impact them. I think that’s a pretty bullish tailwind for privacy, and then, again, bullish store-of-value assets. Zcash just has a lot of good tech. I think I’m pretty confident Zcash is going to go for the next year, whereas a few weeks ago I wasn’t totally sure. But I think it’s set up very nicely right now.

You weren’t sure there’d be another leg up on Zcash a few weeks ago?

Into year-end? Yeah, I wasn’t sure.

Oh, okay. Into year-end.

Yeah.

But now you are? Now you’re sure?

Yeah. Well, I’m never sure, but I am confident. Some of this is biased because it looks really fucking good over the past day, right? But if it can really crank through 1,300—which it’s at now, 1,320—I don’t know. Everything looks good, though.

It’s probably just the certainty, right? People like certainty, and we have the certainty now that nothing’s really going to happen for 2 years.

I don’t think it makes that much difference to the AI trade, actually.

What do you think, Jose?

I mean, I think Yan’s thing is true, but it’s a bit of cope. It was still a landslide destruction. I think it’s the first time Texas has had a Democratic senator since 1988. It’s just a repudiation of all the bullshit, I think mainly the Iran war. We were kind of talking about it in the chat and the graphs.

You can’t say it’s cope if it happens every time, though.

But it hasn’t.

Like—

Texas hasn’t had a Democratic senator since 1988, right?

Right. You have a point.

It’s like you lost Texas.

All right, that’s one thing, but if you look at the actual numbers, Clinton lost more seats in the House, and Bush—or Obama—lost more seats in the House and more seats in the Senate. The magnitude of the losses has actually been worse historically, if you really want to talk about the degree of the sweep.

The magnitude across the House and Senate was worse historically than this Trump loss?

There have been worse instances. Obama’s was worse.

Okay, there have been worse.

Obama’s was worse. Clinton’s first term was worse.

Yeah.

So—

Yeah.

It’s just like, it happens—

I just think that—

—like every time.

But on the—

It happens, but I think the big signal is what it tells you, right? People vote with how they feel. We can talk until we’re blue in the face, and politicians can talk until they’re blue in the face, about how markets are up and the economy’s doing really well. But I think the through line was that most people don’t feel that, right?

There were missteps that happened, as happens with just about every presidency at this point. To Yan’s point, that’s why you often see this flip-flop. But the big things were high energy prices, high inflation, and a real squeeze on consumers. I think it’s a giant reboot of what’s happened over the last 6 to 12 months. But sorry, Jose, you continue.

No, I agree with what you’re saying. It’s just a shame. I never liked Trump, but I was pretty excited about the whole thing—about Bessent and having these smart businessmen around, and Musk being around. It felt like they really had the mandate from heaven to do really cool stuff. I really believed in a lot of the platform, like no wars.

I think he’s actually crushed AI regulation, and a lot of the vibes around the woke stuff are way more reasonable now. I don’t know if we… I think the tariffs were a mistake, but the big one that affected it was the Iran war. It’s just a shame to see, because now I think it’s really tough. Now you give power to the leftist lunatics, the sort who want to basically implement communism, and I don’t think it needed to happen.

I really think he had everything to make it work. So it’s a bit of a shame. But I don’t think it affects the AI trade that much, honestly. I did some research on it this morning. It seems like, at least for 2027, basically 6% of the 35 gigawatts or whatever that SemiAnalysis projects in the U.S. is at risk because it doesn’t have electricity permits yet.

I think there’s only 6% at risk, so it shouldn’t be too bad. 2028 is a bit worse, at about 20%, but still not catastrophic. Obviously, they could pull some more tricks, I guess, but it doesn’t seem terrible.

Even if data centers came in way below target, I don’t think it would be bearish for everyone. Some of the people who built out a lot of existing capacity would do really well. It kind of sucks for the neoclouds, because spot compute prices would get squeezed further and interest rates would go up.

But for the hyperscalers, for Meta and Google, it’s pretty good. For Nvidia, TSMC, and the semiconductors, it only really affects them if we get to the point where so few data centers are being built that it affects them. Right now, chips are still a big bottleneck, so I don’t think it’s that bad or changes that much.

Honestly, I think gridlock is kind of a positive thing, given the way Trump was going. I’m kind of happy there’s a bit of gridlock and he can’t do as much crazy shit. Yeah.

Gerson seemed to think the build-out estimates were going to come down pretty materially.

Really?

Yeah.

What reasoning did he give for it? Any specifics?

I don’t remember exactly. He was on All-In’s midterm livestream. I think he has very good takes generally.

Okay.

The build-out over the next 2 years—or the next year—was supposed to be larger than the cumulative build-out we’ve had so far. He’s marked that down by quite a decent amount.

Like what? Did you say percentages?

I think it was 30% to 40%.

Wow, that’s a lot higher than I thought. All right, I’ll recheck.

For 2028, I could see that. For 2027, I think that’s probably a little overestimated. And SemiAnalysis, to Jose’s point, has come out with some really good pieces. They even had one on the midterms that I was reading through, and it becomes much more of a 2028 story.

One of the big through lines, too, is what we said last week: the premium is going to shift toward companies, especially the hyperscalers and maybe some of the big incumbent neoclouds that already have compute installed—or, basically, operational compute. They’ve not only built the data centers, but they’ve got all the permitting, and they’re actually able to service compute, make money on it, and monetize it.

I think this is going to serve as a centralizing push, because those are the companies with the most bandwidth and capacity to absorb higher costs. A big part of the pushback against data centers isn’t just shutting off or pausing development entirely. In a number of states, additional builds are contingent on data centers giving up some tax incentives, covering more infrastructure costs, and potentially providing subsidies when they drive electricity prices up.

As long as you make those changes—which cost money—and have the capacity to do that, you can get through some of these approvals. They aren’t broad-based, blanket shutdowns. The companies best positioned to do that are the ones that already have the capacity, or the capital and financing to weather it. It’s a capital game, a massive capital game. There are only a handful of firms that can really do that.

Even though SpaceX was down premarket this morning—I haven’t checked since—I think this cleans things up longer term. If Elon had to pivot and find somewhere else to build data centers, who else on the planet would you trust more to figure out a solution?

The estimate was that total U.S. cumulative capacity built out through the end of 2026 is 38 gigawatts. SemiAnalysis’ previous forecast for 2027 was 43 gigawatts, an additional 43 on top of the 38. He thinks 10 to 15 gigawatts of that won’t come online, so that’s 20% to 30%.

Of 2028?

2027.

He thinks 10 to 15 gigawatts of 2027 won’t come online?

Of the 43 that’s estimated. So that’s roughly 25% to 40%.

I thought 43 was 2028. Or is that what SemiAnalysis said?

No, I think—

43.

60-plus is 2028. 43 is 2027. He thinks 10 to 15 of that, which is roughly 25% to 40%.

Yeah, that’s crazy.

That’s pretty material. That’s not a de minimis amount.

I wonder what the breakdown is between new operators and more seasoned data center operators or hyperscalers. A big portion of the supply expected to come online is being built by first-time operators, and we were already seeing delays before this. It’s really hard to do this—to bring that amount of compute online on a short timeline.

To this tweet’s point, I think it winds up being good for hyperscalers and the SpaceXs of the world. If you own compute, the premium on that only gets higher.

Maybe we don’t go up as fast, but we don’t go down as fast, you know? That’s the angle to look at.

There’s always an elevator down. It doesn’t matter. You can either take the elevator up or the stairs up. It’s your own choice.

I mean, we get elevators up and down these days. COVID was an elevator down and up. The tariff war a couple of years ago was basically— That was actually a step— I don’t know. Anyway, do we want to keep talking about midterms? I think we’ve talked enough about those.

Hey, man, as long as my bags go up, I don’t care. I’ll be honest.

Yeah.

I think my opinion on the midterms is—

I’ll be frank.

The market’s up today, so they must be a good thing.

I think markets are approaching a pretty interesting moment in time. The midterms are now behind us. When you look at how equities typically perform over the 3, 6, and 12 months after midterms, it’s pretty much regardless of who wins—though especially with divided government. That’s typically one of the best periods for equities, or at least the hit rates are really high.

Since 1950, the S&P has been up an average of 14% to 15% over the next 6 months, with pretty much a 100% hit rate. There can be drawdowns along the way, but that’s why time horizons really matter when we talk about this. If you’re trading around it, could there be a drawdown in the next 4 to 6 weeks? Of course. But broadly speaking, looking a bit further out, getting past the midterms is definitely pretty bullish.

What’s been interesting, though, is that the equity market has become very top-heavy when it comes to AI or AI-adjacent revenue and earnings drivers. Look at what’s happened with big tech earnings this quarter: a lot of these stocks are getting crushed. Some of them may even beat expectations, but their guidance is off. This sentiment that there isn’t enough ROI, and the questions about whether AI will eventually be monetized in a very big way for these companies, are throwing a little cold water on the AI trade.

I wanted to bring it up because I’m always looking for moments that could be good buying opportunities. The Nasdaq is very AI-centric, but after the recent sell-off it’s now basically bouncing off its 200-day moving average. It’s been the most oversold since similarly sizable pullbacks in July and August. We saw a really big one in early 2025 with Trump and all the tariff tantrums.

Then there are indicators like the Spec Flow indicator that our markets team built out—Scott on our markets team built it. It tries to use intraday trading flow as a rough proxy for potential speculative capitulation among traders and investors. When it gets down to these low levels, it gives you a read that sentiment is not only poor, but that’s showing up in price action and trading activity. Speculative flow is getting more negative, in a sense, and sentiment around it is turning negative.

So you line up all these, and again, they look like classic buy-the-dip moments. Then, when you add in the fact that we just went through one of the biggest deleveraging events, especially for long-short hedge funds, arguably on record—or at least over the last 10 to 15 years—it ties back to something we’ve been talking about for weeks: bubbles don’t pop when sentiment looks like this, when everyone is so polarized and everything is so divided.

My question to you guys is: has anything changed in terms of how you’ve positioned? Are you looking to potentially add risk here after the recent sell-off and pullbacks? What does all this mean to you? Does it provide you with signals, or are you still a little cautious? Any reads?

Yeah, I think it’s a pretty good time to be long. The biggest concerns were on the macro side: war, oil, and inflation. You’re already seeing oil fall off a cliff, and that will trickle down to various inflation inputs. I think it creates a setup for a more accommodative Fed.

A lot of that stuff operates on a lag, but I think this current set of Fed governors isn’t as incredibly laggy as the previous ones, who knowingly used stale data but still anchored to it. I could see a situation where there’s more accommodation. The job market came in a little soft, too, so I do see a world where you get a more accommodative Fed.

If you think about what a blue Senate and Congress can do, they can’t unilaterally block wars, but they can block funding for them. That makes it a lot harder for them to proceed, and I think that’s why oil is doing what it’s doing. You assume we’re at peak Iran war at this point, and I think the administration is probably going to begin responding accordingly, thinking about how to plan for 2028 and do what needs to be done to set themselves up.

The war is unpopular, and its knock-on effects on oil and inflation matter. A couple of years ago, we had incredibly low oil prices before this whole thing started, and people want to get back to that. I think the inflationary pressures and impacts on global supply and movement in those areas should ease. I think all those things are tailwinds.

Combined with the factors you mentioned, this doesn’t feel like the end of the trade or some massive blow-off top. There was a big deleveraging event, followed by a bounce to a very historically supportive technical level on the overall index. It seems like a good time to be risk-on.

If you think about what markets like and don’t like, it’s uncertainty. To some extent, gridlock creates less uncertainty because it’s harder to do things when the two sides are opposed. You assume things stay status quo. There will be individual headwinds to build-outs for data centers in certain areas, but I think the trend should generally continue because no party can impede the progress in any serious way.

I’m very constructive here, but I think it does become a bit of a stock picker’s market because the impacts are pretty unique in terms of what they’re able to do. Whether they’re able to build out or block data center build-outs, whether there are regulatory headwinds—those things matter. I think it leads to a bit of a stock picker’s market, but generally it’s a pretty constructive environment, and I think the Fed should be more accommodative next year.

Yeah, I agree with that. Specifically for crypto, as we’ve been recording this, there’s a tweet about how the Dems are going to go after Trump and his crypto ventures within the first 100 days. I think that’s fine for crypto. Enforcement against World Liberty and all of Trump’s coins isn’t really a big deal to me.

Bullish.

Yeah. One wild card here is obviously the Dems are going to impeach Trump, but it feels like there’s a chance it could go through because the Republican Party now—listen, I’m not American, so I’m not as in-depth on this. It’d be good to get your guys’ opinions. I’ve seen this take a few times now: to repair their party, they might actually want to go through with the impeachment. You need enough Republicans to get him impeached, right? That would be the one market risk, and I don’t know which way it would go. Other than that, there’s probably not much.

I don’t think they will. They need 67 votes in the Senate. They’re never going to get that.

How many do they have in the Senate? 50—

It’d be—

The Dems?

I think, yeah, maybe 52, 54, something like that. I’m trying to—

So they need, like, 13 Republicans?

…54. They’re not going to get it.

Is it an anonymous vote, or no?

No.

Yeah, that’s tough.

No, they’re not going to get it.

They don’t toss out an anonymous poll in a group chat.

Yeah.

These are won and lost on the fringes. You have a very dedicated 30-something-percent base on either side, and a lot of people still like Trump. I think you can straddle the fence by moving away from him toward someone who implements the policies his general base liked, while moving away from some of the stuff he doesn’t like. Proceeding that way is the best way to galvanize the largest portion of the base, rather than completely alienating him, because I think that’ll just rub people the wrong way. Their best bet is basically to let him gradually fade out of relevance.

Yeah. I wonder if he’ll just play golf for the next 2 years. I wonder what the prediction market is. Did Rubio do really well? You have to feel this is like 9/11 for Vance, right? This is probably good for Rubio.

Vance has historically positioned himself as not super pro-war. He’s also been pretty anti-Israel, which Trump obviously has not been. I think he’s been subtly positioning himself that way to run as if this wasn’t his doing and wasn’t something he was for.

Okay, but he’s been getting crushed in the prediction markets. I just looked. I’ll throw it in the group chat. DeSantis has actually been coming up.

Yeah, I like DeSantis.

Rubio went up a bit, but DeSantis got the big pump. He went up like 15% today, from basically zero. Rubio didn’t really move; he’s pretty much flat, and DeSantis just got a chunk of the JD. So, yeah, it’ll be interesting to see.

I don’t know. I think 2 years is a long time, right? A year, a year and a half ago, Republicans won everything.

And isn’t it? I remember when we took office, it was like, “We have 2 years until midterms. We have so much time.”

No, it is. It is, right? Because 18 months ago, Republicans won everything by, like, the biggest sweep across the board in modern history. Then, 18 months later, you make one bad political move and now everybody fucking hates you, and you get this massive pushback in the other direction. I think 2 years is a long time. A lot can happen in 2 years. You could either dig the grave further or make it competitive again in 2028.

You could also make the argument that the Dems winning here is actually worse for their chances in 2028 because—

Well, it gives them time to mess up.

Yeah, it gives them time to mess up, but it also kind of normalizes the crazies. The crazies are never going to be able to get enough votes to win a national election, right? AOC will not win a national election, but she can get a decent amount of that base and win a primary. If you have more of these guys in office now among the Democrats, they can begin to rationalize, “Maybe we can run AOC on a national scale. Look how well we did in these pockets,” some of which are swing states.

I think she’s pretty unelectable unless things really hit the fan and we get a full-on K-shaped economy/socialist movement thing where there’s enough support for her to actually do well. Their best bet was running a moderate who can win over some centrist voters who went right because the left went so far left. But if they run an extreme leftist, it becomes harder to win a national election. You could argue that winning here gives them false confidence to run a crazy in 2028 and makes it almost impossible to win, versus the right running someone more moderate.

I mean, what even is it at this point? If you’re rational—if you’re Trump and you’re actually trying to fix things…

So not that you have a chance to do something in 2028 or whatever. What do you actually even do at this point?

Golf.

Well, I think—

You kind of went all in, right, with Iran? So—

Yeah.

Most people would probably be like, “All right, let’s get out.”

And endorsing Netanyahu for the election as well. He just did this full-throated endorsement of Netanyahu for the Israeli election. It’s like, bro, why?

It’s just, I mean, I think it’s pretty clear, right? You probably have to figure out Iran quickly and figure out oil prices and just—

Well, you have to go back to what he did well, right? He fixed the border. That was really good. He reversed mass immigration.

That’s what I’m saying. If you go back to 8 months ago—

He made tax cuts permanent. That was massive.

Before this, I would have guessed that Republicans would have won even despite the tariff stuff. I do think a lot of people who voted for them were like, “All right, whatever, maybe tariffs, meh, but at least they’re doing the things we voted for.” And then you get this massive 180, 8 months before the midterms, with this Iran war, and you’re just like, all right—

Yeah.

Eight, 12, or 18 months is a long time in politics. There’s a lot that can happen, so I’m not looking too far ahead at 2028 at this point. I just think it’s kind of a moot point.

Yeah. Again, I think it’s anchoring to the wins: shutting down USAID, drug prices at record lows in more than 60 years, DEI is out, Trump Accounts are great, China pushed out of the Western Hemisphere, Maduro’s out. A lot of the stuff went really well.

I think the biggest thing was the fact that the Iran war lasted so long. If he went in there, we set their nuclear program back 2 or 3 years, and it took 2 months, and then you kind of pulled out, it would be much more palatable. But I think it was the fact that it just dragged on.

There was the opportunity when they dropped all those bombers. I feel like that was the time to just—

Yeah.

…take the dub.

So, yeah, that’s why I’m saying the dude cares about what people think about him more than anybody else I’ve ever seen in my life. Maybe it’s enough for him to be like, “All right, we got absolutely clapped. Maybe it’s time to save face a little bit,” but maybe not. I don’t know. It would be great for markets.

Yeah.

Markets would absolutely rip.

Do you think Trump’s still as fully there as he was 2 years ago? Mentally?

No, probably not, dude. You’re like 80 years old, man.

There’ve been reports that he doesn’t really care—

Just listening to some of his recent ramblings and shit, about people’s asses and all this shit. I listened to a 5-minute one the other day. It was hilarious, but it was reminiscent of Biden to some extent.

Yeah. I don’t think he’s full Roomba—

Different vibe.

I don’t think he’s a full Roomba vacuum, but you’re 80-plus years old. Of course you’re not—

It seems like regardless—

You guys are going to get a younger candidate, nobody super old—

I mean, it’s impossible. There’s no one old. I think you’re going to get young candidates on both sides.

I don’t think we should get older.

Yeah, no, that’s what I mean. Both sides are going to be—

Because he doesn’t seem all there, is kind of my point, again, from the recent chat.

I think he’s there enough.

Yeah.

Fueled by McDonald’s and rage.

He wasn’t super all there when he ran, but he’s definitely way sharper. I don’t know. He just talks crazy shit now, man. It’s like—

Yeah, but—

He’ll do a rally, and he’s talking about refurbishing the ballroom and what he’s done to the White House, then he goes into some guy’s ass. Then it’s like—

The guy’s chatting for hours and hours on end. You can find clips of looniness in this. He’s not teleprompting. He’s just—

He’s sharp as a tack.

If you go off the dome for 3 hours and you’re him, you’re going to say some ridiculous stuff.

Yeah, he was complaining about the kickoffs in the NFL. “It’s a terrible thing, what they did to the kickoffs. They ruined the game.”

It’s not wrong.

I don’t even watch. I wait until they’ve done the kickoff; I don’t turn on the game until then.

And it’s a really good time to bring in Omnia, our special guest today, co-founder of Kinetic. So, this Binance hack happens on Thursday, right? We haven’t even talked about it yet. It happens late Thursday. I can’t remember a day when sentiment changed such a 180, from the morning session being almost borderline euphoric with the ICE–Trade.xyz hype partnership announcement. Then you had Druckenmiller coming in and tweeting about it, and then you had Elon comment on top, right, and almost play up the hype.

Then you had Hivemind come out.

Then you had Hivemind come out and talk all about this, totally pre-Bybit hack, right? This happened last Thursday, and then you had the whole Binance hack. Let’s talk about Binance first. It’s a little bit of old news at this point, honestly, because of how fast crypto and sentiment cycles work. Any takeaways from you guys from that, in terms of longstanding impacts?

Right.

Well—

My takeaway is that we’re still just incredibly incompetent as an industry. When the hack happened, everyone was like, “Oh, AI hacking.” And it’s like, okay, it probably wasn’t an AI thing. What was it? It was another social engineering attack. Yes, they’re very sophisticated, don’t get me wrong, but why does an engineer have his signing keys on the same laptop he does all his work on? That’s for $6 billion, right? Come on, guys. It’s just bad security practices. That’s—

Did you see CZ’s tweet about banks? Like, if they could hack the blockchains—

Yeah. He’s saying this could happen to banks too, and it’s a wake-up call to everybody. Banks don’t have a dude downloading Homebrew and falling for fake interviews on a computer that moves all the money, right? Don’t get me wrong, man, it was a sophisticated attack. A sophisticated attack like this would get me, okay? But if you’re a centralized exchange custodying all this money, it’s just bad practice. So we really need to be better about this.

One thing I found kind of interesting is that this is the first time the FBI straight-up called out THORChain. I would love to have balls as big as THORChain node operators, honestly. After the FBI calls you out, it kind of seems like a ticking time bomb. I just don’t know what’s going to happen there. Especially if there’s a regime change in 2 years, I could see them being a pretty quick target.

There was also this Matt Levine column, and it was funny. They interviewed some trader who talked about how they made a lot of money taking the other side of North Korea’s trades. In finance, normally you don’t brag about taking the other side of North Korean trades if you’re a financial institution, right? In crypto, it’s just the norm: “Oh, we made a 15% spread because we were selling ETH to North Korea at $2,650 instead of $2,300.” I found that kind of interesting.

The second-order effects of that are, like—

Yeah.

…you’ll get investment in crypto for doing that. Like, “Wow, let’s actually put money into this guy.”

Yeah. I don’t know. I think there’s not much to take away from the hack other than phishing is getting very, very sophisticated. People are like, “Oh, this means you’ve got to take all your money off centralized exchanges,” when a week ago everyone was like, “Ledger, Ledger hack, Trezor thing. Keystone had a leak. You’ve got to put all your money on centralized exchanges.” There really aren’t a ton of good options. It’s at the point now where you can’t really give anybody custody advice. I don’t know, self-custody kind of just sucks, right? It’s never really gotten better over 10 or 15 years.

So I think you’ll see more and more money going to exchanges and custodians. It’s important that these exchanges and custodians have better security practices so they’re not falling for stuff like this.

Yeah, I mean, I hate self-custody.

Yeah, nobody—

I’ve never—

I’m too old for self-custody at this point, man.

Yeah.

It’s like, you know, there’s—

I mean, I get it. It’s great, but for the majority of people, it’s not the solution. My takeaway from the Binance hack is that it completely overshadowed the great news for HYPE, perps in the U.S. in general, and TradFi adopting crypto rails in a way that I think a lot of us have been hoping for for many years. Is HYPE trading at where it was before this announcement—

Yeah, no.

—at this point?

I think so.

I take the other side of that.

You don’t think so?

No. I think it’ll be forgotten fairly quickly because it was handled internally, so there’s no reliance on external funding. Binance is returning everyone’s tokens, and the bullish part of the announcement was bringing everything on-chain to HYPE, to LIT, and to other stuff. Everything rallied on that, and this, if anything, was a further boost toward dealing with non-custodial venues.

I mean, it’s bullish for that, right? Yeah.

Yeah.

I don’t know about that, because a non-custodial venue can’t make you whole for $6 billion. There are only a few—

Right, but non-custodial venues—

For me, it’s like, I want—

—allow you to—

Yeah.

—to have control over your assets. If you’re an institution, you can do this in a way where you can ensure this doesn’t happen.

I mean, it assumes that your security is better than these big centralized venues. I don’t know. I prefer—

Yeah, but when you hear—

Like—

I hear you, but when you hear how they were operating, at first you’re like, “Yeah, there’s no way. These guys have been in the game for a decade, and surely they have best practices.” Then you realize they don’t, so the answer is that you do have better security practices than them.

It was shocking, yeah. Everybody had this conspiracy theory that they were trying to liquidate their users to make themselves whole. Honestly, I think that would’ve been less embarrassing than the way they got hacked, given the size they are. They should’ve just said that’s actually what they were doing.

Yeah. “We didn’t want HYPE to go to $100.”

But the thing with—

“We tried to liquidate you guys.”

That would’ve been so much worse, though.

The only thing I’ll say that’s bullish for DEXes is that when you deposit your capital into a DEX, you have smart contract risk. It’s not purely self-custody at that point, right? You could have a bunch of money in Hyperliquid, and if there’s a bug in Hyperliquid, all the USDC could get drained. Hopefully Circle would freeze it quickly. We didn’t know what the number was, but we knew they had a number where they would freeze, and apparently $800 million is above the number.

That is the number.

If it happened to Hyperliquid, hopefully that would be frozen extremely quickly, too. We’re learning a lot of things from TradFi that we want, right? We want some delays—

Don’t say T+2.

—we want T+something, and—

Yeah, I’d love if there were at least some rate limiting on the bridges. This seems like the most obvious thing that everyone should implement. I don’t know if Hyperliquid has this. I don’t think they do, but I’d be—

No.

—way more comfortable if they had some rate limiting on the bridges.

There isn’t.

Yeah.

The thing is, you’ve got to be lax on this kind of stuff when you’re small and growing, to be competitive. But when you get to the size Hyperliquid is now, maybe you should add some of these security features that you didn’t have when you were trying to get your original users, right? I’m sure they’re thinking about this a lot. Hopefully we can—

It would be very ironic for me to be worried about the Hyperliquid bridge, which is why I didn’t buy it for years, and then finally buy it at $80 only for this to happen. Hopefully that’s not what ends up happening.

Yeah, HyperEVM replaced the Arbitrum bridge, right? I guess that all makes sense, finally.

Yeah. Maybe that’s a good lead-in to Elysium. Hyperliquid, a perp exchange, announced HyperEVM, I want to say two years ago at this point. It was two years ago that they announced it. It hasn’t even—

It was like two and a half.

Yeah. After their TGE, they had announcements for it.

I remember writing about it for our year-ahead report at the end of ’24, going into ’25. I remember thinking this was kind of a weird chain. It had no throughput, and it was a second-class citizen to the Hyperliquid blocks. It was confusing to see exactly how it would take off. I think the answer to that question is that it hasn’t really, because it’s a tough environment.

We’ve talked about this before: ecosystems that have had a big wealth effect have opportunities for people to do other stuff. You guys started as a liquid staking protocol. When did it become clear to you that this was a high-EV thing to put time into—creating the L2?

So it’s been clear for a long time.

When we actually decided to do something about it, it was at a critical point where people were essentially giving up on building on Hyperliquid. Like you said, two and a half years ago, when this was announced, it was supposed to be a new paradigm. You finally end the monopoly—or duopoly, if you want to call it—of centralized exchanges extracting more money by deploying an L2, because people are naturally going to be on-chain. Then you print all the sequencer fees and have fun. Examples of this are Binance and BSC, Coinbase and Base, and so on.

That’s what excited us about it. The specific development that got us excited about building Kinetiq was Hyperliquid announcing HyperEVM shortly after announcing HyperBFT, the consensus mechanism it uses today. By the end of that summer, we had decided to build Kinetiq. From the very beginning, we knew the liquid staking token would be a wedge into doing much greater things for the ecosystem. That was the whole purpose, and that mission was accomplished.

We always knew liquid staking tokens are very uninvestable, and the businesses themselves have historically not been very good businesses. So how do we take the liquid staking business and turn it on its head? That’s exactly what we did. Then we wanted to go beyond that and build something much more interesting, because Hyperliquid was the only thing that interested us at the time. That’s what Elysium has become. We planned to do a lot more for Hyperliquid, and this is the delta we’ve gotten to.

HyperEVM’s failure to capture all the demand for an ecosystem like this is what set the stage for Elysium. I don’t think that was on purpose by the Hyperliquid team, obviously. They’re a small team, famously—or infamously—of 13 to 15 people. It doesn’t make logistical sense for Hyperliquid to work on HyperEVM. They have an exchange making between $1 million and $7 million a day. Why would you move a head over and give up $90-plus million in ARR per head to beef up HyperEVM? It doesn’t make much sense for a team of their caliber to focus on.

But that doesn’t mean it should come at the expense of Hyperliquid’s general-purpose future, its possibilities, or the possibility of it becoming a dominant general-purpose network for spot trading of any assets, whether they’re shitters or majors. That set the stage for Elysium a couple of quarters ago, when we decided to do this.

The first questions, as you can imagine, were, “What the fuck do we need an L2 for? What do we need another chain for?” That’s exactly what we asked ourselves at first when we were thinking about it. The table stakes were that it couldn’t be extractive to Hyperliquid in the slightest.

If you think about it, you don’t even need to think much about this; it’s obvious. HyperEVM contributes maybe 0.00001% to Hyperliquid’s valuation. No institutional investor is sitting there saying, “HyperEVM is exactly why we bought HYPE.” You’re buying an exchange that makes over $1 billion a year and plows 99% of its revenue into the token. It’s very simple.

Hyperliquid’s economic model and monetary soundness, if you will, come from the exchange, not from HyperEVM in the slightest. Does HyperEVM contribute some functionality, like we talked about earlier, by replacing the Arbitrum bridge? Yes. That’s exactly what the purpose was. The Arbitrum bridge is bad. HyperEVM, your own general-purpose network, is very good.

An example is Circle minting USDC natively on Hyperliquid. It’s not done on Hypercore, which is the exchange; it’s done on HyperEVM because it’s a very simple environment for them to use. It’s an EVM, so it’s very simple.

With Elysium, it’s the first time in history that an L2 will be accretive to the network it’s built on top of. We don’t eat into Hyperliquid’s core business in the slightest. There are no perps and no order books. This is what Hyperliquid does best, and this is where it makes its money.

What we will do is usher in a new era of what’s possible for general-purpose composability across any blockchain. In short, that means putting an extremely performant general-purpose blockchain on top of an extremely performant set of on-chain order books with the deepest liquidity anywhere on-chain. Now you get to do really cool stuff: fully on-chain options, prop AMMs that can hedge directly into Hyperliquid’s liquidity from Elysium, and arbitrage between Elysium—which is within Hyperliquid’s orbit—and the exchange itself.

There’s a lot of cool stuff that BSC or Base should have been to Binance or Coinbase, in our view. We get to do this symbiotically, where Kinetiq itself is not the only party accruing all the value, as is the case with any other L2 today. Hyperliquid itself accrues value in ways that it doesn’t make sense for Kinetiq or Elysium to compete with.

For the sequencer revenue, is it base fees only, or is it MEV and priority fees too?

Yeah. Since it’s built on the Arbitrum stack, we’ll be using PGA, so the priority gas auction—

Okay.

—that will contribute to it as well.

Okay. And then all the DEXs on Elysium can tap into the spot liquidity on Hyperliquid?

Correct.

Or do you—

Yeah, Aerodrome also uses this a lot.

Yeah.

You’re going to want to have LPs for sure. A good example is Bitcoin: it’s very liquid on Hyperliquid. I think it’s apparently the most liquid venue for Bitcoin on-chain, but that liquidity isn’t being utilized. Last month, Hyperliquid hit a 15-month low against Binance in weekly spot volume. Spot itself on Hyperliquid is struggling a lot, and there are a lot more people in crypto who trade spot than trade perps.

This is really about complementing Hyperliquid’s offering and mostly expanding the TAM. You’re going to have a much easier time getting people to trade on aggregators, which need AMM liquidity, than teaching people how to use order books when they don’t really care. So Elysium is very necessary in this sense.

One could also argue that HyperEVM fulfills this need: you can deploy Uniswap, Aerodrome, and whoever else there, and you’d be right. But the problem is exactly what you touched on earlier: it needs a lot of TLC, which I don’t think is technically possible because HyperEVM is built to fit within Hypercore’s block times. Therefore, it has this weird dual-block architecture. It’s at about 7 TPS, so 3,000 or 4,000 people hitting the chain at any time can send gas fees to between $30 and $100.

During the Elon tweet, I got a really crazy, hilarious snapshot of a swap costing $100 on HyperEVM. It took me right back to TempleDAO in 2021, when I paid $1,000 for an approval and immediately got extremely pissed. Imagine people wanting to do anything you see happening on Solana, Base, or Robinhood on HyperEVM. It’s just not possible. Elysium changes that completely.

What do you see having the most activity at the start on Elysium?

It’s a good question. I think it’ll be a mix of things. Naturally, people want to gamble, so quite a lot of launchpads are getting ready to go. Uniswap will be there on day one, as will Aerodrome, FOMO, Pump.fun, GMGN, Axiom, and all the big terminals. They’ll support it starting next Monday, when we go live.

Naturally, you’ll have a lot of really interesting spot activity. There’s also other really cool stuff being built, like perpetual options from a project called Bound. Risk will be coming over as well, offering covered calls and cash-secured puts, so it’ll be a really cool place for options. As you just showed, OpenSea is coming over too, which is interesting. All the aggregators—0x, Matcha, Kyber, and others—will be there.

I think what’s misunderstood about Elysium is that there are a lot of animal spirits that have been waiting to be unleashed on Hyperliquid in ways beyond perps. People see what’s been done with Hyperliquid perps: just a casual $140 million oil long yesterday. People on Hyperliquid are very well-capitalized, whether they’re institutions or retail. They come from centralized exchanges.

But there hasn’t been a platform where sizable spot traders could come in and have a good time. Elysium will help them do that, and it’ll also be a place where sophisticated traders can generate yield or run strategies. Prop AMMs are kind of perfect for them because you don’t need to split your account between a centralized exchange and Elysium, for example. You can just go and hedge directly on Hypercore or arbitrage prices that way.

No, this is music to my ears, because it’s the thesis I’ve had for a HyperEVM season at some point. I never knew what it would look like. A year ago, you weren’t really sure, but you could see the wealth effect that’s happened on every main L1. You had the ETH wealth effect; all that money flows through. You have meme season, you have NFT season. People get super rich on ETH, and they kind of become ETH maxis to a degree.

It’s the same with Sol. Sol had a massive wealth effect. You had some massive airdrops like Jito and meme season on Sol. People who made it on Sol kind of want to be Sol maxis. “Patrons of the chain” is kind of what we wrote about in our year-ahead reports.

When I was looking at HYPE, I thought, “Oh, well, obviously this is going to happen on HYPE.” But there was just no way for it to actually happen until somebody tried to do something, which seems to be what you guys are doing.

Yeah.

I guess one question I have is: You clearly have this alignment ethos, with a lot of fees going to buy back HYPE and Kinetiq tokens. Obviously, you said you want everything you do to be accretive to HYPE rather than extractive, like we’ve seen elsewhere.

Do you think that’s something other teams will be forced to do when launching on Elysium, as table stakes, given that it’s kind of being pushed indirectly by Jeff? They lay the blueprint with everything. If you guys do it, and teams launching on day one do it too, does that socially force other teams to adopt a similar kind of alignment? Is that your hope?

You brought up an interesting point. We actually buy no HYPE. Our revenue is generated in HYPE: from the LSTs, and from Elysium, because gas is paid in HYPE.

What we end up doing is using 50% of the sequencer fee to programmatically buy KNTQ, just the same way the Assistance Fund buys HYPE with revenue. The other 25% funds our treasury. We’ve famously—or infamously, however you look at it—raised very little money for Kinetiq: less than $2 million in a seed round at the end of 2024, on purpose, to make the protocol investable.

Historically, over the last five or six years, all liquid staking tokens have been completely uninvestable because of investor overhang and whatnot. We purposely went against that grain. The only place it made sense to do something like that was Hyperliquid, given the no-investor ethos, which we didn’t fully buy into but also didn’t fully go against, if that makes sense, by taking tons of investor capital.

So, just exactly what was needed was to bootstrap the protocol and then go from there. 25% goes to the treasury for operations, and the remaining 25% goes to builders, earned based on consumption of block space. There will be criteria, right? Nobody could just sit there and say, “Jason builds a contract, spams the shit out of Elysium, and prints HYPE.” That’s not going to be what happens. There will be ways to verify that teams are doing this and make sure there are real users, et cetera. That’s how we go about further distributing the value generated at the sequencer level.

We generate HYPE from this activity rather than buying it and then distributing it, if that makes sense. So this goes into KNTQ and so on. But you made a really interesting point about the wealth effect. This is what we expected as well, right? Like you said, you saw this on Ethereum a few years back. You saw this most recently on Solana, and it’s yet to happen on Hyperliquid. It’s not that it just won’t happen; it has yet to happen. There’s, what is it, $20–$40 billion of capital basically sitting on Hypercore, staked or unstaked, waiting to do something, and Elysium does change that.

I’ll give you an example. Projects that have launched on Hyperliquid have historically been put in a really bad spot because if you’re conducting an airdrop, for example, you’re giving free money to perp traders, and I think you can do the very basic math as to how that works, right? Most perp traders lose money. If you’re giving them free money, they’re going to top up margin. It’s that simple. So it has historically been extremely negative EV for teams to launch on Hyperliquid and do a TGE on Hyperliquid.

But again, we didn’t really care. We TGE’d a month after 10/10, which is the date we kind of committed to, and we didn’t care. We weren’t trying to time the market. So you still see a lot of this happening in crypto, I think. But for teams that build on Elysium, the prize is very different than elsewhere, right? Ethereum money is not really moving. I don’t particularly see Solana money moving outside of the shrinking pool of PVP capital there or on Robinhood Chain.

But on Hyperliquid, you have a really interesting prize if you actually build something interesting: you get tens of billions of dollars that may or may not be willing to move over, accelerate product usage, and become your customers. Whether it’s retail or institutional, you have this very concentrated presence of sophisticated users on Hyperliquid who will move over if it’s something interesting enough. That was another really interesting part of the Elysium thesis: maybe the first users themselves aren’t perp traders, which is totally fine. It’s okay as long as we capture a sizable percentage of the people who are on-chain across other ecosystems, from Base to Robinhood to Solana, et cetera. If something really interesting does happen, there’s a dam of tens of billions of dollars in capital that can come over, move, and have some fun.

Yeah. Well, you guys are launching November 9, so 5 days from today, 4 days from when this podcast goes out. You guys have a lot of good momentum, so good luck with all of that. I know, disclosure, I own Kinetiq. I know some other people do too. And it seems like you guys have gotten a lot done in the last month or two, especially in the last few weeks. All the partners—it’s been pretty impressive, the launch lineup you have. So you guys are definitely getting a good start off the ground floor.