On-Chain: What Is Really Happening On The Blockchain? - TOKEN2049 Singapore 2022

welcome welcome thank you so yeah thanks very happy to have all of you today so Alex key uh uh Michael so I think to give everyone in the audience kind of a introduction to what you guys do you get some context so like I'll let you guys start and introduce yourself briefly um I'd like to start first yeah my name is Alex swanovic I'm the CEO and one of three co-founders at nunsen Nelson is a blockchain analytics platform that investors and Traders use to find out what is really happening on the blockchain um hello my name is Ki yongju um cryptoquan is the institutional create um online data provider um we are focusing on institutional friendly institutional grade index um to evaluate crypto assets so we just announced our Partnerships with CME Group our data is available on CME data domain so I'm happy to um have been here yeah hey I'm Mike raniger I'm the CEO and co-founder of genelysis uh analysis have been around for the last eight years or so and we are basically looking at the blockchain uh look at all the data that's out there look at this as a treasure through of data actually and on that we build different products for investigations work with a lot of governments we build for compliance work will basically every crypto exchange out there and last but not least we have a lot of data on business intelligence like tracing nfts doing other things so everything that happens on the blockchain is stuff that we care about all right thanks for the intro Alex I see you have a pretty cute penguin on the stage rocking it party fan Queen so yeah I mean uh so I mean you guys you know you guys track on-chain data topic of a discussion today is what's really happening on the blockchain I thought I give you guys an open question share to us an interesting Insight that you guys know that you guys have from the tools that you guys have built yeah happy to go first um so this one anecdote that I we were just talking about Miami backstage I was meeting one of our customers there uh who would set up nonsense smart alerts and in May when the Luna or Terra USD d-pag happened uh they got an alert that a lot of people were pulling liquidity out of the curve pools with wrapped UST and because they had that early warning with liquidity flowing out of curve they also pulled their own liquidity out of those pools and then as we know shortly after UST depeg so I think that's an example of how you can if you actually follow what's happening on the blockchain you can stay ahead of some of these large events it's not only for increasing your portfolio but it's also to defend your own portfolio when there are big Market movements um the problem the problem with aggregated on-chain data is noise so whenever exchange flows is skyrocketing um it doesn't mean anything right we need to figure out who sent those money why they're saying money uh what's the purpose behind this transaction so we have a bunch of other granular level Matrix like minor flows and you know utxo H pens so if you track down over seven year old Bitcoins um whenever these old coins moves um it's highly likely from I would say criminals or you know people who use gambling websites to work websites so whenever we we spotted that you know this whenever this illegal transaction happens it went to some kind of mixer um so it's highly likely uh bad news for this market so you you might want to check out spend output H pens over seven years um on our website and it when whenever this data spikes it perfectly matches the top which means like a significant selling pressure so yeah um it's pretty quiet uh lately so just don't worry over we're good so yeah cool yeah I think there's a lot of things you can find on the blockchain and one of the things when you ask the question I couldn't help start thinking what is really cool and I think one of the things that I I found really hard to uh to to analyze over the last couple of years and now we do a yearly report around it that's what happens in crypto in the real world so what countries are actually deep into crypto and what happens there and we've been following that year or year and for example number one countries is one of the neighbors to to Singapore it's Vietnam and then that's been state number one as for for many years basically and then we have a runner-up that's uh number two as the Philippines they came from like being number 15 to number two last year and it's it's still driven off like part of it is is the businesses but a huge part of that has been basically pay to play gaming so gaming is happening and you can see that on the blockchain is yet another example of interesting things you find there all right sounds good like a lot of interesting Insight just too much data going along like hot but it's like looking and finding right Intel in there um Alex you guys at nansen uh I guess you guys popularize the term smart money Trader uh kind of like trying to follow the smart money and you know uh maybe go for a attacking move to find like good position to have uh spawning or like you mentioned defensive move to kind of pull out like Luna uh for example like I'm just wondering right like how do you define smart money like can I be a smart money for example how do I be a smart money and and that's it does it still work to follow these smart money addresses like I mean at the start like when nobody's following yeah I can see how it can be an alpha but if everyone's kind of following it does it still make sense to follow does it still work yeah so I think like he mentioned earlier there's a lot of noise on the blockchain right and when we started announcing we wanted to find a way to surface the signal and all of this noise and so if you think about the blockchain or the blockchains in plural now you have maybe hundreds of millions of addresses of course many entities have multiple addresses and so on but what you really care about is a small subset of these addresses and what they're doing so what we're trying to do do with nonsense is to basically curate what are the addresses that you care about when it comes to investing in trading and so there are kind of two high-level ways to think about smart money one is just based on known entities so if you think about large funds that you will know Alameda Pantera a16z and so on most people would consider them smart money because they spend a lot of time in resources uh into figuring out what's happening in space developing investment Theses and so on the other way to think about it is from a behavioral perspective so you asked can you be smart money and yeah I actually have people that I know and you know friends and so on who have had the nonsense algorithms tag them as smart money I haven't checked your wallet maybe I'll check it afterwards but so definitely there is kind of one component which is social and you know about who this identity The Entity actually is and then it's a behavioral part and of course the behavioral part you can decompose it into nft flippers you can look at yield Farmers you can look at who's good at trading certain tokens to generate uh very highly positive p l and it's really a combination of all those things there's no silver bullet or one line kind of magic formula but it's really a combination of all these different signals that we group together and we think this helps investors navigate the landscape and understand what the smart money is actually doing makes sense makes sense so I mean talking about Wallace and also you have the algorithm where you kind of track the behavior but you have to have a way of tagging all these wallets and like from known entities and and that's kind of a very long list and entities change their addresses very frequently I'm sure you have like hundreds of thousands or maybe millions of addresses tag and I trade the same for both of you so I'm curious to hear from the three of you how do you tag these entities how you find these entities how you docs all these guys because they some of them go to some Great Lengths to kind of hide their Trails so yeah maybe you start with uh Alex and then we go down to my career yeah as I said there's no Silver Bullet right so you have to combine man and machine you need to have research analysts that are actually monitoring what's happening and it means in some cases reading about uh fundraising rounds and mapping the token distributions back to the lead investor and so on based on what you see happening on chain but you have to have algorithms too and more than 99.9 percent of all the 130 million plus addresses that we tag come from algorithms so it's not feasible to manually just sit around and tag all these different entities and addresses um yeah so it's really a combination of those two things man and machine and there's no Silver Bullet and I'm sure the guys here have some interesting ideas that we could borrow perhaps in the way we tag addresses as well um my time turning data to quantitative hedge funds and traditional finance and the on these people they don't like on data on changing data so our focus is building our algorithm to figure out um wallets for for example for exchanges we sent a small amount of money to exchange user deposit wallets and see where this money is going um so for crowdsource label like nonsense um is amazing feature um so I think it's a it's good for the community um so we are using dosing algorithm and sometimes machine learning to figure out these wallets yeah I think I think it's the best analogy to how to do this is um how do you run a machine learning algorithm you need training data so first first of all training data is in the human as Alex said like you you basically train the algorithm yourself there are certain things you can do it's obvious you can find the pattern and then you can reuse that algorithm in many ways but then when you want to do it at scale and look at billions of transactions over many many years the only thing that really works is to have access to ground tools data and our like SQL source to many ways is like basically the huge network of customers we have so like any transactions that being sent is being vetted for compliance reason and it means that we also get like a very good insight into what does a big service look like what is a big wallet what how does that look like and every time they send and interact with someone that's not a big wallet interacting with a personal wallet and what does a personal wallet look like and how do they move cross chain and so on that's again data you can infer from that and then like there's of course all the services out there that's right calculated and transacting and so on but then we have a team of I don't know 50 people or so that are transacting with anything that you should not transact with that's like dark net markets it could be terrorists uh cells it could be other things and they help like law enforcement Partners to figure that out so that builds like our data set of personal wallets uh interest in in in in like bad stuff on the blockchain and then all the stuff that you know and love and like we all use in the blockchain for every day so Micah um I mean I believe many of your clients such analysis uh law enforcement agencies and you have a lot of governments you know help with investigation ones like things go completely wrong for example uh man and one of the reason the largest hacks that have happened is uh the running hack uh 650 million dollars was stolen from the the the bridge the running Bridge so I'm curious to hear like can you share any insights from these incidents and how much of the money that has been recovered so far how much have you guys helped I'm just curious to hear any updates from this running hack yeah it's it's a great question it's like if you look at the Ronin hacks so so basically it's a bridge you have this hot wallet sitting in an infrastructure that's supposed to be safe as everything is it wasn't uh it got hacked by by the national group from for North Korea and uh what happened after that is like different schemes to launder the funds and how do you learn the funds well first the first thing you can do is being fast if you move the funds out and you move into an exchange and they're out and you have accounts that's been set up over many years and just waiting to be used for this purpose then suddenly you have to find this out that works then the next piece of this is like well some point people will figure that out and they're alerted so now you move the funds to a mixer this one this example tornado cash was like the main mixer being used here and then that sets everyone that raises the funds back in time and then you have another Advantage so what you need to do to be first of all figure out who did it and secondly being able to seize the funds which is really what you want to do is number one you need to have the data to identify the funds have the capabilities to trace through a mixer like a tornado cash which is like something we could do relatively trivially and then the next part of it is doing it fast really fast because if you can do it fast then you can actually get ahead of it and say hey ex exchange a b or c These funds are actually someone else's money and you should cease them and then law enforcement can collaborate on that and today 10 of the funds have been recouped that way because of collaboration between the industry between law enforcement and Analysis investigating the case so so you spoke about tornado cash recently the US government kind of put it down taken it down but I you know it's kind of a whack-and-mole game like right tornado cash goes down this is a matter of time before the next one comes up do you think um do you think what do you think of the tornado cash sanction is that the right thing to do can government should government sanction smart contracts in the first place and has anything changed on chain since the tornado cash ban happened the sanction happened yeah it's like it's a good question right and and again I would say um sanctions is like it's it's it's it's a big net you throw it at throw it there and it has of course an effect and to start with the last question yes stuff has changed there's hardly any funds moving through Connecticut anymore another thing that changed is there's copies of tornado cash out there so you you cut the head of a dragon and it gets three new ones so was that really what you wanted to accomplish that's the question maybe it has a little bit of help but we all know that this was actually a a method of laundering funds that was like getting better and better understood and the tools to to solve it was getting better and better so it would probably have meant the ability to see small funds by not doing it but again you want to show action and it did work like tornado can't stop stop stop acting and then to the first question you had was like what do you think about what is tornado cares about can can government do this and there's many many like views viewpoints on this I would say the the normal thing you would say like this is a service for privacy but is it really because what is privacy actually and and the way I usually divide what privacy is is that privacy is that your neighbor doesn't know what you're doing and anonymity is that no one knows what you're doing and we've used over hundreds of years banks for example and other institutions as our providers of privacy so our neighbor didn't know what my salary is but government could figure it out if they need to or if the bank sell them to something is going on so they kind of saw that and tornado cash only solved one part of that equation they sold like no one figures it out anymore and that was one of the the weak points in in a in a mixer in on that service so I would say is it right or wrong I don't know and you want to add anything onto that location I mean I think it's uh important to think about the ethical part of that which is a subjective statement and you have to look at the facts right and it's true like Michael said I mean there's less usage of tornado cash now uh in fact tornado cash was also used I think less than people realize uh if you look at the total number of addresses that have used tornado cash I think we're talking less than 25 000 addresses in total right so um I think that's an important point the ethical judgment on how highly you value privacy and how you view these different things that's a subjective thing that people have different opinions about but it's important to understand of course the technology the behaviors around tornado and I think if you see people just or criminals moving to other protocols instead then it probably doesn't have the intended effect right yeah I mean that was one of my next questions actually I mean we I mean we used to only live in a world where when you start actionalities not just Bitcoin and then you just want to track and talk Bitcoin addresses utx or and all those things and ethereum came about and they're like okay evm based account based let's start a track and easier to track a lot easier probably but but you know different and then in the last one year we kind of moved from ethereum evm based and the data launching data to a multi-chain world right so like there's like 60 over different chains that we track our coin get the terminal for example and I'm sure that all these generates all these swaps on all these dexes multi-chain Bridges across all these different 60 dexes it must be massive headache for you guys right like tracking all these data so I'm cursing here like how are you tracking in the multi-chain what how many chains you do um anything interesting on these chains as well yeah I'm pretty free for all questions I would say yeah so first of all this is a good occasion for me to plug nonsense portfolio which is a portfolio tracker that spans I think 45 different blockchains so it has very broad coverage to track your own funds and your own portfolio for the nonsense Pro product we tracked about 10 different blockchains if you think about the most granular access of transactions and smart money movements and things like that um and yeah it is it is a headache from an engineering perspective it means you have way more data because these other chains if you if you think of them compared to bitcoin or ethereum they typically have much higher throughput which means you have way more data so we I think we scan close to one petabyte of data across the different chains that we support every single day in terms of querying load and so on um but I think it's it's not going to reverse this trend towards multi line and I personally used to think that ethereum was going to kind of dominate everything and it was the only ecosystem that that would um have genuine organic traction I think what we're seeing now is different chains optimizing for different things making different trade-offs and they will find use cases for for you know various different things that users want to do I think one interesting thing to highlight I was looking at the nft data recently and you see magic Eden and Solana nfts actually growing quite a lot in activity compared to the rest of the nft market which has been largely going down so this is another thing that when you are tracking what's happening on the blockchain if you only look at one ecosystem you might miss a lot of activity in the other ecosystems so I think all of us here have to monitor what's happening on multiple different blockchains you can't limit yourself to just one ecosystem um from institutional artificial understand uh point of view I think um for us it was pretty important to just focus on bitcoin and ethereum so we've been providing Bitcoin ethereum data but um institutions they just overcame Bitcoin hurdle and then they are looking for ethereum data and then I guess looking for other data stuff so I think launching data providers they are dividing into multiple layers we're more like application layer so for infrastructure layer like the graph and some of the raw they blockchain data providers uh we we're using some of the third-party data providers for now but um once these institutions um you know on board those chains for quantitative trading and others I think um we will build some data influencer for them so yeah um things are changing right right now yeah and I would say like what what is it basically it's it's a it's a lot of data that you need to look at and there's a lot of like the trend is that the number of transactions is going up and like that's been growing and growing and that will only grow even more than than it is now so I would say our position in sinalis has always been like we need to check every single blockchain out there because every single blockchain could potentially be important for a customer in one way or the other and that will happen I do think though that like if you think about all technology and how it evolved we have like a rule of roughly a handful right so today I usually say there's three computers in the world it's Asia it's TCP and it's AWS so distribution is three computers it's nothing more than that that's the size and then if you look at like operating systems it boils down number of Phone platforms and so on so you typically see a curve where there'll be a lot and then there'll be consolidation I think we will see that but you cannot plan for that right now right now it's the other thing and no matter what happens in terms of consolidation all the features that us looking at data might hate about this too many transaction they will go into another blockchain then so it's not going to stop so it does make sense to invest in the capabilities to look at that and Trace through it yeah I mean I mean I I think I I hold a view that anything that can be tokenized will be tokenized so we have millions of tokens yes and when that happens existing blockchain just don't have the capacity to track everything so all these things that's happening on web to database a lot of them have moved to a blockchains we're powering the next sort of applications we will see app specific blockchains coming up and then we'll have to see an exponential growth of blockchains especially when evm make it so easy to just copy paste all this blockchain so it's going to be a massive exponential growth in data that a lot more work for all of us here on stage I suppose yeah oh yeah I have one question that I want to ask right so it's one of the questions that uh one of my followers on Twitter us uh Team Copeland from decrypt our slide how does the so we are in a month 10 of the current bear cycle the peak that uh was in November 2021 so we are kind of like went through 10 months of downwards down only cycle I suppose how does current on-chain activity you know in September compare against the same period during the last best I could say around mid-2018 curious to hear your point of views on this Mike yeah I told you to be a personal one so so I think that first of all one thing that's a huge difference is there's way more addresses and transactions that are out there now than there were before and way more use cases than there were before so going back to 2018 there was like small compared to now far far less less data it was basically just like ethereum Bitcoin that was it and now it's so much much more and there's so many more users that's involved in this another thing that's important is that I would say transaction volume has not really declined a lot the value that's transacted yeah that's down because that's like part of a bear market right but the number of transactions that's roughly the same so I don't really see a bear Market at least non-bear Market yeah yeah just to chime in on that so if you compared to 2018 um one major difference there are many different uh things to point out here but we have stable coins now like more than 100 billion dollars of stable coins and just the injection of U.S dollars into the crypto economy is very significant and so if you think about icos that raised money in 2017-18 most of them raised money in Eve that they had to sell off you know the next 12 18 24 months uh now most of the raised capital is actually in US Dollars and as we know looking at you know Fiat currencies around the world the US dollar is doing pretty well right so I think that's a major difference and you can see it on chain with the stable coins more than 100 billion dollars in circulation mostly from custodial stable coins like tether usdc also binance USD backed by by paxos um and of course algorithmic ones like die so that's a huge difference the other thing is as Michael was saying the transaction activity is there because you have use cases now you didn't have that in 2018. you have nfts right and transaction activity for nfts is still very high but it's spread across more marketplaces and more chains than it was one year ago uh I think D5 is maybe the area where you see less activity now there's been a lot of these kind of liquidity games and incentives that of course two years ago led to D5 summer you see less of that now but I think so the debate for D5 is more about how big is the Tam right because D5 makes a lot of sense but it's unclear how big the Tam is if you compare it to other types of financial products and so on yeah I mean if you just take a look around this conference like compare this against the previous conferences that we were here we were around in 2018 like I mean this is a massive conference we have Montana of a bear Market There's 7 000 of you guys here attending paying 500 1 000 per ticket so Kudos on you guys [Applause] what bear market right yeah and also I think I think it's worth mentioning like there's a lot more things to do on uh in crypto these days previously which is Bitcoin ethereum you can't that smart contract wasn't really the thing uh trading or decentral action just wasn't a thing nfts wasn't a thing so many more D5 applications uh in here I think nft was kind of a game changer in my opinion uh previously it was a very male dominated uh world but we don't really very Finance as well very nerd very CS guys we get a lot of uh creatives a lot more female participation these days so yeah I think I think it's interesting uh a lot more use cases uh for sure um eight minutes more at a time so curious uh let's talk about nfts big fan of nft um any interesting insights on nfts in the last six months and you kind of mentioned here and there so we have a section announcement called nft Trends and indexes which is actually free and open to use for anyone um and what you see there like I was mentioning earlier generally nft activity or rather nft volume in U.S dollar terms is down quite a bit I think most people are aware of this but you do have some some uh some portions of the nft market that are doing better than others you see Blue Chips holding up better you see art nfts holding up better whereas things like gaming nfts are generally down pretty bad probably because there was a lot of um hype and speculation surrounding it and there were frankly quite a lot of Supply like a lot of different projects and the thing people realize is that it's very hard to build games and so it takes a long time to actually produce what they are promising in their pitch decks and so on so generally see more of a downtrend on the game nfts I mentioned earlier that you you do see activity on Solana going up quite a bit it used to be only ethereum and openc now it's a bit more diverse which I think generally is healthy for the space but so it's it's a bit of a mixed picture I don't think the headline of you know nfts are dead or anything like that is very accurate you have to look at the actual sub sectors within the nft space and I think I think it's interesting like um at the moment almost everything is correlated with Bitcoin like ethereum is correlated Bitcoin I think when blockchains start decoupling from each other that's when we start seeing more interesting things so like we're starting to see early signs of Solana's activity decoupling from ethereum uh looking forward to seeing what the app tours and sui guys are doing there will be a pretty interesting new L1 coming up will they be the Solana Killers for example and then decouple from the activity there will there be a new use case that is kind of very specific to these chains and that we start seeing a lot of different things so I think all these things are very interesting and um will be very interesting to monitor what's happening all these genes so I guess for for three of you right I'm really curious to hear there's so much data what are the three top metrics that you kind of enjoy trekking maybe on a day to day or a week to week basis like what are three things there's so much data to look at so Mike yeah I think I think this is again um there are many things you can look at at the blockchain many pieces of data there I think one one of my and this is kind of the the the trend uh in the blockchain that's been concerning for many years was like it's a criminal activity going on in the blockchain what's going on there and for us like one of the metrics that we track year or year and it's being quoted like basically by any any news Outlet you can find it's like what is the percentage of criminal activity going on this year what was it last year how is it declining what is happening and so on and take for example last year uh everything grew in the crypto space all the all the uses was growing 5x uh and illegal activity was doubling so it was growing but not to the same amount and now we are talking basis points in terms of like counting criminal activity so it's just one of these things where anyone who sits in a bank in the room anyone who sits with a regulator or anything it's actually not that bad it's getting getting far better and that's been my number one metric track the other one is to use I said that before like I think understanding what's going on in the world of crypto and really getting an Insight is where things are moving is another important piece and then the last one that's getting into like the the core of more like the the nitty-gritty details of what's actually going on what are the trends is a bear Market coming is it not should you be concerned and that's typically what happens to exchanges indexes what are the what are the whales doing what are the retail doing what are the trends and that's kind of you could call it exchange influence on outloads but it's like that's another thing that we always track and look at it being like what is going on right now is it just noise or is it real um for me Bitcoin I'm more familiar with Bitcoin so uh for Bitcoin I think mvrv and spent output age bands and utxo realized cab agent um so these data is all comes from realize price which is some kind of the entry price for blockchain wallets so if you calculate on the when the transaction happens you calculate the aggregated uh the average price when those transactions happens or you look at some kind of Enterprise for the all the blockchain wallets so for Bitcoin there's nothing much to do with Bitcoin if you move Bitcoin that's highly likely to selling activities so um whenever mvrb goes below one which means people are underwater they're in loss um the market is in in the long term is like highly likely where um at the bottom and when this mbrv goes on above 3.

7 which means all the blockchain wallets p l is 20 270 percent uh where like cyclic talk this is technical though so for ethereum another chains and there are there are so many use cases just moving ethereum is not selling ethereum right so it's important to see granular level realized price and vrb so we're just extracting all the transactions um going back and forth with centralized exchanges and transaction in decentral exchanges and calculate realize price you will get selling uh you'll get realized price based on just selling activities so we're building a flexible analytics tools to calculate this type of Matrix so it's going to be launched at the end of this year yeah yeah I think the thing that makes blockchains very different from traditional Finance is that you can actually go down to the most granular level and look at individual transactions that's quite unique when you compare that to traditional finance and so in traditional Finance you're forced to look at high level metrics because it's the best you have but with blockchains you can go really deep so I think I'm not a huge fan of the concept of metrics I think most metrics are actually quite simple and the example of smart money I would use is kind of a suite of metrics or signals that you can think about so what are the inflows that you see into and out of smart money wallets that would be one thing I checked every single day effectively if I look at which tokens they're buying which ones they're selling the metric is very simple it's literally just inflow or outflow into those addresses but that can often lead to a further investigation into which specific addresses are buying right is this is this someone who has an activity or a history of holding these tokens or are they more of a flipper if you think about nfts and so on so the first family would be kind of smart money related signals and metrics and then the other thing which key also mentioned briefly is around centralized exchanges and liquidity so looking at the flow of tokens into and out of exchanges and that's I think think relevant both for specific cryptocurrencies and tokens like eth but even stable coins too because you need to understand how much U.S dollars are basically flowing into exchanges and at this point of course as we know many people prefer using stable coins than Fiat so I think looking at the flows of stable coins into and out of exchange is very important one to monitor all right um I think that's all the time that we have for today so thank you very much Alex key and Michael for what is a panel yeah thanks Bobby thank you thanks for having us