Investment Highlight: Numerai

Numerai is launching the Erasure protocol

Numerai is launching the Erasure protocol

Numerai started out as a hedge fund that used Blockchain and crypto to crowdsource predictions from data scientists around the world. Erasure is a middleware protocol on top of the Ethereum network.

Numerai the hedge fund trades traditional instruments like stocks based on quantitative models. A quant hedge fund would normally employ a number of math PhDs to develop and optimise the trading models for the fund. Instead Numerai is using crypto incentives to crowdsource this work. Data scientists participate in monthly tournaments by submitting predictions based on encrypted financial variables. Winning participants is rewarded in NMR tokens. The encrypted data are normalised and cleaned, looking like this:

Numerai’s tournament data

Numerai’s tournament data

As a hedge fund you don’t want to give away the data for others to work on. That’s why the data is obfuscated through encryption. The trick here is using a novel kind of encryption to preserve the structure of the data even as it is encrypted. This means that the data scientists around the world can build models to make predictions on this encrypted data. Thanks to encryption and crypto incentive models like staking, Numerai is able to crowd source machine learning data scientist from around the world. The result is that Numerai doesn’t have to give up their proprietary data and the data scientist don’t have to give up their models, only their predictions.

So far Numerai has paid out over US$6.5M in rewards and Numerai has been one of the most used protocols on Ethereum while backed by some of the most prominent investors in the space like Union Square Ventures, the cofounder of Renaissance (one of the top hedge funds in the world) and the cofounder of Coinbase.

While Numerai is really impressive and innovative, we felt that optimising on running a hedge fund, Numerai, wasn’t the right fit for our fund with its primary focus on open finance infrastructure projects. Numerai was too narrow and too centralised for our liking. That’s why we were very pleased to see the announcement of the next phase of Numerai; Erasure.


Erasure is an open platform for data prediction that can be used by anyone. By having participants create an immutable record of predictions on a blockchain, coupled with skin in the game in form of staking, Erasure creates a general market place for all kind of data that is open to anyone in world. There are huge amounts of data in the world but the good predictions can’t easily be leveraged and potential buyers of those predictions don’t have a reliable way of measuring accuracy over time. Erasure effectively allows anyone to time-stamp data, allowing people to develop an immutable data record. Further, participants can signal their belief in the data by staking NMR.

Numerai built the first two applications on top of Erasure, ErasureQuant and ErasureBay.

ErasureQuant is a stock market prediction market on top of Erasure. Anyone can submit a prediction of a stock in Russell 3000, that prediction is encrypted and sent to IPFS (a decentralised file storage protocol), while a hash is committed to the Ethereum blockchain. Over time you can create an immutable record that combined with staking prevents you from submitting multiple conflicting predictions. ErasureQuant is live and usable today, you can check it out at

Anyone can build their own markets on top of the Erasure protocol and Numerai recently put up a grant of $1M for developers to do just this. 


ErasureBay (deriving its name from The Pirate Bay) was the second market on Erasure that was announced. While the core building blocks are the same as ErasureQuant (IPFS, Ethereum and staking of NMR), ErasureBay is a general purpose data market, not necessary finance related. ErasureBay is designed to allow innovative uses for general purpose data time-stamping to create an immutable record.

With the Internet, data were let free, it became widely available to almost everyone in the world. ErasureBay is trying to solve the trust issues — i.e. which data should we trust. It is hard to predict what kind of data ErasureBay will be most used for. Will it be used by the next generation of stock pickers or whistleblowers. Perhaps the next Edward Snowden will use ErasureBay to build trust with the public.

Apollo’s CIO Henrik Andersson met with Numerai in Singapore during Invest: Asia 2019

Apollo’s CIO Henrik Andersson met with Numerai in Singapore during Invest: Asia 2019

Numerai recently made a public promise to quickly decentralise as much as possible while reducing the total supply of NMR — which is at the centre of everything in this new emerging ecosystem. The ethos and decentralisation of Numerai is quickly getting recognised by some of the top crypto companies. The Crypto Rating Council classifies crypto assets on the likelihood of them being securities. A rating of 1 means an asset has no resembles of a security while a 5 strongly suggest it is a security. Notably, Numerai scored a 2 — indicating it has little resemblance to a security. Only pure cryptocurrencies like Bitcoin and Litecoin scored lower.

In summary, we have always been fascinated by Numerai’s clever use of structure-preserving encryption and with the launch of their new platform Erasure coupled with the team’s strong ethos of decentralisation and track record of delivering, Numerai has a clear a place in our portfolio.

Thanks to Richard Craib for reviewing this post.

Henrik Andersson is the Chief Investment Officer at Apollo Capital. Based in Melbourne, Australia Apollo Capital invests in the crypto assets that are powering the next generation of computing infrastructure. For more information, please see

A New Framework of Ownership

Image | via Lorraine Grubbs

Image | via Lorraine Grubbs

Ownership of crypto assets is about control. ‘Owning’ a crypto asset allows the holder of the private keys to control future transactions of those crypto assets. Ownership of crypto assets doesn’t feel the same as ownership of other assets, which usually comes with physical possession and/or legal rights. 

Let’s look at an Ethereum transaction. I provide Bob my Ethereum public address and he sends me 5 ether. My public address can only be controlled by the corresponding private keys, which I store securely. A helpful way to think about public addresses and private keys is like email. Anyone can send to my email address (public address), but only I can send from my email address with my password (private keys). 

I now ‘own’ 5 ether and this ‘ownership’ affords me the right to control those 5 ether. They have been assigned to whoever has the corresponding private keys, which in this case is me. The 5 ether will always reside on the Ethereum blockchain. I cannot take my 5 ether off the Ethereum blockchain, store them on my hard drive and at a later date bring them back onto the Ethereum blockchain to sell or transfer them. They are digital assets that live on the Ethereum blockchain. The 5 ether I own are assigned to me, or transferred to me, such that I now have control of 5 ether that I did not have before. I, and only I, have the ability to control these 5 ether through my private keys.

Ownership of crypto assets doesn’t feel like ownership in the same sense that I own a house or I own a gold bar. Ownership of other assets comes with more than just control, usually physical possession and/or legal protection. 

Let’s consider other assets:

  • Gold - owning a gold bar likely means I have physical possession of that gold bar, which I can store in a vault or bury in my backyard. I continue to own the gold bar until I sell it or forget where I buried it

  • Property - owning land gives me control of the land, physical possession and legal rights

  • Shares - after buying shares in a company, my name goes on the company’s register. Nowadays, this is largely a digital experience. I have control of these shares until I sell them, usually on an exchange. Sometimes I do not have control of these shares, where for instance I am a minority shareholder and a critical mass of shareholders decide to sell all the shares in a company to an acquirer, in which case I am also forced to sell my shares

  • Cash (paper money) - I control and have physical possession of the $20 note in my wallet, but ownership of cash in circulation resides with governments and central banks. I control the cash until I exchange it for goods or services

Ownership of crypto assets is limited to control. There is no physical possession or legal protection. This might feel like a weaker form of ownership, but it is arguably the strongest form of ownership. Crypto assets are the only assets in human history which are unseizable. Cash, gold, shares and property can all be seized by the centralised actors behind these assets. Unfortunately, this is not an uncommon experience where governments cannot always be trusted to act in the best interests of their citizens.

Crypto assets are digital assets that reside on a digital ledger where ownership is the assigned right to control future entries on that ledger. If I buy or receive more crypto assets, I have the ability to control more future entries on that ledger in the future. This may seem incredibly simple but it is, in fact, incredibly powerful.

Investment Highlight: Algorand

This is the first instalment of a series we call Investment Highlights where we briefly describe some of our current portfolio investments.

If we define Bitcoin as a first generation blockchain and Ethereum as second generation, then Algorand could be defined as third generation. Algorand was founded by Silvio Micali, a Turing award winning cryptographer and professor of computer science at MIT.

Algorand’s consensus algorithm is called Pure Proof-of-Stake. At its core is something called a VRF or a ‘Verifiable Random Function’. Micali, Algorand’s founder, was involved in introducing this cryptographic primitive in 1999. Much of Algorand’s security, scalability and decentralisation properties are thanks to this VRF (Dfinity, is another Apollo investment that utilises this primitive in its consensus algorithm). For each block, the VRF essentially selects a random set containing 1,000 tokens where holders validate the block. Each block is produced in below 5 seconds. The result is a protocol with potentially a very attractive set of tradeoffs between fast confirmations, robustness and security (Algorand requires that 2/3 of participants are honest). Despite being a Proof-of-Stake network, Algorand doesn’t have slashing as they don’t feel this is necessary to protect against bad behaviour. In our discussions with Algorand they do seem to be flexible on this point — this is something that could be implemented down the line if required.

To keep the network as lightweight as possible and to avoid unwanted complexity, Algorand will try to do more in layer 1 instead of at the level of turing complete smart contracts (layer 2). In our dialogue with Algorand, they expressed a goal to replicate ~95% of the functionality of current Ethereum use cases with layer 1 technology.

While Algorand has been criticised for not being completely permissionless, it is our understanding that they do have a strong emphasis on quickly becoming highly permissionless and decentralised. While a non for-profit foundation in Singapore governs the network, they have a for profit company in Algorand LLC which will provide support services to the network à la the Linux/Red Hat model.

From left to right: Algorand founder Silvio Micali, Apollo Capital CIO Henrik Andersson and Algorand LLC CEO Steven Kokinos

From left to right: Algorand founder Silvio Micali, Apollo Capital CIO Henrik Andersson and Algorand LLC CEO Steven Kokinos

The goal of Algorand is to ‘democratize finance’. This is an ambitious goal that falls well within our Investment Thesis. Our sense is that Algorand will initially focus on tokenised assets, stablecoins and different types of financial contracts.

Since the network went live in June, there has been an impressive list of announcement from Algorand in a short period of time:

  • Tether has announced they are integrating with the Algorand Protocol.

  • ISDA membership to enable financial organisations to use their existing templates and tools to easily create decentralised financial instruments.

  • Proposal to implement new fungible tokens and execution of atomic multi-party transfers (AMPTs) in layer 1.

  • New 200M VC fund, Algo VC Fund, focused on the Algorand blockchain platform.

  • IDEX, the leading DEX on Ethereum, will work on implementing a platform on Algorand using its layer 1 technology.

Apart from the above list, Algorand has shown impressive agility and innovation when it comes to token economics. Their first Dutch token auction in June contained an interesting buy back mandate. Algorand since then implemented an early redemption program for auction participants as well as a 200 million Algorand staking program. Talking to the team, it is clear that the 3bn Algorand to be injected into the ecosystem via auctions and other means will be done so over a minimum of 5 years. We also note that there has been a recent proposal to extend the token distribution to relay node runners from 2 years to 5 years which should, if accepted, lessen sell pressure in the medium term.

Algorand has a place in our portfolio as a base layer 1 blockchain that nicely balances a number of attractive tradeoffs and has so far impressed us with what they brought to market in a relatively short time frame.

Henrik Andersson is the Chief Investment Officer at Apollo Capital . Based in Melbourne, Australia Apollo Capital invests in the crypto assets that are powering the next generation of computing infrastructure. For more information, please see

Bitcoin's Ever Increasing Scarcity

When Bitcoin’s creator Satoshi Nakamoto created Bitcoin, its inflation schedule was hard-coded in the software. It can not be issued beyond what the software specifies.

According to Satoshi Nakamoto:

The root problem with conventional currency is all the trust that’s required to make it work. The central bank must be trusted not to debase the currency, but the history of fiat currencies is full of breaches of that trust. Banks must be trusted to hold our money and transfer it electronically, but they lend it out in waves of credit bubbles with barely a fraction in reserve. We have to trust them with our privacy, trust them not to let identity thieves drain our accounts.

When Bitcoin launched in 2009, 50 bitcoins were issued every 10 minutes. That number is cut in half every 210,000 blocks —with 10 minutes between every block, that’s about every 4 years. This is the issuance of Bitcoin since 2009:



As we can see, close to 18 million bitcoins have been mined so far. If you look carefully you can also see two points in the history when the graph decreased its slope, in 2012 and in 2016. These points are called “halvening” points and marks the supply cuts. We are currently coming to the end of the third period when the supply is cut from 12.5 bitcoins every 10 minutes to 6.25 bitcoins. 

It is straightforward to calculate the total number of bitcoins that will ever be mined:

Screen Shot 2019-09-02 at 6.05.26 pm.png

Around the year 2140 the last bitcoin will have been mined and we will have the full supply of 21 million.

Lately there have been attempts to model this increasing scarcity in what is called the Stock-to-Flow model. For the below modelling we thank the pseudonym Planb, who writes on Twitter under the handle @100trillionUSD.

The stock of bitcoin is currently close to 18 million while the flow is around 0.7M per year for a Stock-to-Flow ratio of 25. The hypothesis is the Stock-to-Flow ratio drives value for a store-of-value type commodity like Gold or Bitcoin.

Source: PlanB

Source: PlanB

So far in Bitcoin’s life the price of Bitcoin has followed the Stock-to-Flow ratio remarkably well as can be seen in the chart above.

If we zoom out, we can look at where on the Stock-to-Flow ratio Bitcoin falls compared to other Stores-of-Value:

Source: PlanB

Source: PlanB

In the above chart, the coloured dots are Bitcoin while the grey dot is Silver and the yellow dot is Gold. Bitcoin has so far moved up and to the right as scarcity has increased. Note that the y-axis on the above chart is logarithmic and that the ‘market value’ of Gold is close to US$10tr vs Bitcoin’s current market cap of US$175bn.

I mentioned before that we are coming close to the next halvening event. In May 2020 Bitcoin’s supply will be cut in half again. Bitcoin’s inflation can be calculated as the inverse Stock-to-Flow ratio. After May 2020, Bitcoin’s inflation will be around 0.35/18=1.9%. That’s is below the target inflation of the global central banks —and from there it will only go lower.

We expect that the market will start pricing that event in closer to the end of this year. We think it’s wise for investors to consider a position as one of the world’s most scarce asset gets even scarcer!

Henrik Andersson is the Chief Investment Officer at Apollo Capital . Based in Melbourne, Australia Apollo Capital invests in the crypto assets that are powering the next generation of computing infrastructure. For more information, please see

Crypto is Based on Thin Air

President Trump recently shared his thoughts on Bitcoin and other crypto assets. We were unsurprised to learn that he is not a crypto enthusiast. The President’s concern that crypto assets are “based on thin air” is not uncommon. Crypto assets are not backed by anything. But the premise that an asset cannot be valuable if it is not backed by something is short-sighted. Gold, fiat currencies and paintings are all valuable assets not backed by anything. The true determinant of whether something is valuable is the tension between supply and demand for that particular asset. 

First, let’s start with valuing assets. How do we value assets? The simple answer to this age old question is the value of anything comes down to what someone is willing to pay for it. The financial services industry has developed models and techniques to determine the value of a variety of assets. While these models might help determine a range of potential values, the inputs to the models inevitably involve a great deal of subjectivity. (Let’s consider the alternative. If the models didn’t involve subjectivity, everyone would use the same inputs, the models would provide the same answers and asset prices wouldn’t fluctuate). Valuation is art as much as it is science and while these models may help, the true value of an asset comes down to what someone is willing to pay for it. Any other definition can only be theoretical.

How do we work out what someone is willing to pay for an asset? The answer: markets and the forces of supply and demand. Markets exist to facilitate the trade of assets. Less liquid assets are traded infrequently and price discovery is more difficult. Supply and demand are the forces that drive markets. Supply comes from issuing new assets and from sellers of existing assets. Demand comes from buyers looking to purchase at an attractive price. The value of any asset comes from the ever-present tension between supply and demand. 

Markets tell us that crypto assets are valuable. Crypto markets are currently worth around USD$250bn. That’s a lot of value based on thin air! As is the case with every other asset class, supply and demand determines the value of crypto assets. Supply comes from new assets, increasing supply of existing assets and sellers. Demand comes from the increasingly large number of people who are realising the value of crypto assets. The question is not “how are crypto assets worth anything if they’re not backed by anything?” Rather, the pertinent question is “what is it about crypto assets that makes them valuable”? The answer: crypto assets have unique characteristics that are unlike anything else known to mankind.

Crypto assets often have the following unique characteristics:

  • Scarcity - using Bitcoin as an example, there will only ever be 21 million bitcoin in existence. If we think of Bitcoin as money, this is ‘hard’ money. Unlike fiat currencies governed by central banks, it is impossible to print more bitcoin, thereby devaluing all the other bitcoin

  • Digital - crypto assets are digital. The concept of assets in the digital ether is often difficult for people to grasp, they cannot touch or feel it and nor do they receive a share certificate. Native digital assets can be traded digitally with less friction than alternatives. As the world becomes more and more digital, digital assets make sense

  • Utility - crypto assets can be used for a variety of purposes, with different assets suited to different means. People can use Bitcoin as a personal, unseizable bank account, and all they need is a smartphone and internet connection. Crypto assets are often programmable, leading to a number of novel use cases, often disrupting and disintermediating incumbents. The unique economic model of crypto assets means they can capture enormous amounts of value in the process.



Let’s compare Bitcoin to gold. Why is gold valuable today? The short answer is because of supply and demand. There is demand for gold because it has always been valuable, because thousands of years ago, a group of people decided that gold has a number of unique properties which meant it would function well as a form of money. Gold is somewhat scarce, so it cannot be duplicated and printed (unlike today’s fiat currencies). Gold is durable, so people can store their wealth with some degree of assurance. It is malleable so it can be shaped into functional coins. Gold is fungible, meaning any piece of gold is replaceable with another piece of gold. Gold, particularly when shaped into a coin, is portable, meaning it can be carried around and exchanged with other people to facilitate trade. Gold became valuable because of these properties. The people of the day decided that gold is valuable and should be valued. This cycle developed further and further until gold was recognised as being valuable, by people all over the world. 

It is important to note that this process did not happen overnight. Imagine presenting a group of people a new asset and proclaiming that it should be immediately valuable. It takes a long time for people to realise the value of a new asset class, as they become familiar with the asset and as network effects develop. Just as gold did not become valuable instantly, nor did crypto assets and nor will they realise their full value for some time. 

If we compare Bitcoin to Gold, Bitcoin’s properties are superior. Bitcoin is more scarce than gold, easier to transfer, more divisible, equally fungible and easier and cheaper to store. On top of this, Bitcoin is programmable and is impossible to seize, unlike gold in the United States in the 1930s which was seized by President Roosevelt. The only advantage gold has over Bitcoin is it’s shiny. In time, more and more people will realise Bitcoin’s superiority to gold, demand will increase and the value of Bitcoin will increase. The estimated market capitalisation of gold is between US$8 and US$10 trillion dollars. Given Bitcoin’s current market capitalisation of around US$170bn, we believe there is a strong case that Bitcoin will capture some of the value from the gold market which is around 50 times larger.

Fiat Currency

When we talk about crypto assets as currencies, many compare to fiat currencies (US Dollar, Australian Dollar etc) which is “backed by governments.” I often smirk when I hear  people say the US Dollar is backed by the full faith and strength of the US government. The problem with this argument is this ‘backing’ exists until it doesn’t. Fiat currency is valuable, until it’s not. If macroeconomic problems creep in, people start to doubt the value of the currency. Once doubt creeps in, it can be a slippery slope until the currency is worthless. 

Sadly, there is no clearer example of this than the current state in Venezuela. Hyperinflation has recently been reported at 10 million percent and the local Bolivar currency is worthless.

venezuela streets.png

Venezuela is not an isolated incident. Hyperinflation has plagued many countries over the course of history.

hyperinflation table.png

I’m not suggesting that the US or any other major economic force is soon to be affected by severe inflation. I am suggesting that because something is ‘backed’ by a government gives it no more value than a crypto asset that is based on thin air.


monet water lilies.jpeg

Let’s consider a painting by Claude Monet. A Monet painting is not backed by anything other than a canvas. The painting has no intrinsic value. Yet, it is undeniable Monet paintings are valuable. Different paintings by Monet will have different value, depending on the particular beauty of those paintings. And of course, paintings by Monet are scarce. Monet died in 1926, so we know that there will never be any more Monet paintings.

We can draw a comparison to crypto assets. Crypto assets and rare paintings are backed by nothing, they have no intrinsic value. Yet this does not stop them from being valuable. Investors often ask why crypto assets are valuable if we can simply create new crypto assets. We would say the same about paintings. I can paint a picture of some water lilies at my local pond, but this will likely be no more valuable than any crypto asset I produce. The crypto asset that I create needs to have unique characteristics that are deemed valuable by the broader crypto community, just as my painting needs to be deemed valuable by the art community.


The advantage that gold, fiat currencies and paintings have over crypto assets is they have been recognised as valuable for centuries, even thousands of years. Crypto assets are ten years old. It is highly unlikely that the value in the US Dollar will evaporate over a short period of time. While we think it is also unlikely this will happen in crypto assets, any unravelling of value in crypto assets is more likely than the same for assets which have long been recognised as valuable. In this regard, crypto assets are riskier than the traditional assets discussed, and this is well known and understood.

Crypto assets have unique properties that make them valuable. They are not backed by anything, they might be based on ‘thin air’, but this does not mean they are not valuable.

The Inevitable Future Enabled by Crypto Assets

With the release of Bitcoin’s white paper, Satoshi Nakamoto unleashed an unstoppable force.

With the release of Bitcoin’s white paper, Satoshi Nakamoto unleashed an unstoppable force.

The book The Inevitable by Kevin Kelly talks about 12 technological forces that will shape our future for the decades to come. The 12 forces that Kevin uncovers are inevitable because they are rooted in the nature of technology, not the nature of society. It’s a great read for anyone interested in our future and the big trends ahead of us.

I remember reading quite a while ago that one of Google’s design policies for new products was to imagine a world with unlimited bandwidth and ever decreasing cost of computer memory. They knew that technological advances would keep making Internet speeds faster and that, similar to Moore’s law, computer memory would become cheaper and cheaper every year. I.e. by realising the inevitable road ahead they could focus on designing the best products for this future world.

In a similar vein, Balaji S. Srinivasan, former CTO at Coinbase, recounts how the Internet was mocked back in 1988:


The Onion didn’t see the inevitable future of cheaper bandwidth that 20 years later would make streaming services like Netflix ubiquitous.

I’d argue that when Bitcoin was launched in 2009, Satoshi Nakamoto had showed the world the inevitable future of non-government based money. This was something that Nobel Prize winner Milton Friedman had foreseen 10 years earlier in 1999:

With the realisation that money is not something arbitrary that humans choose to adopt but rather that money with the best properties tends to win out over time — Bitcoin’s future looks bright. Fiat currencies have been inflated over time as governments can’t resist printing more. Secondly, security is unprecedented for a decentralised digital system like Bitcoin:


Many may dismiss Bitcoin due to things like:

  • Volatility — Crypto is known for being highly volatile

  • Custody — Doesn’t fit into existing custody models

  • Hard to understand — Crypto is a new paradigm that takes time to grasp

  • Poor UX — Crypto is a new technology where UX hasn’t been prioritised

We should ask ourselves are these really things that can’t be overcome with enough time and engineering effort? We won’t get into the details here but we certainly believe that the above issues can and will be solved over time.

Just as Bitcoin made non-government money inevitable and unstoppable, Ethereum showed us that smart contracts are inevitable. Smart contracts are contracts executed by software instead to humans. Some of the use cases of smart contract might take a long time to play out and some more exotic versions of the future enabled by smart contracts might never come to fruition.

However, what is exciting to us is that you can see a first glimpse of a new, more decentralised world, powered by smart contracts. Early applications include things like Ethereum based Compound where you can borrow and lend in a trust-minimised way:


With trust-minimisation we mean that we don’t have to trust the solvency of a financial institution. Compound and many other emerging applications in the Decentralised Finance space are built with smart contracts on Ethereum. While these platforms are still not mainstream, you have to on-board people to Ethereum and in many cases install a browser app like Metamask, having global access to financial services flowing as freely as information on the Internet is a glimpse into a more inclusive financial future.

The book The Inevitable lists 12 verbs that will define our future: Becoming, Cognifying, Flowing, Screening, Accessing, Sharing, Filtering, Remixing, Interacting, Tracking, Questioning and Beginning. Thanks to crypto assets like Bitcoin and Ethereum, we would propose to add the verb Trust-minimising to the list of inevitable forces that will shape our future.

Henrik Andersson is the Chief Investment Officer at Apollo Capital . Based in Melbourne, Australia Apollo Capital invests in the crypto assets that are powering the next generation of computing infrastructure. For more information, please see