Scroll to view more →

dYdX Market Maker Debate


On August 2nd, we hosted a debate with Daniel Matuszewski (Partner, CMS Holdings), Bryn Solomon (CEO, MGNR), and Evgeny Gaevoy (Founder and CEO, Wintermute). This is the sixth AMA that we're doing in our series focused on market making and institutional trading.

Daniel, Bryn, and Evgeny discuss:

  • Their backgrounds and the stories behind CMS Holdings, MGNR, and Wintermute

  • Theses for 2H 2021 and beyond

  • Crypto market structure and spread pricing

  • Advice for pro traders looking to get into crypto trading and market making

  • Order books vs. AMMs vs TWAMMs vs. RFQ and the future of price discovery

  • NFT market making

A revised transcript is available below:

Vijay (dYdX): Hi guys. It's great to have everyone here today. So for everybody in the line, we have three of the leading trading firms and market makers in the crypto space with us. So first Dan who is a partner at CMS, then Bryn who is CEO at MGNR, and Evgeny who is CEO and founder at Wintermute. All of them are involved with the DYDX perpetuals market to various levels and all participated in last week's poker tournament as well, I believe. So the plan is to spend the first 20 to 25 minutes on quick introductions and a few round tables with quick answers on various topics from each of them. And then the plan is to spend the rest of the time so about 50 minutes or so on three meaty debate topics before a rapid-fire session and audience Q&A to close things out at the end. As a heads-up the audio here will be recorded and transcribed afterward.

So first things first, can you guys give us a quick introduction on the founding story of your firms and your backgrounds. And talk a little bit about the types of strategies and products that you focus on. Maybe we'll start with Dan and then Bryn then Evgeny.

Daniel (CMS): I've been trading crypto now pretty much the entirety of my sort of professional trading career. I jumped into this fairly early, when it was sort of just a very, very small asset class with some potentials. This was around late 2012, early 2013. I worked at Kraken as my first stint, Kraken has been around probably since 2011, I guess Jesse just released a sort of 10 years recap. But from there I went to work at Circle where I worked for five years running their OTC desk. OTC now is much different than it was then. Not all, but a very, very sizeable amount of the order flow was going bilaterally back then. Exchanges were a lot dicier. There were a lot more like concerns around compliance and regulatory sort of risks with facing these offshore venues. Just as a proxy, we would typically do sort of about what you'd see Bitstamp doing today, sort of on the high touch. And that scaled pretty well with 2017 when it got really big.

And then the desk obviously grew into all the asset classes and things of that nature. So that was primarily my trading background. And then spun off CMS on the back of that with a guy named Bobby Cho who ran and worked at Cumberland DRW, a big crypto trading shop started by Mike Komaransky back in the day. And a guy named Julien Seguin who actually was the back-end sort of infrastructure guy with me through the whole time with Circle Trade.

So we trade partner capital, but we run two sorts of books. One is a liquid active sort of trading strategies. This is pretty heavy derivatives-based these days. We do a bunch in the options like flow as well. We don't take counterparty, so we're not taking risks from other people and trading that way. But we do a bunch of different stat arb type stuff, and then funding-related trading a lot. And then we run a venture book, which is probably where people have seen us more because that's where the sort of headline stuff comes through. And that's mostly longer tail of issuance for anything with one to seven years sort of recruited the horizons with a blend of tokens and equity, but really sort of heavily skewed towards the token side as we know that. So that's our sort of where we fit into this space.

Bryn (MGNR): So, hi, my name is Bryn. I'm the co-founder and CEO of MGNR. Our background is, myself and my co-founder are both from the TradeFi derivatives trading world. So we worked at wo different spin-offs that started from ex Optiver guys. Worked in Hong Kong for about four years as an options market maker there and worked on various sorts of quant strategies, indexes, and single stock arb strategies. And then later moved to Chicago and worked at another Optiver spinoff, but in the more competitive and more liquid US market. At that shop, I was involved in treasuries and fixed income options market making, and a number of other quant strategies around that. Cut into crypto in probably early to mid-2017. And it was actually my co-founder who got me into it. He was passing through town and came to stay on the couch at my apartment, and him and another friend at the time had been kind of getting into Bitcoin and he educated me about that.

And then actually just started building a model, a market maker bot on Bitcoin. And so just as a hobby project, after we sat down and had a look at it. And after looking at the market 2017 for a few minutes, you can see how much edge there is, and especially hanging out at the time basis. And there were just kind of standard futures bases on BitMEX at the time, just had a ton of volume in it and we were just all over the place. So it was pretty attractive. And that's the first time we started getting into that. Since then, we've just continued to grow for a few years and become one of the larger market makers in the space where we have a super strong focus on automation in our core business. So everything is completely automated on the CeFi side of the world. And that's not quite as easy on the DeFi side of the world. Because of the nature of the space, things are changing so quickly. So we kind of try to be a bit more dynamic there.

I'd say we were a little later to the DeFi side of the world, we probably got into DeFi in maybe Jan or Feb 2021, so we missed the super juicy 2020 summer. And now we have various DeFi books and do some opportunistic trading strategies. We do industrial yield farming. And we have a venture book as well that we kind of got into since about February, March time.

Evgeny (Wintermute): Hi everyone. It's Evgeny from Wintermute, I'm the founder and CEO. So my background is pretty similar to Bryn. So basically prior to creating Wintermute, I was working at Optiver, which is the largest market maker, primarily in the options space. And I was there for about 10 years trading Delta One spot. But with a special emphasis on ETFs. So stocks, commodities, bonds, I tried pretty much everything. And I have basically built their market-making desk. And basically, in 10 years time, I kind of felt like Optiver became too big and, not to say bureaucratic, but just not interested enough in terms of growth. I went straight into crypto actually with two guys. And we basically started back in 2017. Then it was just like getting well pretty hot in terms of growth and Bitcoin going up and ever since. We started with just doing simple arbs between exchanges, like Coinbase and Kraken. And actually, our first idea was built to be a market maker for projects. And that's kind of what we started with. And from that we kind of evolved to be a more traditional market maker. So I think we were basically in 2019, that was when we started looking into DeFi, started with dYdX actually. We were one of first market makers there. And 2020 was when we grew our DeFi capability quite a bit. And in general, it was a massive, massive year of growth for us where we grew well, basically from less than 10 people to over 30 by end of 2020.

And where we are now, we are trading all the major centralized exchanges. But more importantly, I think we have a pretty much present on all the major Ethereum protocols. Trading-wise, we are doing both we're doing central limit order books, like dYdX, we're doing RFQ provision on Paraswap and 1inch. And in general, that's the core business. So we are doing market-making and liquidity provision. So there is also a significant strategic part where we still provide liquidity for, well, we basically do market making for projects. And we do have a venture book as of end of 2020. And do a bit of farming as well, but it's not as big yet.

David (dYdX): Awesome. So to kick things off on the round table side, wanted to get your views on where we are in the overall cycle. Obviously, we're in the second half of 2021 at this point and wanted to get your view on your top three theses for the second half of the year and beyond. So maybe Bryn, you can kick things off, then Bryn, the Evgeny.

Bryn (MGNR): Sure. Well, actually, I try to stay away from macro predictions or theses that I get to attach to it. It's probably just trying not to tunnel vision ourselves too much. I think it's not the same for every trading firm, but if you come from a higher frequency kind of shorter time horizon quant side of things, you often focus on very high probability trades. And so, if you kind of become too intellectually invested in certain macro thesis, working out a day, it can sort of hamstring your agility or just mental ability to kind of change paradigms. But, I will give a couple of predictions or things that I think might happen in the second half of 2021.

One of them relates to the options markets specifically, I've kind of been writing this thesis for a while and haven't heard a lot of people talking about it. But the way that I see crypto options market or a reasonably high probability of the crypto options market evolving is that eventually, finally, when we get a proper ETF approved in the US, I think that people would just start trading equity options on that Bitcoin ETF as a proxy for crypto market options. That's not a certainty, I mean, there may be other crypto options markets that continue to evolve in the crypto space. But almost unintentionally or ironically, I think most of the Bitcoin options action will end up on listed US equity exchanges. And I think Citadel and Optiver will just become the main market makers there, thus almost shutting out or making it hard for crypto native market makers to be involved. And yeah, no moral judgment there, it's just what I think is likely to happen.

Another thesis that's kind of probably consensus at this point, but just continued institutionalization and regulation in the space. I think it's actually going to make DeFi a lot more boring than it has been. So I think, unfortunately, we've probably been through the most fun year or two years of the history of DeFi. And every year it will become increasingly vanilla and increasingly sort of derivative and copycat the sets of ideas and the concepts that are played with as things also find in place and network effects kind of establish themselves. So that's a little bit sad, but try to make the most of it while things are still in the fun stage right now.

David (dYdX): Dan, over to you.

Daniel (CMS): I'm back by the way. I have now learned that if I leave the Twitter app, I get booted and I become a listener, so I'll stop doing that. So I'm actually curious just to piggyback on the last comment, how much option order flow for gold is GLD related to sort of the CME products? It would be like a great, interesting data point. Just curious how that played out.

So as far as what I see going, I would be a little more bullish. I sort of have been obviously yelling on Twitter about this, and obviously, if you ever asked me personally, I would tell you too, I think you really need to look at how much money is moving into the fundraising side. There's billions of dollars almost weekly at this point that you're seeing sort of raised either by projects or by funds that are going to them, raise with projects, right? There's a tremendous amount of private money moving in, definitely on the equity side, the corporates are just sort of, I don't want to say anything's too rich because that always has come back to haunt me, but there's just a lot of money pushing through this system and it's going to find its way into the liquid stuff at some portion. This is just historically how we've sort of seen it. There have been cycles of this, right. The corporates get a ton of financing and it doesn't make sense relative to the asset class and then the asset class balloons. And then you get the sort of flip back and forth, I don't know. So I think that's going to continue, I don't see any reason for that to abate. I mean, we have a lot of conversations in the background really just sort of, I've said it before, we take just phone calls sort of for no reason, but we talked to a lot of these funds because we want to know what they're doing with their raising. And I say funds traditional sort of money, guys that run mom short equity books. And they're like, "Yeah, we're looking at this crypto thing." And it's like, well, this is new, this is interesting. And like, we've always had phone calls like this over the years, but now we're seeing them actually raise money or punching into some of these startups as their first checks. Because what tends to be as the easiest way for them to get exposure to this stuff is to just shoehorn it into something that exists in their existing world. Somewhat so they're backing venture beds on corporate equity, which makes a ton of sense because that's very easy to fit in their profiles.But they're going to move up the risk curve, right. This is just the stepping stone of them being like, oh, now we own some Bitcoin and Ether spot. Oh, now we're doing this token project. Or, now we're forming this. It's all just like baby steps in that direction. And I think you're seeing a ton of money pile into that.

I tend to agree, I think DeFI will get more boring as time goes on, it'll just get arbitraged out. The interest rates sort of imbalances of the real world and DeFi are going to have to come in. So this is another big thesis I have is that, there's this huge disconnect, right, risk-free even though it's a much different definition of risk-free of cash lending in crypto, it's just on a different planet in comparison to what you see in basically negative-yielding world outside. But those are going to converge, there's going to be a ton more money moving in, barring any regulatory shock that stops that flow of money coming in. Stablecoins are just going to continue to go up into the right indefinitely, massively, as long as there's yield opportunity or just opportunity in general to earn anything. So I think that's going to get you. But the backside of that is going to crush sort of these juicy, ridiculous rates that everybody's gotten too accustomed to in this industry.

The other thing that I have, because I really don’t want to shill, obviously what we're talking about and our host here. But I very much think that the push to in particular perpetual products, but really all sort of futures into DeFi is an inevitability. Look, I think the big shock that's going to push it is, there's so much regulatory pressure coming from so many different directions on so many different venues for so many different things that, at some point, people just bounce to those for whatever reason. Most likely that they just can't operate on whatever venue they were looking at at any given point or even just they're worried about the potential of not being able to operate on that venue. And they move into these DeFi products. And there's a real win if you never have to leave and you don't have to worry about that problem in the future. And I just think that is the stickiness to get people to move over and then they just won't move back. I don't think that they're going to beat out their centralized counterparts, but the DeFi derivatives space is going to grow in order of magnitude. I'm a pretty big buyer at that thesis. And I think that you're just going to see a continued onslaught worldwide on a lot of these centralized venues. That's going to be the real, just quick reaction to get people to move there. But then once they're there, I think they're staying, I just don't think they're leaving. So I know those are the biggest things that I have conviction about right now sort of in the market.

Evgeny (Wintermute): It's kind of funny because, for me, I'm somewhere in between, I guess. So I'm not so optimistic but I also don’t have this pessimistic message that every single trader will move to the CME. But I think what we'll see is, well first, democracies as I would have is they'll see kind of two worlds go have this traditional finance world of CME and EUREX with options and sort of futures and everything else product trade-ins there. And basically the majority of institutional investors like big funds still trading there because that's how they used to do it. And at the same time, you'll see this massive Cambrian explosion, continuing both in derivatives and pretty much everything on a DeFi side of things. And I guess once we see enough of those institutional investors taking a leap of faith or taking a leap of regulators. I don't know which one is going to be, maybe both, going into DeFi at some point as if the flood or just burst and CME will die. And I think that would be a pretty nice development to see, because I think it's well for us last market-makers in crypto. We’d rather continue this space to develop continues this technology to develop and grow. And basically overtakes the CMEs as well. But it's also possible that it will never work out, and CME will take on some of the best practices. I mean, maybe put some of the settlements on chain or something. And to just continue to be CME dominance like we saw in the past. Because if history teaches us something then it's CME has a pretty strong position and the traditional exchanges have a pretty strong position.

But jumping into more in the micro world, what stuff I'm quite excited about is just how the derivatives on DeFi re going to manage the capital efficiency, especially when it comes to options side. And well, in particular as dYdX here, I'm quite curious whether present dYdX can copy the Deribit model and just list options on all the perpetuals it trades on basically therefore establishing this capital efficiency for market-makers and kind of basically pretty much copy paste from Deribit, that would be a pretty interesting approach. But also, anyone could do it anyway. But in the pure DeFi fashion, I think it would be really interesting if instead protocols Opyn for instance, or some other ones, basically once they allow everyone to liquidate everyone, I think it can also be another way for us to see this capital efficiency, as long as there are enough market makers who can manage those liquidations and make markets efficient and make sure that the whole protocol is properly collateralized.

And then the final one, I would say once, and I'm really bullish about, is basically index investing. I think that something that didn't really get a lot of traction in the DeFI and crypto in general yet. And partially is because it's still quite tough to build DeFi native indices on a cross chain assets. But partially just because, I don't know, I think the majority of people on the crypto Twitter things, they can outperform the market still and a lot of them do. But I think as we see a lot of new participants coming in, it will be harder and harder to do. And so they'll just see, well, basically my big bet is that we'll actually see big flow of funds into those DeFi ETF products whem once people realize they're just much a better way than investing longer term.

Vijay (dYdX): It's a good spectrum of views there. Moving on to the second of three round tables that we had, could you guys each talk us through a kind of a high level how you model your spread pricing. So given that market makers tend to, the spread tends to be the engine of profit there. How do you guys think about that? And maybe if it's pretty similar, then if you guys could also speak to kind of how that compares to how you modeled it in the sort of the TradeFi world, and if there are any kind of stark differences in terms of how you're doing the crypto space. So maybe Dan, if you want to go first.

Daniel (CMS): Yeah. I mean, I'm the least pure market maker sort of on this panel. So my answer will be obviously muanced. Our liquidity provisioning that we do in any capacity is pretty much all DeFi related. And it's obviously being an AMM is, you can say that liquidity providing, but really just as these DEXs are getting bigger on derivatives side, we're trying to be more and more what we used to sort of traditionally do in the market making side of the business back when I worked at Circle on the OTC side. I mean, I could speak a little more to that world when we used to do this where we were pumping out liquidity sort of everywhere. I mean, it was just fundamentally a function, at least at that time when I was doing it of how off the exchanges were to each other. That was a proxy of what we needed to be making. Because there was just a long time in the industry where you were just paid for pushing plumbing around, right, buying here, selling there, brain dead, just getting the money through the system. And we're like, all right, well, that's the edge that we sort of need to justify "Doing this". That's like how much we need to make in order to get this going forward. But it's more like market-microstructure of how we like...would set bid-offer... spreads now, in stuff that we do, like in the DeFi land. I think it's really just a little bit of a finger in the air being, look this is what the vol is, this is how much volume we're seeing punch through this is where we think it's an acceptable level of what we should be getting. It tends to be more that not, "Oh, we need to optimize this." the last thing it's all right, we have this opportunity in this subset of the market at any given time. And we see this larger opportunity so much further away, where does it make sense to have capital. And borrowing has gotten a lot better. So this is less of a problem, but there were definitely days back in 2017 where it was every dollar we had that wasn't spoken for, had to go to Bitfinex the next that morning, because there was a 7% wide spread to Coinbase and trading any spread on any exchange individually didn't make sense because that money was better used just cycling through the pipes. So you have a lot of stuff like that I think pushing against market makers at times that will indicate sort of like it's weird, right? That Coinbase-Bitfinex spread basically set parameters for how we would quote any individual exchange but it's not necessarily intuitive of why that width would be related to that. But that was like, put the capital problem on the backend, but that's also not as big a deal these days because there's just so much more money and so much better access to borrow. So I don't know. I definitely tiptoed around your question a little bit there, but that's sort of like my background to it.

Bryn (MGNR): I can speak to that a bit. I think it's interesting that your question is about spreads because if you broke market-making down into two high-level components, one would be the valuation, which is at any point in time what do we think this instrument is worth? And then the second part would be the spread. So how much edge do we need to do a trade on this instrument? And so you can quote a spread around your valuation. And so, from a quant side by far the majority of the effort and thought goes into determining valuation rather than the spread itself can be somewhat arbitrary and you'll still have perfectly acceptable outcomes if you have a good idea of the valuation.

But because we are talking about spreads I won't give away too much secret sauce, but just thinking about a few different high level inputs that are quite well known and would make sense in determining how to work out the spread. So fees are a starting point. So you have to at least make that, the fee. So whatever money you wanted to make before you linearly add the fee into your spread, and so that works in both DeFi and CeFi, right? If a protocol has better rebates, by two basis points, then we're two basis points tighter. And if a protocol changes their fee structure and taker fees are an extra five basis points, then we need an extra five basis points to do a taker date. So from the protocol or exchange perspective, it's pretty easy for them to understand how to generate the volume that way.

Some other ideas, as Dan mentioned, volatility is a big factor that you might want to consider in your spreads. In general, if it's a more volatile market, you need a bit more edge because there's higher uncertainty in the outcome of any given trade.

And I think another useful framework to think about is... I wish I had a pen and paper. I like to kind of draw this in a four squares sort of grid, but different... On one axis, you have the amount of turnover in a product. So ranging from low turnover to high turnover. And then basically the... The higher the turnover product the tighter you want to be with your spreads in general, because you'll make more money trying to capture two-way flow versus in other products with lower turnover, you won't be able to flip your inventory as frequently and so you'll be carrying inventory. And in those products, it's better to try to take a bit more of a position-taking approach and try to have positive carry on the positions that you're accruing. Yeah. Let's see if Evgeny has anything to add to that.

Evgeny (Wintermute): Yeah. I think you covered it pretty well. It's actually pretty much what I wrote myself preparing for this. Besides taking the fee into account, you do need to know about volatility and you need to know how much you can build the scalp and between your bids and offers. So at any given moment, all this plus your price is the secret sauce of any market maker. And if you nail that and then you will be successful, if you don’t, then you won't be that successful.

I think to add a bit, there is a bit of a speed to being taken into account as well. If we know that we are, I don't know, maybe not the fastest market-maker on ay given exchange, we would also quote wider or if there is an element to it where the execution is not very, it's kind of random. I don't know our order gets stuck for instance. And we can’t get executed in a minute time on some really awful exchanges, we might also quote wider just to account for that. Because if we give too much functionality to people who might be just arbing it manually, or maybe the exchange reason, it's a market-maker in every sense possible in the Crypto, but, yeah, in general. Yeah, I think that's it.

David (dYdX): Great, great insights there. I am just moving on to the last round table discussion. What advice would you give to pro traders, quants or ex traders looking to get into crypto trading or market-making, and how does having LP capital versus managing a proprietary capital influence your overall thinking here?

Daniel (CMS): Yeah, so I'll take this one because I actually think for us those two are uniquely intertwined. I would say this, like the highest edge is going to be the furthest out the risk curve and for lack of a better word the weirder sort of aspect of crypto that you're involved in. Right? All the money, not all the money, but all of the very easy money that was around when DeFi first got started, it was partially just a function of the fact that either it was very complicated and hard to know about it and, or it was really impossible to have traditional pools of capital interfacing. And as time goes on, those things get edged out, right? People see big opportunities and they started to knock them out. And they find ways to structurally be able to trade on those venues.

And they get compliance comfortable with all right, here's a solution and a geo and a trader sitting in a desk somewhere that we can use to get those inefficiencies out of the market. So the larger, more regulated pools of capital, and this is always going to be a function of whether or not you have outside capital as well, because when you take other people's money, it's a whole different ball game. And now suddenly you have a ton of other rules that you have to answer to. And you also have to answer the fact that, you can't just randomly jump into the first food farm and lose a ton of your money. And then you had to explain that, right? There's consequences more to your actions other than "well that sucks." Now I'm out that, now you have to answer to certain people, so you're going to move slower.

So my advice to be, if I could do this all again, I would try to be as nimble as possible and to be playing as far out and whatever is the most bleeding-edge aspect of the market in any given time. What you definitely don't want to do is you don't want to go backwards in the sense that you don't want to be like, "all right, I'm going to our CME futures against Coinbase ", that game's over, that's lost, yes, there's a ton of volume going through it, but larger pools of money have taken that. You're not... Maybe you’ll be the greatest thing, but it's a lot lower of an expectation that you're going to win that market versus would be like, this weird thing is happening in NTFs teams that, five guys on Telegram are talking about. And I happen to be able to get access to them. So I have an opportunity here and it doesn't take a ton of money, but there's huge upside on it. That's where you really can sort of make money as an individual. And also frankly, if you can keep that pace, that's a much better seat to be in, in my opinion, versus going building this traditional sort of like shop that at this juncture is always getting harder. So I don't know. One of the unique things about crypto is it's extremely open. You can get to pretty much anybody at any point in the industry with one hop, even direct a lot of times, it's incredibly easy to get access and information that you need. I've said this a couple of times, you can just hang out in random Discords and make a very nice living just by paying attention to the information flow and knowing what's going on.

So that level of edge really doesn't exist in other asset classes as much maybe there are weird niches of certain stocks that you can sort of do the thing but where crypto, it's prevalent everywhere, and you don't need a lot of money to do it, and that gives you the full autonomy to be able to grab whatever you want to do. So I don't know that that would be my advice to somebody that was looking to get into this. It'd be find some... Whatever the bleeding edge of the market is in any given point and just become obsessed with it and try to just deploy as much of your own capital into it as you can. Obviously if you don't have any money, even if you don't have any money, you can probably do it, right? I mean there's just airdrops for money in this industry sometimes. And people would've done ridiculous things that and turned outrageously small sums of money into outrageously large ones. So I don't know if... That would be my outgoing advice to anybody who asks me what they should do in this conjecture.

David (dYdX): Thanks, Dan. And we'll get back to the NFT piece later in the debate section. Bryn, over to you.

Bryn (MGNR): Yeah. I largely agree with Dan on that one. I think it's the exact kind of right approach. Just if you're a home gamer or if you're even a trad guy and you're looking to get into crypto, just, I mean, so much information flow happens on Twitter, in various Discords and it has really strong emerging market characteristics. So, you're going to find so much edge there. It's almost just comparable to sort of lurking around on Reddit, working out what the pump squeeze of the week is, all that same stuff happens in crypto and this is far less efficient and easier to execute manually. And I also think we've talked about this a few times in our team, but the tail upside in DeFi, if you actually know how to kind of screw around with protocols, it's really huge because you only need to find one thing a year that you can take a big swing at, and maybe there's a hundred million dollar upside and maybe you don't even need much capital on that because of flash loans. So when I say take a big swing out, I just mean, if there's a poorly managed massive pool of capital like the FEI protocol, for example, and if you happen to be able to work out a way to take advantage of that, I mean, you're done for life on that one trade. And so you could easily justify five years of lurking around in Discords with that kind of upside hanging around.

Yeah. And from the quant side, I would also encourage people to just have a go and just deploy something. The barrier-to-entry is just rock bottom in crypto, even on the CeFi exchanges. So if you take a big venue like Binance or FTX, I mean, you can sign up an account and then sort of get an API key and be interfacing with the exchange in an hour. And if you know how to code it's... or you can go on Github and there's so many other people who've already open-sourced bots and just take someone's bot and just actually start running it with a $100 of capital. And watch your orders going to market in real time and watch what's happening and just start to think about what you would change in the model or the parameters.

Evgeny (Wintermute): Yeah, I would say on my sides, it kind of depends where that particular person wants to end up. Yeah. If it's all about we'll make enough nice money on some opportunities in crypto. Yes. There's definitely a lot of them. There's definitely going to be a lot of them in next one, two, maybe even three years, just to make really comfortable living and just retire somewhere. But I guess if your goal is to actually build a company, then you really need to focus on a particular niche for rather longer amount of period of time. And I guess DeFi in general allows for more opportunity than compared to centralized exchanges because centralized exchanges, right now, in order to be competitive, you need to generate quite some trading volumes to get to really nice fee tiers. And it's not necessarily easy, without start in capital.

So, and DeFi is a bit better in that respect as long as you, well, as long as you can find this capital and finding the capital and basically this whole debate about against whether you want to run a hedge fund or prop trading, the answer for ourselves is that prop trading is answer simply because it's in the long run. Look, if you are building... If you're in a building stage and you are run your algos and not making any money, or you're doing something experimental and losing it in some experimental farm. Yeah. You’ll just get redemptions and that's not nice. And vice versa, like if you're a super successful then this is... Running a hedge fund would feel like it's a really, really expensive cost of capital because you don't make as much as LPs. So yeah, it's much nicer to just run your own capital and make your money. But to get there you might need to start with some... This is some capital from... WinterMute when started, we did employ somewhat profit share capital initially. And that's how we, that's what allowed us to grow as well in 2020. But once we go up to a level where the revenues appear sufficiently profitable, we basically recalled all profit share capital and moved on because in the long run, that's a much better way to do it.

Vijay (dYdX): Awesome. So switching gears a bit, we had a few debate topics that we've lined up for you guys. So the first one focuses on market structure and price discovery in general. Obviously, we've kind of had a significant amount of innovation from a market structure standpoint in the crypto space and from centralized older books, to a variety of AMM models, to RFQ models. And most recently the time-weighted average market maker model that the Paradigm team and others had released. So, with that in mind, from your perspective as a trader and market participants, what are the trade-offs between these types of liquidity models for traders, and are some better suited to certain products than others? And maybe if you guys, as you answer those, if you could keep in mind what these models might look like at scale in five years and sort of which old ones are they being built on. And also, how do you think about that from a price discovery standpoint? So with that said, maybe one of you that has a particularly strong view or kind of background on the TWAM. If you want to kick things off, that'd be great.

Bryn (MGNR): Sure. I'm happy to chat about that. I'll try to keep it high level. I still have some questions about the TWAM blog posts that I wanted to ask Dave, but I didn't want to ask him until I've checked with my team, just trying to tap out local resources first. In terms of AMMs, I mean, the first time you hear about the normal kind of XYK AMM model you think that it sounds very elegant and it is and it has a lot of kind of nice properties to it. But I think coming from a traditional market-making background, it becomes apparent quite quickly that it's a really poor kind of way to model reality. I suppose it's like, if you were trying to be a market maker yourself, and you decided that there was a single input to your model, which is how much size is being traded against me and nothing else mattered. And that's really bad, you'll lose a lot of money really quickly. I mean, having a market making model with no prices is just a recipe to get run over immediately.

So obviously these things have been evolving and people are trying to evolve them. And I think that's noble. And one thing I think is... I sort of keep coming back to a similar thought process, like every time I see someone make an iterative improvement on XYK AMMs, changing the shape of the curve or trying to add a feature or add an input, I feel like the steps are too small and that people are maybe just missing the forest for the trees like this. They're focused on these really micro iterations when the end goal is already known, which is kind of like a fully-featured order book where people have all the information and they can decide what types of orders they want to put in where, and when. Now that being said, the really most elegant thing to me about AMMs is that it allows retail or kind of passive capital to play the role of market maker.

So, yeah, I don't know exactly how the future looks, but if there's some way that somebody maybe more closely aligned to ourselves or another sophisticated market maker was to kind of design an AMM model with maybe 10 to 15 features that actually capture the things that matter in terms of liquidity provision. And then we're able to open source that or to place that on a protocol and allow passive capital to be the funding pool. I think that would be really successful and suffer a lot less toxicity. It's non-trivial for sure. I think it probably needs to interface with an oracle. I think it probably needs to operate on a chain that allows pretty cheap compute or fast compute, so that, you're either talking about Solana layer one or reasonably efficient Layer 2 Ethereum in those two ecosystems.

The TWAM thing. I don't know, I'm a little confused about it because it seems to me that it's more of an order type that's being applied to an existing AMM... The name TWAM sort of implies that the AMM is different, but as far as I can tell the AMM is the same. And it's just a new feature, which is you can do this TWAP order type. And so both of those things already have names. You know, the order is called a TWAP and AMM is called an AMM. And I'm not sure why they're being conflated here and what happens in the future when you try to add the next order type? What if you want to do a VWAP? Do you have to rename the AMM? But yeah, it's all just a working paper. And I think it's definitely a problem that's being faced right now by a lot of people, ourselves included. It is very annoying and difficult to execute a large size on less liquid DeFi poools. And so, hopefully someone can come up with a neat implementation of a TWAP or TWAP-ish feature.

Evgeny (Wintermute): Yeah. On my side, I kind of view it all from a more general point. If you look at traditional finance, it's basically traded in traditional finance, one way to look at it is basically market-makers fighting with informed flow, funds basically. And then in that regard, market-makers like huge HFT firms like Citadel or Optiver and they're kind of winning, especially if you look at retail side of things, you see the Robinhood-Citadel combinations, which basically completely takeover pretty much most of the flow from US exchanges. And also, and if you look at the institutionals and traditional finances like old institutional trading the size are going to very liquid derivative exchanges, like CME or Eurex, or it just goes by OTC desks or via OTC platforms like Bloomberg RFQ.

And I don't really see a reason why crypto would evolve in a different way, like combinations of platforms or protocols can be different. But I totally see, for instance, a combination of Paraswap or 1inch being in the role of Robinhood effectively, or like Zerion, for instance, being in a role of Robinhood and barely connecting all the liquid AMMs and bunch of market makers. And this is a completely valid model that can work for sport and derivatives and what not. When it comes to AMMs and the TWAP AMMs, I really don't see how it's basically... Kind of like mirror Bryn’s thoughts as well, the business of market makers is just so complex. It's way more than just like 10 or 15 or 20 parameters in it. And if you're a successful market-maker and you have this like secret formulas, it makes your money, why on earth would you release it in an AMM way instead of just making money for yourself. And I think that's why it will be quite a few years until we see any significantly complex and successful implementations of AMMs.

On the TWAP specifically, i’m not very bullish on it simply because it does... If you do a TWAP, you don't want anyone to know about it because you can be frontrun. And so far, what I've seen in... From the paper is the ways they want to address it as they want to, they basically want people initiating TWAPs to basically do some market manipulation... And basically pretend that they're buying and then, screw the people who are front running them. Right. I think it's just kind of poor design, because yet you just want to use TWAP. You don't want to manipulate markets on "screw those people". So I think those TWAPs would be really interesting how they taken by platforms like Paraswap. If they can implement this kind of functionality and do it in the hidden ways so the whole market doesn't see, and just query a bunch of market makers to take it over. So that would be a much better model.

Daniel (CMS): Yeah. I feel this is such a flame war in crypto, and I don't know why this whole order book AMM thing... it's getting rehashed. It's just odd that the people that do all the trading, feel that the order book makes the most sense. And they, at least in my opinion, have laid out very good reasons about their business that they know the most about. And then everyone else is fighting at them that they don't know what they're doing. It's just a very interesting... We're not completely impartial on this thing. Right? We've backed Serum, Solana. We have a lot of bets in the idea that this is going to ultimately land on the order book model. That is very much where we see the end state of all this stuff. And I was very dismissive of AMMs out of the gate, which was wrong. I will make one little caveat to what some of my colleagues said here is that... I think they actually have been very successful. I mean, there's billions of dollars in TVL and people are using them, they're moving real volumes. But I think that's primarily a function of, there's just a lot of capital in the system that wants to earn yield and has incomprehensible risk tolerance. Right. So there's somebody who has ETH and Bitcoin and they're like, "I'm just going to put it in this pool. What do I care if the ratios move around, I'm happy having more Ether, having more Bitcoin or I'll take the path dependency. And maybe I'll earn a little bit." In that pool of people, I think has largely not... obviously, there is opportunity. And you can obviously blend it bilaterally and whatnot. But it's just been very easy to slam it in a pool and leave it and be like, "Oh, that's cool." And that's stuck. Right? People love this. And obviously, yield farming-related stuff has jacked up a lot of this across different stuff. But that whole subset of people, I think if any order book design in the end, should find some way of incorporating that desire into it. Because that was clearly a tapped demand that wanted to go out there, and maybe that's just a better way of and maybe juice it if you are willing to take some risk on some bound of some effective option that you're selling or something. That, coming in, is, I think, the big takeaway.

As far as the TWAP AMM model and design, I am just skeptical still of anything large that matters being executed on chain in the current design. I still think that most of the liquidity that for sure and then beyond that, there's just too much information leakage across the board. I'm not entirely sure how you would do something material and real without, not only people that want to argue taking out, but just having Twitter all over it yelling within 30 seconds. And that's just not a good design. So I'm just skeptical from that. But in the high-level mechanics, I haven't dug a ton into the depth of it, but I would bet that there's just better options currently out there, if you're looking. I mean, we do execute large slugs and we have better ways generally of doing it. I mean, if you're obviously the market maker, you have a really good way of doing it.

And I think when Wintermute did that SHIBA trade, they were very transparent about how they're doing the whole process. I thought that was a really cool sort of breakdown. I don't know if that was officially a TWAP but it was functioning effectively, as one in a capacity. I don't know. That's where I fall in on the whole thing. I do think order books stuff, but I don't want to spend a year fighting about it on Twitter, so don't come at me about it.

Vijay (dYdX): Makes sense. If you guys had to comment on where you see price discovery coming from, say a few years from now, let's say the market's more built out both in DeFi and CeFi, you have a variety of product types out there. Where do you guys see the market pricing being led from?

Daniel (CMS): I think the trends your friend on this one. I think it's going to be the perpetual futures will continue to dominate price discovery, unless there's a regulatory shock that forces that to go away. I just think that's, especially now that you're seeing those instruments dominate things beyond just the larger cap names, I just think that's largely where you're going to see price discovery. Especially in aggregate risk of if there's any sort of like macro pulse to crypto, that's where you're seeing that stuff play out in, I don't know. I don't think that's going to change.

Bryn (MGNR): Yeah, I mostly agree. Over the last three or so years after we started looking at crypto stuff, we had a number of models which were pretty, quantitatively keyed into all the different venues and liquidity sources. There have been changes or evolutions over the last few years, which venues or products lead price discovery. But for a long time it's- or a good rule of thumb that will normally work is just "Where is the most liquidity being transacted?" and that will generally lead you to the leading instruments that are driving price discovery.

Evgeny (Wintermute): I agree as well. That's what you see in traditional markets as well, where the futures products are basically driving the price action for pretty much everything. I think the only difference could be in less liquid tokens, for instance, like in the long tail assets, which might not be able to accommodate any kind of derivative market. And that's what you see, I guess, in single stocks as well, and traditional finance as well. But whatever top 20, top 30 tokens, derivatives will definitely lead to price discovery.

Vijay (dYdX): Got you. And the last thing to touch on here before we switch over to the second debate topic, if you guys have to comment on what you see the future of dark vs. lit kind of OTC pools. In traditional markets, you've had the development of these dark pools where a significant amount of volume trades and also a lot of these OTC platforms, similar to how you were extremely active in early on in crypto, Dan. And so if you guys are looking at that and sort of looking at the future of crypto and DeFi, and DeFi of course, extremely transparency focused, do you guys see volume ultimately all moving towards these lit pools effectively in DeFi? Or do you still see a role for bespoke and hidden OTC pools?

Daniel (CMS): I don't think there's enough end order flow that's looking for that currently to justify it. I think it's a tremendously retail-dominated market still. And I think that gets missed in a lot of people that try to build these products out. If you want to hide something, I mean, you can just wrap that through an OTC desk and they'll internalize it half the time anyway if it's like a large cap, or go to any one of these other guys on the panel and be like "Hey, I want to do this. Do you want to work with me?" There's options and there's not- if you built a dark pool for pretty much anything but the top five names, I don't think there'd be anybody using it. There's just not enough people looking for that at any given point, that are willing to pay for it, that would use it. Look, there's a world that that could get there, I just don't see it on the horizon currently.

Bryn (MGNR): I don't have a lot to add there. I wasn't exactly sure what we were talking about. The dark pools in TradFi were a failed experiment in my mind that there was a period of time when there was a lot of literature and buzz around them, and it doesn't really seem like any of those things stuck or attracted a huge proportion of flow, even though they hang around and I'm, well, it depends on what sources you read, but I'm reasonably confident that the balance of research about dark pools is that the execution is less good than a lit market.

Evgeny (Wintermute): Yeah. I think pretty much agreed with the guys as well. It would be really interesting if somebody figures out a way to direct flow, to llit pools in the liquid names, but make it not very obvious because the whole problem and the hole reason why people would go to dark pools or to OTC desks, for that matter, is nobody wants to leak information. So if somebody figures out a way to basically direct this flow on different platforms without anyone noticing, I think that'll be quite an interesting product to build.

David (dYdX): Great. Switching gears now to our second debate topic and this relates to one of the comments that Dan made earlier around NFTs. Somebody just posted a rare crypto punk NFT for sale for $90 million, that if accepted, would become the biggest crypto punk sale ever. The NFT was previously sold for $7.6 million so a very significant increase in the listing price. I was curious to hear your guys' thoughts on the punks price action over the weekend, there's a whale that's been buying up some of the limited supply. Do you guys see NFTs and in-game assets eventually becoming attractive or liquid enough to get involved in from a trading or market-making standpoint?

Bryn (MGNR): I can touch on that first if that's all good, I understand you're just introducing the topic, but just to make sure everybody's on the same page, when you offer something in the market, the ask price is not the traded price so it's titillating to see a $90 million ask on a crypto punk, but I too have listed an NFT for $90 million and I don't think it will sell. The punk might though. It's interesting that you ask the question, will it ever be liquid enough to get involved in quantitatively because we were already doing that in February and I think we were probably one of the earlier and more professional firms to be doing that in the NFT space. That's a little more in our wheelhouse, maybe more opportunistic, nimble kind of stuff that lends itself nicely to scripting things up quickly, but maybe sort of throwaway scripting that you don't intend to build out into a massive system.

I can give some specific examples because we were doing a lot less of that now, it kind of became less interesting, maybe this month we should get back into it. We did a number of sort of snipe bots on assets with fairly established price floors. So there was a long time, or a couple of months, when there was fairly frequent fat-finger order entry errors on OpenSea. I actually think a lot of that was from people trying to go direct to the contracts to enter the price. If you've ever tried to trade an NFT or play around directly with a contract on the Ethereum blockchain, there are various kind of decimal place rounding issues that you need to deal with and I think on a lot of these NFTs, they were either nine or 16 decimal places. And so if people are off by one when they enter the price, they'll sell you a punk for 3,000 instead of 30,000 and if you're the only bot who's listening to the blockchain for those specific assets, then you can do just pretty well every so often sniping those things to buy them. We also just played around with some other fun concepts at the time. We had valued all of the plots of land in a few metaverses, like sandbox and decentraland. This isn't a super new concept, but it's just the same as the real-world real estate market. Most real estate, you can estimate the price of it using quantitative inputs and all the same principles apply to the metaverse land market that there are- Location matters mostly and then there are certain features about location that matter more. In some namespace, for example, is it a waterfront plot? What's the maximum height you're allowed to build? Et cetera. So you can just build even a basic regression model and then just set up some bots to try and snipe land that's under-priced. One of the big things we went into was the first copycat of the Crypto punks, which was the Binance bunks, that was just kind of heard about that one night on Twitter and then the whole team was there just manually clicking it in before they sold out and we got a few hundred of those things and kind of were able to flip them. I'm not sure that at the moment, or maybe ever, there's enough juice there for a big trading firm like us now. We have too much AUM to screw around for a few thousand bucks. I think that's another great example of a reasonably real and doable edge for someone who might be on their own or a pair of coders or a pair of people trying to learn the space but maybe won't be attractive enough for guys like us anytime soon.

Evgeny (Wintermute): Yeah, for us, we are definitely not as deep in NFTs space at the moment and it's primarily because it's just not attractive enough for us to bother with it. There's always a lot of opportunities and mispricing but, just like Bryn mentioned, a lot of code is throw away and we are primarily focusing on coding stuff that hopefully we'll still be reusing in six to 12 months time, but on the NFT front, two things I'm looking forward for us to get involved in is, I guess first is, if somebody starts running fonts of punks, whatever, and they will be derivative and it would be actually possible for us to make markets on them and second, digging really, really into metaverse stuff. Once you have enough in-game assets that it's actually sufficient for us to get involved in as a market makers, that would be pretty cool as well. But I don't think it will be, at least six months, until we see some significant volumes going through in any of those metaverses for us to get involved as a market maker. But that would be really cool.

Daniel (CMS): Yeah. This is a really good opportunity set. What I was touching on before for somebody to carve out a very good living, just knowing all these intricacies of, even just all the punks and like playing around, there's just drop money everywhere, all kinds of random things, but to get to the point said before, it's not enough money to do systematically for size that I think shops that are larger can justify doing. I think if there were ever to be anything we would get involved on this front, besides investing in infrastructure that would help these tools or pieces of the ecosystem, would be if there were any index-related products that were liquid, if there was some way of aggregating trailing sales of punks to some instrument. If it was tradable in some fashion for size, in aggregate, I mean, I don't even know entirely like what that would look like necessarily. And I know some people have tried some things around NFT related indexes. It's just hard if there's 10,000 punks and you're effectively picking up pennies, trying to figure out little arbs between all of them at any given point or why stuff is systematically priced or under priced, it's just hard. I mean, you could take a big slug and just buy it for outright risk. That's definitely still in the cards, but it just trying to trade them, it's hard.

But like that being said, this is a great example of something that- look, I was personally dismissive of this, not like dismissive, but I was like, "This is cute but I don't really care." And the money got real fast for a lot of it and I saw it and I even saw the evolution of it as it was going through and I'm sitting right here. So, also the idea that there's always smart money, doing this stuff or that knows what's going on, sometimes you just have no idea where stuff is going to grow out of, which gets back to my original point of just pay attention to the cutting edge of it all because you have no idea what's going come out of this stuff. I am very, very interested though in how quickly the, I don't want to say mania because that implies it's mispriced wrong and I don't necessarily believe that, but just how fast things move in Crypto that it's ubiquitous now, "Oh, you got a punk, you've been around for a while or you're smart or you get it." I feel we were only talking about them realistically eight months ago and now it's everywhere. You've seen them and watched them. The pace is outrageous to me of how this stuff. And it's great. I love it. But that being said, the things that flow through people and their risk tolerance and their momentum and their speed and the way stuff gets adopted and all of those things that happen in crypto punks or just in NFTs in general, that same sort of ethos runs through all of the other aspects of Crypto. And that's why you see things just happen at such a large, fast scale and it's worth paying attention to, because it's no different than any of the other cycles or things that we've seen play out in this industry.

David (dYdX): That's great. And I guess just to follow up on that point, Dan, do you think that Crypto has a worsening liquidity and attention fragmentation problem? And how do you guys get over that hurdle more broadly?

Daniel (CMS): I mean, we get over it by, we just haven't done anything. We're just having interacted with it mostly because it is harder. I think it's probably also the reason I don't know a ton about this, but are there funds that trade art? I would guess probably not a lot when if they do, they're probably very small and boutique in themselves. It's just a hard asset class to be shifting around risk in or where expected value changes of future prices or actionable. So I don't know. I would bucket them in the same thing. I'm not dismissive of them. I actually find the whole thing, very cool and fun, especially as an onlooker of it all. It's hard as a trading asset for us to have much to do. There's guys on the desk that are very deep in this stuff and are talking about land auctions and all kinds of crazy stuff. Sometimes there's been a couple of trades we've gleamed off that, if there's a native token that has a purpose beyond the actual NFT, but otherwise, it's still really hard to have anything to do from a risk side.

David (dYdX): Thanks guys. Great thoughts. I guess, just to wrap things up before we turn it over to the audience for a few questions, we have two closing questions. So the first is, are there any exotic perpetuals you guys would like to see in the market or anything specific in DeFi?

Bryn (MGNR): For me that the one thing, if there's ever going to be a chance that people really want to trade volatility stuff in DeFi, which we haven't seen that appetite yet from end users, I really want to see someone do a clean implementation of just a realized volatility future. If you can work out how to implement that nicely, it's so much simpler to understand for traders and people looking at pricing, and it's easily scalable to any number of assets, no matter how illiquid, because the only input you need is the historical prices.

Daniel (CMS): I would love to see more index products. I'm a little selfish in this regard because, here's a problem that we find ourselves in a lot, we have obviously risk parameters that we keep ourselves in any given point, but let's say I'm like, "All right, I think we should buy a ton of Eth" for whatever reason, ignore the purpose behind it, but, we want to get long Eth. I want things to sell against it so that I can do that trade larger. And what the problem is, you can use the index futures on FTX, and those have been a big use for us, but the exchange on the funding ris eally hard, they're not that liquid, people aren't in it, and they're clunky sometimes in the sense that a certain name will dominate them, which is not great for us.

I really want to see a good index product that the industry can lean on and use for trading purposes. It's just a way to hedge Beta out. We used to do this with SPDRs, obviously when I traded equities like a million years ago, but I would love to have something like that. And if it was DeFi native, even better. If it was like, all right, you can only access this stuff or that's where its pricing indicatively came from and all the exchanges just keyed off that and get funding based on. It would be great. So I don't know, that I would really be what I want to see out there. And I'm not even entirely sure, what my optimal design of what it would like would be, but I would love to see more use of things like that. This also goes back to the NFT thing. If there was some sort of indexable component that was liquid and tradable, and I know some people are working on some stuff out there, but at least getting them actionable and people trading on them with help too so that there's liquidity. That's a big thing I would love to see out there.

Evgeny (Wintermute): I really second that. We really want index products to be, well, they're trying to push them as much as we can on all kinds of exchanges we are talking to and in general, if we could invest into whatever index on Solana, on Polkadot or Polygon, that's such a big opportunity in general and somebody should just list that. And I think they will be a lot of interest from both the buy side, as you just heard, but also from the sell side. We as a market maker would love to make markets in those products, and it's actually a series of products that will be liquid and to actually be a good substitute for the market in general and be a really good hedge instrument in the long run.

David (dYdX): Perfect. So just to close things out, we're going to have a “FUD or signal” rapid session. I'm going to say a few themes that are prevalent in Crypto, and then we'll walk through one by one and say, whether you think it’s FUD or signal and then a short explainer why? So to kick things off, ETH 2.0. FUD or signal?

Daniel (CMS): So, not giving directionality on it in any short term. I would say it's signal. It's incremental progress in the right direction for Eth. It's a big deal. To be dismissive of it in the sense that it doesn't matter for the price at all, is definitely wrong. That being said, you may think something has signal, but you may think that the market's overpricing said signal. You got to remember, everybody wound up on Eth and there's a reason it's outperformed the shit out of BTC over the trailing three or four months because of the back of this. That's part of it. So, signal, but not necessarily means it's going up. Is that answer your question a little more? I think it's relevant, to ETH and definitely relevant to eth's price going forward, but I worry that it's a wound up trade that's already gotten crowded. And I think you did see some of that in the washout that happened, I think it was a little more exacerbated in eth because there was just so much levered up into that sizing, which maybe helps now that they're on the backside of it. I think it's going to be interesting to see how the burns start going, if they're higher than people think then I think it's going to really fly.

Evgeny (Wintermute): Yeah, also, well at this stage, closer to FUD for me, because, well, pretty much for the same reasoning. I think it's already been priced in multiple times and nobody knows when that's going to actually happen. So from that perspective, it's just a FUD to me. I don't take into account in my assessment because I want to buy and I don't think anyone does really because it should be priced out by now. And then he talks about it, it's more like when it's going to be finally coming. You see all these huge ecosystem grounds staking instruments for instance, with this leader for instance and this stuff is happening. There is a lot of value locked into it. So, any additional use, kind of FUD to me, I guess.

Bryn (MGNR): I don't have a lot to add there. I kind of agree a little bit with each of the guys, short term, it's definitely a signal, it's moving prices. Crypto prices seem to be pretty reflexive in the sense that, which sort of leads the momentum. So whatever is hard, continues to get harder. So I think in that sense, it's hard to price pricing events a little bit. It's not impossible. I think the Uni V3 launch was super priced in. That was a pretty easy post-event short but this eth thing I can see it going either way so I don't want to draw a line in the sand.

David (dYdX): Great. Next one is the US Infrastructure bill and its impact on Crypto. FUD or signal?

Bryn (MGNR): I think on that one, signal only in the sense that it's part of a broader regulatory shift and has been happening for a long time and probably will continue to happen. So I don't think that bill itself is super important, but maybe it is, it's not my area of expertise. I'm not like Neeraj over at Coin Center or one of those people who's deep in the regs.

Daniel (CMS): Now we go, so Signal, but I'm going to ignore the regulatory side of it. I think it's just like the continued pattern that they're just going to keep injecting an incomprehensible amount of money into the system. And I just, I think that's the big takeaway here for all asset classes and crypto is just going to get dragged with it, kicking and screaming. Like this just isn't going to slow down, Right? There's just this idea that there is going to be any pulling back or fiscal responsibility is pretty much gone, at least stateside. I think this is just showing more and more of this coming through. So I'm excited as somebody seeing trillions of dollars move into sort of the ecosystem and the ecosystem being the U.S. economy and then somewhat whatever small piece of that hits crypto, I think will matter. So I don't know that's in the regulatory side, like I said, I don't know, but I think this stuff tends to usually just like work itself out.

Evgeny (Wintermute): Yeah, I feel like we are going to agree on pretty much all of those points between the three of us. It's I think it's definitely part of the bigger like regulatory pressure frameworks that, it's really hard to see where like how big of an impact they'll have, but it certainly will have an impact on, for instance, like the number of institutional counterparties is moving into space because some might be worried that a particular bill might hit them. So from that point, like the more of those, like those bills, the more signal.

David (dYdX): Great. Next up miner extractable value (MEV) - FUD or Signal?

Bryn (MGNR): For me. I don't know. I'm kind of like a little fatigued about hearing about MEV. I feel like there is a lot of discussion about it. I think, I mean, it's important to understand and to work out, but I don't know. I feel like it's a big black hole or a big sinkhole of a lot of smart people's energy, which is kind of not super productive.

Daniel (CMS): Yeah. I'll just touch on it. It burns a lot of resourcing for us, for problems that we don't want to have. It's just like this endless, I don't know if it's endless, but it seems like it's just this constant cat and mouse nonsense that we do. That being said it costs us real money. I think we were close to a million dollars just in like sandwich attacks when we finally got around to fixing the problem, but then it's like you fix the problem and then there's a better mouse trap. And then you got to fix that. It's just like, it does seem nauseating, nothing gets built at the end of it. You're just like constantly sort of having this arms race. It feels like a little bit, I guess, like the race to speed and HFT sort of things as an analogy, but yeah, I mean, it's real. I don't know enough about each roadmap to know how any of this stuff may or may not change, but I'll tell you it does like burn real resources and like money internally, which is like, so it's real in that sense.

Evgeny (Wintermute): Yes. The analogy with high frequency trading like arms races is correct. It feels like exactly what, like when I was Optiver, as a market makers, we're doing this first like cables and the microwave towers and radios bouncing off the atmosphere. It's basically the same. It's just like a lot of wasteful spending, a lot of waste of time going into it. And I think the good thing is it's kind of like the guys who are behind any of protocols, whoever is building Ethereum to Ethereum 2, they can take this into account, they can fix it. But it's also like, we actually could have predicted this. This would be happening like years ago as well. So it's, it should not be like a massive surprise to anyone that this is happening.

David (dYdX): All right. And then the last one is Layer 2 growth versus other layer 1s like Solona - FUD or Signal?

Bryn (MGNR): I don't know how to categorize that exactly. I think it's kind of a no brainer that Eth as a Layer 1 is not sufficient, so L2 growth is a good thing, but I don't want to make a strong guess about whether Solana or some ETH L2 is the emergent winner in five years time, could go either way.

Daniel (CMS): I think you could have both, right? Like you're going to L2s that have that have some real volume go through them. You’ll obviously have L1s at least getting value for like in aggregate. I think you can't be dismissive of that. Whether or not like you can say all the economic activity on it is like true and pure like matters or is competitive. Like that's like a harder sell right now here. Here's what I think is going to be an interesting problem L2s are going to have is that a lot of value for these ecosystems is driven just around the fact that their L1 has value. It's like people care about ETH because they care about ETH’s price and they care about they're vested in it. Right. And like, it'll be the same thing for Solana or Avax or Algo or whatever. A lot of these L2s there's not necessarily a clear value prop for using the L2 or that you're invested sort of in it. And may seem like a sort of irrational economic argument and be like, why do you care? You're just trying to like, get stuff done. It's a lot of this industry is built on that concept of I own this and I use it and I appreciate in it. And I want to see like, so I think that is like a weird problem that L2s like may or may not find themselves in when they're competing against like other L1s. But you can have both because Eth is so strong and people have so much value tied to that, that they're going to want to build things on it and keep those things running and use those things that are bolted onto it to make sure Eth like maintains some dominance. So I think that's like sort of how it plays out, but I, you just kind of have both.

Evgeny (Wintermute): Yeah. So to me it's kind of two problems. The first is, there is a really big fragmentation of liquidity on the spot side of things. There's also different L2s there is like Solana. Like the more L2 solutions we have and the more like an L1s like Solana we have like there more fragmented liquidity will become. And I think it actually plays really well into derivative kind of products because they don't care. So you can actually continue developing them and then speed is the only thing that matters. And the second sign is everyone is talking about how fast you can settle translations, but the real problems that they will start seeing, like both on L2s and L1s like when in any like fast chains is basically like reconcile everything because we actually saw it ourselves already on Binance Smart Chai. But now it's like very, very unusable because it's really, really hard to read it. Like you, it's really hard to sync your nodes, if you run your nodes and it's like a real problem. And I'm actually quite curious, like how L1s and L2s will solve that as well as things that's. Yeah. That sounds so they should be watching as well.

David (dYdX): Perfect. Dan, Bryn, Evgeny. Thanks so much for your time. That wraps up our prepared questions we have about five minutes or so for the audience. So if anyone has any questions, please raise your hand and I can bring you up on stage. Okay. So we have one question here, Michael, bringing you on. Michael, the floor is yours.

Michael (audience): Hey, hope everyone can hear me, sorry. This is my first time using Twitter Spaces and I'm still getting used to it. But thanks for the great conversation, really insightful. I work for a firm that provides backend crypto services to retail, brokerages and one interesting kind of, I guess development we've seen recently is obviously a lot of demand from retail customers for altcoins like SHIBA, DOGE, ones besides like Bitcoin and Ethereum. And as we start to try to explore finding market makers and liquidity providers for some of these altcoins, it's just, it's been proven to be pretty difficult to find some that are, you know, some of these more institutional and retail or institutional liquidity providers that have the capability to provide liquidity for SHIBU or SUSHI or something like that. So I guess my question is, do you see some of these bigger DeFi options, being able to provide that with kind of filling that space to provide liquidity to more institutional and retail firms like mine, because that's an option that we've looked at, but I think the problem is just from a compliance and risk standpoint, it's just kind of hard to, do AML KYC and stuff with more of these DeFi options. Hope. I explained that. Well, thank you.

Bryn (MGNR): Hey Michael, Bryn here from MGNR. Yeah. It's interesting. I'm not sure exactly how far down the totem pole of shit coins your clients are going. But you know, I think at some point it's a reality or, or we need to accept that, that it gets less and less likely that big institutions are going to be market making certain types of coins. It kind of reminds me of one time I had a person come to me via Twitter and said, “Hey, I know a guy and he's made $6 million on this shitcoin and are you guys able to give us a price so we can get out of it?” And I forgot what the coin was called, but it's some something Ape and I pulled it up and the price was sort of 0.0000001 and 0.0000002. And his entire net worth was dependent on getting filled on the ask instead of the bid. And there was a $6,000 of turnover a day. And, so if you get that far down there down the bottom of the barrel, like you just, the reality is you're never going to find a way out institutionally. And I don't know if you guys can interface with, directly with Uniswap or if that's what you, you were implying, you already have set up, but that's probably your best bet at that point. Yeah.

Evgeny (Wintermute): Yeah. I would say it really comes down to like how far, well, what's your risk tolerance on the regulatory side of things? Because yeah, if you do need to have like hardcore KYC and AML, especially if you're regulated. Yeah. You need to, well, you cannot be facing the hardcore DeFi and you cannot interface with aggregators. So then, your only option is, well, basically try to work with the market maker and well, I mean, you can drop me a DM after this session as well. We can explore how we can potentially help there, but like I can tell you the market marker that covers this like long-term list. They will not be like super hardcore regulated by, all kinds of regulators because they are actually trying to be much more nimble about it.

Daniel (CMS): Yeah. I mean, like as a rule of thumb, I would say like once it has a perpetual future on Binance or FTX, it's probably fair game that a larger market maker you can face will trade it.

David (dYdX): Thanks guys, Brett, over to you.

Brett (audience): All right, thanks a lot and enjoy the talk so far. So I have a question on kind of DeFi blending in with real-world assets. There's been a, I think like a really early stage trend right now to bringing real-world assets on chain, like tokenizing pools of mortgages or invoices kind of assets from the real world. I've always kind of had a thought that we'd see like public stocks and real estate and other kind of mainstream securities tokenized on chain. I'm curious if you guys think this trend is inevitable and then also maybe kind of what, some of the barriers are. And then also, I know you guys are talking a lot about derivatives if we'll see kind of perpetuals on chain for stuff that's kind of not crypto native, like some of the stuff that FTX is launching, like lumber futures and that type of stuff. If you see traditional assets as, as a growth area for DeFi, thank you.

Daniel (CMS): I think so I will say that I think there's an inevitability of like a blending of the two worlds, but I would say that it is going to happen in a much longer timeframe than like anybody predicts mostly because things that exist in the real world, like first of all, they've been around longer and they have very dominant, entrenched regulatory regimes that are going to be very hard to move into crypto specifically because like like that it doesn't jive a lot of times, right. Like everything in crypto, not everything but a lot of things are just like bearer assets like if you had them, you had them and that's like the end of it. That's not true about a house like if I put my house on chain and somebody like steals it, like doesn't really matter. So things that are going to be a little incongruous at times with portions of how the system works like in DeFi. And also there's just such a complex setup of like regulatory bodies that oversight all the things that like exist in the real world currently. It's just going to be very slow. I mean, I think the things that are quasi tokenized already, like stocks are probably the easiest one that you can see in analog to putting into something like crypto, but look there's like a lot of rules around like touching securities. Like our industry in general does a lot of work to make sure everything we do doesn't look like a security. So it's possible. And I do think it will happen. Like I'm a buyer of that thesis longer term and I just would not hold my breath that you see this in like the next five years.

Brett (audience): Yeah, definitely. Thanks Dan. Maybe, just one quick follow up on that. What about something like, I mean, Elon Musk has been in and around crypto. I mean what about tokens, like NFTs representing automobile pink slips, right. If kind of stuff like that, that just, brings more of the real world into DeFi, you kind of see some of that happening, maybe nearer term?

Daniel (CMS): Well, the NFTs stuff seems to have like a ton of attraction because that's like in general stuff that isn't in some regulatory like regime, right? So like you have like sort of open ability to do it. Like, I think that's why you see so many U.S. Corporates and just corporates in general, on embracing that in some capacity, because there's not like a ton of risk involved in doing so. I'll tell you what, no entrenched incumbent is going to do anything to like put their existing businesses at risk for crypto, for something that has like an unclear market, like opportunity. Unless you have like a very, very like, sort of aligned management team that views this stuff as the future and wants to do it for reasons that are not rapidly transparent, like why the economics would be there. I mean, like a Saylor type situation. So I think that's like sort of why you've seen uptakes in creative sort of related stuff and NFTs is coming out of the age. It's like, look like nobody wants to rock the boat to put their existing businesses at risk if there's an unclear market opportunity.

Brett (audience): Yeah, definitely. I can pass it on to the next question asker, if there's somebody.

Jordan (audience): So thanks for the great conversation. There's a lot of good information passed on. So I was just interested in hearing your guys' thoughts on moving crypto to the next level of consumer adoption. I was watching Vitalik speak at the conference about a week ago and he's around the boat, obviously like DeFi is great, but like what's past that the benefit towards consumers like earning a yield is awesome, but most people don't really care about that who are just everyday consumers. And obviously, we have gaming taking off on chain and I'm just interested in obviously NFTs that's a whole different ball game, but what do you guys see in the future of bringing consumer adoption to crypto?

Bryn (MGNR): Thanks, Jordan. Yeah. Good question. I think most people would agree at this stage DeFi and crypto is still very geared towards crypto native users and power users. And, and I think something that's actively being worked on by a large number of teams and is probably the focus of a lot of Silicon Valley venture investment at the moment is ways to improve UX and UI such that they can face retail consumers with no strong knowledge of crypto. I've seen a few other real-world examples of things kind of crossing over into crypto lately. I mean the Lollapalooza that just happened in Chicago, I'm pretty sure I saw, like, it was definitely sponsored by Solana, but maybe they actually had a merchant store or an NFT store online and it's very easy to see the leap to doing concert tickets. And other, as someone mentioned earlier, things like pink slips and such on a blockchain or via NFTs.

I think also this one's a little bit more under the radar, but I believe there were a number of apps that are being built on the Terra blockchain that are doing a better job of being consumer-friendly consumer-facing apps that than anything I've seen on the Eth or Solana side so far. And I don't know all of the kind of background behind that, but it's, if you start digging around some of these things on Terra, like they're just really nice, like the UX that they have apps that look a lot more like Acorns or Robinhood, and you can just imagine pulling them up on your mobile and it just sort of says, here's how many dollars you have. You can make this many dollars of interest just without people needing to understand things about blockchain and Metamask, all of that.

Evgeny (Wintermute): Yeah. I think, I think like to add to that, I think it's all about the gateway. So currently you have Coinbase and FTX and I was really active as well as gates into this like consumer interest in crypto. And I think it will, those as like companies they should be paying attention to in that regard. I think it would be quite interesting if somebody interfaces properly with DeFi by potentially bypassing like Coinbase and FTX and Circle can do it for instance. Just like allowing people to go directly to a nice UX or whatever Terra. I think those UXs and websites will drive their adoption from the retail side of things.

Jordan (audience): Thanks guys.

David (dYdX): Okay. So I think we have time for one last question Dema before is yours.

Dema (audience): Thank you. I was wondering what are your thoughts on options? I recently saw an article from paradigm on everlasting options. Does this has a place in crypto and what are the advantages?

Bryn (MGNR): Yeah, I've discussed that everlasting options paper with a few different teams that we work with, portfolio companies, and it's a nice idea. And It's quite novel and elegant compared to options that expire, there are a number of sort of nuances and disadvantages to having to roll options. But yeah, I'm not aware of anyone having actually built that out yet or, or being at a stage where they're seriously intending to build that out. I also still think that we just, haven't seen a proof that there's a really strong demand for options yet in the crypto space. So, you know obviously Deribit is the largest venue. And even those volumes, just like relative to the portfolio volumes are incomparable compared to traditional markets where in some markets option volumes greatly exceed futures volumes. Part of me thinks it's just a very retail-oriented market. Crypto is. And even where we see crypto demand for trading options, which is like a Robinhood. They've been super successful with this last year or so with options on GME and Tesla and such, really those things, I just view as a proxy for bored people gambling. And so if you can be a bored person gambling with shitcoins and perps in crypto, you don't really need to try and do options because it's more, options are a bit more esoteric and any regular retail, consumer picking up some sort of options GUI is going to be pretty scared by, by what all the things mean and how, how everything works. And so it probably does make a lot more sense for them to be gambling on shitcoins than gambling on options.

Dema (audience): Thank you. Thank you. That makes sense.

David (dYdX): Awesome. So that wraps up this debate, Dan, Bryn, Evgeny, thank you so much for sharing your thoughts. Thank you audience members for tuning in and all the great questions we'll be posting a transcript and then a recording of the audio on the dYdX blog later this week. And thanks again everyone for joining.

About dYdX

dYdX is the developer of a leading decentralized exchange on a mission to build open, secure, and powerful financial products. dYdX runs on audited smart contracts on Ethereum, which eliminates the need to trust a central exchange while trading. We combine the security and transparency of a decentralized exchange, with the speed and usability of a centralized exchange.