No Rest for the Liquid The Role of Market Makers in the Ecosystem - John Gu

come in uh so today I am talking about the role of market makers I'm from alpha lab Capital so really quick just want to introduce Alpha lab capital on our background we're founded in 2017. uh we have about 80 people globally and we're market makers in centralized decentralized and nft exchanges so first of all what is market making Market making at its most fundamental is the act of providing liquidity to marketplaces so that people can transact what they want when they want and in the sizes that they want I think everyone here has probably interacted with a market maker in the real world so the example that I always like to use is the money changers that you find at airports right so if you're flying in from Europe you give them some euros they give you some US dollars if you're hopping on a flight to Paris you give them some US dollars they give you some euros the the function of these money changers at airports is so that people can transact the currencies that they want to transact when they want to transact and in the sizes that they want to transact in let's take a moment and imagine what the world would look like at airports if these money changers didn't exist so I'm about to get on a flight to Europe I'm running around the airport I'm trying to find somebody who just came in from Paris or Berlin so I can swap my USD for their Euros but maybe I want to change a hundred dollars and they want to change 500 or We can't agree on an exchange rate right it would be extremely inefficient it would be extremely expensive and it would be a huge hassle for everybody so the role of market makers is really to make exchanges easier and market makers exist in all kinds of markets so FX markets is one equities markets crypto markets but they also exist in markets like ride sharing Mark uh e-commerce marketplaces like Amazon so I'm gonna focus in on the financial Market specifically and talk about how Market making Works in financial markets so most centralized all centralized exchanges and also some dexes use something called the central limit order book or club for short on a club market makers post limit orders that other people can trade against and another term for this is resting liquidity so a limit order just says I'm willing to buy at this price and I'm willing to sell to you at this price and there's always a gap between the two prices and that's called the spread uh what is a maker versus a taker so in marketplaces in every transaction one of the parties is called the maker and the other one is a taker so let's say I say I'm willing to sell at a dollar right and somebody else says well I'm willing to buy at any price better than a dollar and one cent so there we have a match and the exchange matching engine is going to match us or fill those orders one of the the one of us is going to be considered a maker and the other one's considered a taker if you place a market order on an exchange you're always a taker if you post a limit order you can be either a maker or a taker and typically the algorithm that exchanges used to determine who's a maker and who's a taker is based on price time priority so who offer the better price and who was there first okay I'm gonna throw uh some more terminology at you right so these are the kpis that as a market maker we care about so we pay attention to the spread the spread is the distance between uh the best bid and the best ass price on the order book we care about the depth so the depth tells us how much can we transact if there's not much depth on the order books you can't transact very much uh the term that we use for that is saying that liquidity is very thin or the order books are very thin we also care about uptime so uptime is what percent of the time uh are we there quoting on the exchange as you can probably infer speed is incredibly important for market makers why is speed so important for market makers because we're basically quoting around the midpoint around the price and the price in any Market is constantly fluctuating it's constantly moving so we have to move along with that price and if we're too slow we can get wiped out all right I'm gonna dive um for a moment into a very high level look at Market making infrastructure and really kind of lift the veil and tell you guys like what the architecture of a market making firm looks like so their exchanges we connect to the exchange via an API in traditional financial markets is typically a fix API in web3 it's more commonly a websocket or rest API there are a couple of exchanges that offer fix but it's not common then from that exchange via the API we get in real-time Market data and then we have to normalize that market data because every exchange has a slightly different way of formatting their data and we just need to normalize all of it across all the different exchanges um so we're tracking that data in real time and we're also Gathering you know this historic book of Market data uh the historic Market data goes into something that's called the simulator so the simulator is something that we build ourselves which is meant to simulate the behavior of an exchange it's basically like a shadow exchange like a Sandbox exchange that we can play around in and test out different algorithms in um wow the exchange simulator is how we build our trading Algos so on this graph everything that's below the dotted line is research and back testing everything that's above the dotted line is what's happening in real time right so we have these trading Algos that were built using the using uh simulations and then in real time what we're doing is we're feeding the real-time Market data into those trading Algos and they're spitting out orders right buy this much at this price sell this much at this price those orders go into what's called a normalized OMS OMS stands for order management system what the omms does is it translates our orders into the specific format that's needed for each specific exchange because again every exchange does things a little bit differently and then it feeds that back to the exchange again via the API um so as I said before speed is incredibly important to market makers our end-to-end latency which is this entire Loop in the top right from when we get data from an exchange to when we spit an order back out to The Exchange that latency is less than three milliseconds okay so what do market makers do in the marketplace what's their function uh I want to be very clear here about what market makers do and what they do not do so market makers can help you solve your cold star problem so the cold start problem is if you're launching a new exchange or you're launching a new token on an existing exchange you don't have very much liquidity in the beginning right you don't have very many buyers and sellers so if you don't have very many buyers and sellers it's very hard for people to transact we solve that problem by as I said before providing resting liquidity and adding depth to the order books we tighten spreads I have another slide later on where I'm going to show an example of how we do that and we smooth price Discovery so smoothing price Discovery what that means is if you look at an order book without a market maker you'll see what's called price jumps where the price is at a dollar and then it jumps to dollar and ten cents or jumps down to 90 cents that makes it very difficult for participants in the marketplace because maybe they wanted to transact at a price in between but they weren't able to do that so market makers one of the things we do is we minimize price jumps what we do not do is control prices control volume there's a whole host of unethical and illegal things like spoofing wash trading Etc that market makers do not do if you're ever looking for a market maker and you talk to somebody who tells you that they can help control your price or they can guarantee volume run away from that market maker um so here's an illustration of some of the things that market makers can do for marketplaces right so the first one is tightness reducing the gap between the uh bid and ask price so you can you know imagine if the if the Gap is tighter right then it's less expensive for people to transact if there's a very large gap then it's quite expensive because it means that there's a big difference between what buyers are willing to buy at and what sellers are willing to sell at second one is depth so uh if you look at some of these order books that don't have market makers there might be like a couple hundred dollars of liquidity on the books which means that if you want to place an order for ten thousand dollars worth of something you just can't or you're going to move the price a huge amount or it's going to take you a very long time to place that order and then the third thing is resiliency so after an order executes how quickly does that spread snap back how quickly do we get back to an efficient market so this is an illustrative example of what market makers do okay this graph here is the average spread in the U.S equities Market starting from 1994 1994 over there 2014 over there in the 1990s that's when electronic trading and electronic Market making first emerged right didn't really exist before that so you can see that in the 1990s even in the U.S equities Market which is the most liquid Market in the entire world the average spread was about a hundred basis points or one percent one percent is huge nowadays the spread in the U.S equities Market is typically like one basis points or one one hundredth of one percent or even less than that and why this happened is because of market makers okay so I'm gonna close out the conversation by talking about defy because this is the D5 stage um everything that I've been talking about so far applies to clubs or Central limit order books which is the type of uh trading that happens on centralized exchanges now defy has um given us a whole wonderful new world of all different kinds of exchanges I'm going to quickly talk through some of the different variations right the pros the cons and the role that market makers play so you can think of this as a taxonomy of different dexes um so the first type of decks that emerged was uni V2 the original the Pioneer this is called a constant product amm it's a very simple formula right by which you create a liquidity pool and then two people can swap assets without any active market makers uh existent in the market because as is in the name it's an automated Market maker there are no market makers in univ too the pros of this is very simple the downside of this is is extremely Capital inefficient and it's expensive transaction fees on univ 2 are like five to ten times what you find on centralized exchanges uh then we had the Innovation after that of univ V3 so univ3 gets closer to what a central limit order book looks like with the innovation of concentrated liquidity where instead of just providing liquidity to the pool you're providing liquidity in a specific range right so that is much more Capital efficient but it requires active management and it's very difficult to do to be honest unless you're a professional Trader um another variation of amms is what's called stable swaps so stable swaps are great because they have much lower slippage but you can really only use them for assets that are pegged to each other like stable coins you can't use them for for non-stables um there are some dexes that are trying to do central limit order books on chain I think this is I think this is really fascinating to be honest I think this is the future of where we're gonna go with dexes but right now the infrastructure is is still problematic right because in order for a central limit order book fully on chain to work you need much higher throughput you need much lower transaction costs so dydx is probably the most well-known Club on chain and um they're now you know migrating to being fully Unchained but when they started the matching was off chain and then the settlement was on chain and the reason why the matching was off chain is because it was just impossible to do it in on chain it's too slow too expensive um another variation of dexes is oracle-based like GMX is probably the most well-known example of an oracle-based Dex um so as a market maker our role working with oracle-based x's can be to stream data to oracles um oracles are fantastic because they're informationally efficient right they don't start from this naive place of no knowledge about reference pricing the way that amms do but the downside is that they are susceptible to economic Tech so if you guys have heard of uh you know what happened to Mango markets last year that was an example of an oracle-based Dex that got manipulated and then finally a request for quote dexes so this is basically the opposite of how a club works so with a club The Exchange sends you Market data and then you say okay I want to sell or I want to buy at this Price Right with an RFQ the market makers whoops my computer died um with an RFQ the market makers are the ones streaming pricing in so for example hash flow is one of our partners we are constantly streaming prices then saying we'll buy at this price we'll sell at this price and then you can go to Hash flow oops sorry right so Traders can go to Hash flow and there's no slippage because you're seeing prices that are already being streamed in uh by market makers at which they're willing to transact so the main benefit of an rfqdex is that there's no slippage and it's Mev resistant um this is just what the Dex landscape looks like today I know that there are a lot of projects very actively working on innovating in the space coming up with new ideas I fully expect that in the next few years we're going to see you know things that are not here on this slide and it's going to be super exciting cool um thank you everybody for listening if you have any questions about Market making happy to take happy to take those questions