交易原则、Friend.tech 的「完美用例」与比特币抛售潮 | 1000x

I think that if you're not paying attention, you're going to be missing out on a big trap. This is one of those things that goes mainstream, and ultimately I think it's genius. I'm refreshed by it. This is a perfect use case for crypto.

Finally, we get some volatility. It's been boring as hell this entire summer, effectively since we hit $32,000. We finally got some volatility. Bitcoin traded down, ending up breaking down from that $30,000 range to $26,000, and now we're teetering here on the edge, with a lot of arguments to be bullish and a lot of arguments to be bearish. But Jonah, where's your head?

We were literally just trying to come up with ways to stay awake this summer, and then bang, a 15% sell-off. When I look at price action like this, it makes me nervous, and I think it could go a lot lower. Just as a total finger-in-the-air view, watching it nervously stabilize around $26,000. But if you pull up the open-interest levels and look at how much length was liquidated over the course of this sell-off, the picture looks really clean.

Despite being a crypto bull and feeling nervous about longs right now, I think this is a time to buy with both hands. If you have dry powder—and hopefully you do—you're supposed to be adding meaningfully here. I think what happened is that a lot of participants got long, or even levered long, ahead of the anticipated GBTC-versus-SEC decision last week. The market was expecting the judge to come down in favor of GBTC, or Grayscale, rather, and then nothing happened. Maybe there was a bit of cascaded liquidation as people exited positions, so I think all of that weak length is gone. I think it's a clean slate that you're supposed to buy.

I'd agree with that. The only reason why I'd be slightly more hesitant is if I believed that the equity markets were in trouble. I think September is going to be macro-podcast season, unfortunately. I really enjoyed my break from not listening to any macro podcasts and not really paying attention.

Discount is a macro podcast.

No, because we're not opining on macro the entire time. We're talking about something else. I do think that listening to an hour-long podcast on inflation is very tedious, in my opinion, because people just tend to repeat themselves. Once you've heard the same argument once, you hear it again.

But I do think it's important to start paying attention again. I think equities provided you with a period of time to flip from bull to bear. If you look at Bitcoin and then overlay Coinbase and MSTR, you'll see that both Coinbase and MSTR were nuked before Bitcoin was nuked. When I say “nuked,” I don't mean they were down 3%; I mean they genuinely nuked hard, 10% to 15%, while Bitcoin was ranging.

That gave you some indication as to where the interest was in the market, and the interest was not in crypto. There's a lot of alpha in that statement.

I don't trade like that. I try to trade around a core position. I think that anybody out there who is in crypto is probably a bit bullish on crypto, and anybody who had the guts or the courage to get tactically short in this summer-doldrums period, playing for this dip, is truly lucky or truly talented.

But what you bring up is interesting. I disagree with that. I think it was prudent once you got that move up to $30,000. We were ranging for a long time, but once you got that move up to $30,000 and rejected, and then equities started nuking while Bitcoin was still above $29,000, that was a chance to at least protect yourself.

That was a chance to say, “Okay, this was two weeks ago when we recorded the pod. We were trading up, we tagged $30,000, I said I was bullish, and then we rejected. We traded basically 4 or 5 days sideways, and during that 4- or 5-day sideways period, Coinbase was nuking, MSTR was nuking, and all these other things were nuking hard.”

I view those as indications of institutional interest in the asset class. What you were holding on to was the idea that there wasn't enough buying power to get us above $30,000 right now. We got rejected, and there seemed to be institutions and other holders exiting through Coinbase and MSTR pretty aggressively. Given that you had a pretty reasonable stop-out above, call it $30,200 or $30,400, it made sense to start protecting yourself.

That's also one reason why podcasting is important for all the listeners out there. It's very important to take into account why somebody is saying the thing that they're saying. What I try to do is lay it out very clearly, so that when you hear my viewpoint, you understand why it might change and can factor that in.

We only do this once every 2 weeks, and obviously, as an investor or a trader, sometimes my bull and bear takes will change day to day depending on new information that I get, and that Jonah gets. You just have to update your information.

I will say that once we didn't get above $30,000 and equities started nuking, if you're trying to directionally trade Bitcoin over a 2-week or 3-week time period, make sure you have Coinbase, MSTR, Bitcoin miners, and the other crypto equities on your watch list. I think they're actually reasonably predictive in certain scenarios.

If Bitcoin is ranging sideways and it's a low-volume time period, you can catch those inflections of demand and supply through those pieces of equity. That's just something that I find particularly valuable.

Do you find it valuable when Bitcoin, the miner stocks, Coinbase, and MSTR are all trending in the same direction? Do you find that the equities are useful then, or is it only when BTC is range-bound and the equities are basically trying to pick a direction?

Bitcoin range-bound. I'm trying to pick a direction.

Okay, so this is our equities-stronger-or-weaker framework. This is critical. You listen to podcasts all the time and get pulled or paired up on the basis of what the podcasters are saying. I think, Avi, what you're trying to mention here—and what you hear about on a trading floor all the time—is process.

What is your process for arriving at a trading or risk decision? Is the process consistent or improving over time?

Yeah. What I hate is when people say, “You can use equities to trade Bitcoin.” Great—you can use equities to trade Bitcoin in certain situations. They're sometimes predictive. Okay, when? How? What are you going to do if MSTR is up 5% today? What does that tell you?

It doesn't tell you anything in context unless you put it in context. The only place I've found it valuable is when Bitcoin is going sideways and needs to pick a direction, and you're trying to figure out what direction it's going to pick. That's the only scenario in which I've actually found Bitcoin equities, or crypto equities, useful in terms of being predictive for Bitcoin.

That's interesting. I hadn't thought of that. What I was thinking when I was watching miners and Coinbase during this sort of slower summer for crypto was, okay, crypto's not moving, but the miners and Coinbase have just gassed higher to local highs. All right, now they're retracing a little bit from local highs, whatever. They were decorrelated from Bitcoin on the way up. Maybe they'll be decorrelated on the way down.

They weren't decorrelated. I mean, they actually—one thing that I will say, I'm just going to pull up the chart. We're not trying to take it tick for tick, though, dude. Bitcoin literally didn't move for a few weeks, pretty darn close. I mean, they topped on a very similar day.

Yeah, it wasn't exactly on the day that Bitcoin traded $32,000, but they basically topped on the day that Bitcoin traded $32,000.

When you have all the things you look at in the market, what I try to do a lot is place them into very specific contexts and then start to overlay. For example, you can look at breaking the 200-day moving average. What's more important is whether breaking the 200-day moving average is more predictive of momentum to the upside or momentum to the downside.

It's actually a very easy problem. I'll leave it to the listener to decode. These are the types of things where you section off an indicator and ask, “In what context does this thing work, and in what context does it not work?” Then you start to piece things together.

I actually have a Notion with all these different rules for the things I look at, where I see them tend to work, and where I see them tend to fail. I'll review them, take a look, and try to apply them to the current market. I'll ask, “Do any of these apply right now? What's important?”

That's interesting. I have a slightly different process, honed from TradFi. Basically, the process that you just described—or any process, for the listeners—is something you can cling to when you're losing money. It's an objective framework that helps keep you from stopping out of positions when they're temporarily not going your way.

That's why you need to have a process that's somewhat static over time. Otherwise, you're just going to be buying on the way up and selling on the way down and chopping yourself up. I did that for a period of time in my career, when I made the excruciating transition from being a market maker working at a bank to an Avi Felman-style prop trader working on the buy side.

I had to unlearn a lot of bad habits and stop myself from day trading without a framework. I was just blowing in the wind. I screwed up quite badly and almost got myself fired a couple of times. On a last-ditch effort, I had to develop a process.

For me, I like to take a 6-month to 2-year view on my asset class and find ways to hang on to that, because I'm usually pretty right over the long run. Over the short run, on the time frames you're talking about, Avi, I'm pretty weak. I'm not a very good trader.

So I basically put a yellow sticky on my monitor that said, “Do not trade. Do not day trade unless you have backtested your strategy.” I had that on my monitor for years.

I'll say that that's an important part. It's not just part of the process; it's asking, “Okay, you have these certain situations that pop up. Can you quantify them?”

There you go. What's your hit rate on these types of things? A lot of trading is just coming up with strategies that you sometimes observe organically. Then you go to your quant—if you have a quant—or maybe you are your own quant, and you say, “How do I put this into a framework that's repeatable?”

For example, is there alpha if you have 2 correlated assets, Arbitrum and Optimism? If Optimism is up 8%, or if the Arbitrum-Optimism pair is up a certain amount, will there be mean reversion at some point? How much does it have to blow out before mean reversion starts to kick in?

Maybe it's if it's up 8%, maybe it's 10%, maybe it's 15%. Once you get to 15% or 20%, maybe you're at a 3-standard-deviation move. If it's up 20% over the last 2 weeks, maybe mean reversion is probably going to kick in there. Then you start to quantify these things, tag them, and look for them.

I think there are a lot of idiosyncratic upside opportunities for DOGE and a lot of idiosyncratic downside opportunities for BCH. That's not a mean-reversion thing; that's just a secular trend. These are slightly different, but can you juice alpha in these types of situations?

You have the DOGE potentially dying, unfortunately, which is bullish for it. You have Elon launching a rocket with DOGE on it. You have Elon rebranding the accent on BCH. You have a completely manufactured pump that sent that to $240.

If you can find an efficient way to get short altcoins like BCH, there are a lot of those trades. There are a lot of longs, too, but Polkadot is another one that you can short. Good luck getting short Polkadot without waking up one morning, realizing that the perp blew out to negative 200% APY funding for a couple of cycles while you were asleep, and finding that it ate up all of your P&L. These markets are hugely inefficient.

One thing that I really like is that you get these scam pumps from things like BCH, Aptos, and Sui. They're often characterized by perp volumes blowing out, a random spot bid appearing from nowhere, and a tremendous amount of open interest. You'll get BCH going from 120 to 240, for example, and think, "How do you think about shorting that, and where do you close it?"

What I tend to do is wait for a period when it's down 20% or so from the highs. It bounces, then retraces back down to that 20% from the highs, and that's when you normally want to short it: when the bounce has failed and the scam pump is down 10% or 20%. That's somewhat arbitrary.

I see what you mean.

Here's the part that's not arbitrary: Where do you tend to close it? Sometimes you get caught in these situations where it starts going back up, so where it normally starts to go back up is the interesting point.

Let's use an example of an asset that goes from 100 to 200. It goes down to 180, then back up to 195, and then back down to 180. Great—that's your trigger point. You're going to short at 180. Where do you close? Do you close at 100? No, you don't, because the question is where you think fundamental value should ultimately be—not fundamental value, but where the risk-reward makes sense for other traders to come in and take the other side.

These markets, especially these types of markets, tend to trade very technically. Let's say the top of the pump is 200 and the bottom is 100. You've shorted at 180, and it trades to 150. At that point, you have to think about what other people are seeing in this chart and how they're seeing the trade. They see 150, and what they see is a stop-out below 100, where the scam pump began, and a target of 200, where the scam pump ended. The risk-reward is about equal, so there probably aren't that many people who are going to step in there.

At 140, it starts to become a little more interesting. People will look at a chart like that and say, "If I buy at 140, I can stop out at 100 and target 200." The risk-reward starts to look a little better. At 133, it's 2:1, and generally 2:1 is where I find people in crypto like to step in when they see those types of setups: risk 1 to make 2.

In TradFi, some of these things that pump like crazy actually have fundamentals. For example, the March-April natural gas spread: March is the last month of winter, when natural gas is in high demand in the United States, and April is sort of the first month of spring, when natural gas is abundant and can't really get squeezed by too much demand and not enough supply.

That March-April time spread is usually worth 30 cents per MMBtu at an extreme, and usually 20 cents. I remember that in 2018, during some crazy winter polar-vortex event in America, it pumped up to $2. That was the craziest thing anybody had ever seen. Billions of dollars of capital were lost on this, and the whole time the thing was worth 20 or 30 cents. But it went up to $2, which is crazy.

It feels a bit like crypto when it's trading up there. What is it? Why wouldn't it go to $10? It's like oil and gas during the Ukraine war. Could oil go to $500 a barrel if Russia started a war with the West? Sure. You don't know how high it's going to go, but you know that fundamental value is probably a lot lower than the current price.

When you say that something is "scam pumping" in crypto, theoretically your fair value for BCH is zero, right? What do you actually think that token is worth? No, before you answer, I take your point about people wanting to come in and get long because it just traded up. This is a little different from a commodity or commodity spread that has physical end users who will ultimately drive fair value toward fundamentals.

At the same time, when I think about these things, I usually look at them a little differently. It's not, "Oh, 20% off the highs," or, "Where would someone take the other side?" It's literally, "Go to Wikipedia, type in Kelly criterion, calculate how much money I can afford to lose before I get fired or get divorced, or whatever it is." How much money is that, Jonah?

No comment. But at the same time, it's usually just a function of your dry powder when you go into these technical, no-price-makes-sense markets. I use the Kelly criterion—half Kelly.

Yeah, I think that's fair. My long-term value for BCH is probably zero. That being said, the crypto markets are influenced, to put it in the nicest way possible, by a lot of factors that have nothing to do with fundamentals, and it's often easy to change the pricing of these assets in the same way.

Over the short run, that's true. Over the long run, the fundamentals always win. When you see evidence of something moving for inorganic reasons—and those reasons can often be a spot bid that has come out of nowhere, a very sharp price movement on an illiquid coin, or a large increase in open interest on futures contracts corresponding with a large move in price, normally more than 30% to 40%—and then you see the drivers of that move start to disappear, that's when you take this type of trade.

Maybe I was watching the Coinbase-Binance spread, and for the last week this asset had Coinbase above Binance while it was going up. Now it has stopped going up, and Binance is above Coinbase. Maybe that's your trigger. Or open interest went up a ton and the asset went up 40% or 50%; it's no longer going up, and open interest is starting to come down. Maybe that's your trigger.

Now, how do you short that thing? What I just talked about: Where do you cover?

I like that. What you just talked about is very concrete. It's a pioneering framework in crypto for what most people would call quantamental trading. The input would be the Coinbase-Binance spread for a certain token, or open interest. Enterprising listeners should be taking notes on everything Avi just listed and back-testing it.

What is the lead-lag relationship between the Coinbase-Binance spread for DOGE and the price of DOGE? Is there one? These are cool little quantamental factors that you could use to build a systematic strategy.

You can also pull all sorts of random things that affect the price of crypto. Time zone is huge, and time zone correlated with the price spread between exchanges is great. For example, when Binance is in the lead, you tend to get better price action during Asian hours. When Coinbase is in the lead, you tend to get better price action during U.S. hours.

You can start asking, "How do I take advantage of that?" It's not just Bitcoin, because Bitcoin is efficient. It's all these other coins out there. When does Litecoin perform best? Go figure that out. Now you have a strategy. Congratulations. Let's go make some money.

I think we should debate the Friend.tech app that just launched. I've read about it but haven't used it. You have actually gotten yourself on there, right?

Yes. Let me see. I guess maybe we should try to debate this, and I'll attempt to pick it apart. For the record, I think it's cool, and I think it's worth something, but I could pretend for a minute that I think it's not.

I get personally offended when people sell my shares. If you sell my shares, I will see it and be upset about it. I don't think I can threaten people on this podcast, so I will be very upset with you.

You're one of the most unthreatening podcasters out there, Avi. You've got no tattoos.

Are you kidding? I don't know if you know this, but I used to cage-fight.

Really?

Yeah. I'm not going to challenge you to a cage fight. I wrestled a little bit in high school and college.

Oh, really?

Yeah. I played cello.

Call me nonthreatening again.

Come on. Play cello and say it to my face, Jonah.

So, Friend.tech: Let's say that I am a novice, aspiring day trader. I read Crypto Twitter and follow a few influencers. I could theoretically buy those influencers' shares on Friend.tech and get invited into private conversations with those influencers, where I might theoretically be able to learn something about the market that could help me do a better job profiting from trading. Is that a correct interpretation of how Friend.tech might work?

Yeah, I think that's a reasonable interpretation. The way I would think about it is that the financialization of people is inevitable, given the technology that we currently have.

Why does this need to be a crypto thing? I'm playing devil's advocate here.

It's just the easiest way to transfer value on the internet. I think we both agree with that.

Yeah. If you're going to have value transfer on the internet, crypto is the best way to do it.

I agree that people will eventually be financialized, and I agree that crypto is the best way to transfer value on the internet. But there's a difference between financializing people and transferring that value around on the internet, and what Friend.tech does, which is basically selling access to those people in the form of a private messaging channel.

Don't you think that seems potentially dangerous and valueless? For example, if we were trying to have a private, invite-only, paid version of the 1000x podcast where we did what we don't do on this podcast—offer investment advice and give you tips or whatever—that's not something I believe retail traders would be able to profit from. I wouldn't feel good about doing that.

I don't think it works. I think everybody has to have their own investment process, to our earlier point. I don't think you can just follow somebody else's trades and expect that to work for you. You can invest in their fund, and that works sometimes. In the case of profitable funds, of course it works. But getting invited into a channel where someone tells you what to do—do you think that's actually beneficial to retail? Is that a real use case that works?

There are 2 things here. The first is whether it's beneficial to have monetary value assigned to access. I think, yes, 100%. That's how basically everything in the world works.

Can you give an example?

If you want to go to a club and you're a guy, you buy a table. If you want to have lunch with Warren Buffett, you pay $5 million. If you want to have a better chance of getting into an MBA program, you pay an MBA counselor $30,000 to help you with your application.

There are all sorts of examples in life where you assign value to access. There was that one Balaji Srinivasan company that he sold to Coinbase that was 21.

Yeah, so something—oh no, it started as 21. It started as the Bitcoin Computer, and then it switched to Earn.com or something like that.

Earn.com, exactly, where you would pay to get access to people. I view this as a natural extension of that. I think, Jonah, the main worry that I have is that somebody’s going to short me and then try to murder me or something like that. That’s really the fear. I’m being serious.

Well, you know how to wrestle. You could just put the guy in a full nelson.

Yeah, you can’t wrestle a bullet, Jonah.

Okay, all right. You know what? You’re convincing me that this Friend.tech thing works. I sort of believed in it from the beginning, but I wanted to try to play devil’s advocate. As everybody’s probably aware by now, I’m not very good at arguing for things I don’t really agree with. So, Avi, you win this one.

No, I didn’t even argue for it. I just said that it allows people to potentially get murdered for profit, which is actually an argument against it. We actually invested in a platform called Fan3, which allows artists like Zedd, for example, who’s part of the project, or Liam Payne—or Taylor Swift, to use a better example that everybody knows—to issue special-access NFTs and build a website that allows for special access very easily. It’s basically a suite of tools that has some Web3 components and some non-Web3 components, allowing the artist to engage more directly with their fans. The reason I said Zedd and Liam Payne is because they're both part of it.

People are happily paying for this, right? It’s one of those things where I think people are willing to pay for access to people they view as important, useful, good, or fun. People are willing to pay for access to people they idolize.

I think there’s also a gambling element to this. You’re not only paying for access; you can resell that access and potentially, if you get access to somebody early on, make a lot of money by betting that this person is going to be huge. I’m sure we’ve all come across somebody in our lives where we’re like, “I don’t know exactly how they’re going to be successful, but I know this person is going to be successful, and I want a piece of that.”

Well, frankly, you’re looking at that person right here, Avi. His name is Jonah Van Bourg, and he was one of the first 100 or 1,000 viewers of PSY’s “Gangnam Style,” right before it really blew up. Look at me now: I haven’t made a dime off that.

And so, 2 thoughts on your financialization of people and crypto giving you access to interesting events or conversations with those people. Thought number 1: You are the Taylor Swift of crypto traders, Avi. I just want you to know that you’re definitely at the top of your game. You’re young and have a bright future ahead of you.

Thought number 2: I agree with you. I think crypto is a fantastic and proven way for people to engage with content creators and for that value to be fungible and liquid. If you look at Friend.tech, like it or not, it’s the 3rd-biggest revenue generator on the Ethereum network right now. The top revenue generators are obviously Ethereum mainnet; number 2 is Lido, and Friend.tech is number 3 right now. People are spending real dollars on this—or ETH, really. So ultimately, you’re right, and it’s being proven in the pudding.

But to me, my only actual point of skepticism about Friend.tech is that, while I do believe there is value in access, I think that in the space that you and I are professionals in—in trading—that value is extremely dangerous. You can’t just take tips from people and expect that to work for you. If that’s what Friend.tech is selling, Friend.tech is mostly crypto influencers right now, so I would be worried.

I would think the product would need to differentiate between backstage tickets to an artist’s show or a conversation with Warren Buffett—innocuous things like that—versus getting in a chat room with this early Bored Ape adopter for NFT alpha.

Yeah, I agree with you, but I think you’re painting an unnecessarily narrow picture. I think that in about 2 weeks, this thing is going to go mainstream. It kind of is already there. There are a couple of gamers that have gotten on board. You’re right, it’s still mostly crypto by far, but I think, give it 2 to 3 weeks, people are going to come up with some pretty innovative things here.

I think the core thing here is that there will inevitably be a financialization of people. It started with sports stars. There were platforms out there where one of the theses was, “Hey, we’ll let you invest in Messi, or we’ll let you invest in this high schooler who might go to the NFL, and you’ll get a cut of his salary if he does, and he’ll get the money now.” There are all these different platforms that tried to do that with sports stars. This is just that, but at greater scale, with a better actual pool of potential investments, like social media influencers.

Okay, who do you think is going to get big? And not only who’s going to get big, but who is going to deliver the most value to the people that sign up to own their shares?

Here’s a crazy thought: Imagine that instead of buying a concert ticket, they just spin up an account for the concert, and the first 1,000 people to buy shares are the only people that go to the concert. There are so many random things that you can do with this, which is why I actually think there’s a lot of money to be made here.

If you’re not paying attention to Friend.tech right now, you probably should, because even if it flames out, it’s worthwhile staying on top of the developments.

I couldn’t agree with you more. I think something like this is going to succeed, and you should have a good understanding of what that something might look like.

No, you’re absolutely right, Avi. There have been a few iterations of products like this in the past in crypto. They kind of fizzled out, but that doesn’t mean that the technology itself isn’t perfectly conducive to this use case.

For me, it’s just refreshing to talk about something other than Pepe or Barack Obama Sonic Harry Potter coin. I’m just sick of that stuff. First of all, Friend.tech has a great name. It’s slick and catchy.

Financialization of people is a theme that, like you said, is a bit dystopian, but it ultimately appeals to our sensibilities because of what we’ve seen happen in our recent lifetimes with social media, followed by crypto, followed by basically the financialization of other things. Ultimately, I agree with you. I just might be a little scared of it, but at the same time, I think that if you’re not paying attention, you’re going to be missing out on a big trend.

If this is one of those things that goes mainstream, I don’t necessarily know what the best way to profit from this is as a trader, other than to buy shares of people that you think are going to go viral.

Yeah, I guess you really have to think about how to do it. Maybe the best way would be to—I’m just brainstorming here—you build a bot that monitors people who will almost definitely go viral but aren’t yet on the platform. The second they onboard, you buy.

That’s brilliant. I would bet a million dollars people are doing that right now.

Yeah, I’m just trying to think: how would you do that? People are definitely doing that right now. I feel like by the time this podcast comes out on Thursday, people are going to hear that and say, “Well, everybody’s doing that, Jonah.” Just for the record, this was filmed on Monday. I bet he’s actually going to be doing that for 2 days.

Yeah, it’s not quite as easy as, “We’ll buy the Friend.tech token.” It’s a little more nuanced, but maybe there’s a little bit of alpha in there for people who are willing to invest the time and effort and write a few lines of code.

Yeah, it’s pretty wild. I signed up, and people are buying my shares for some reason. Should I sign up?

Yeah, you should 100% sign up. I say “for some reason”—I actually really enjoy the platform. I’m probably going to start doing a lot more with it, so I think it’s genius, and I’m refreshed by it. This is a perfect use case for crypto.

All right, I think it’s about time to wrap this one. I appreciate you spending time with me as always, Jonah.

And likewise, Avi. Thanks for listening to my rants. It was good. I’ll see you out there on Friend.tech, everyone.

See you guys out there on Friend.tech.