Open in app

Sign in

Write

Sign in

Augur | Token Fundamental Analysis

Cyprien
17 min readSep 5, 2019

--

Highlights

  • Interesting market with high potential, but lack of active users
  • Solving a big issue with a bad UX and worrying mechanism design flaws
  • No break-through technology and a convoluted resolution process
  • Currently way overvalued on the fundamental side
  • Augur V2 (to be released in September) might solve some of the issues of the model, but will liquidity and participation rate increase?

Business Description

Augur is the largest fully decentralized and open-source prediction market platform built on top of the Ethereum blockchain. It was founded in 2014 by Jack Peterson and Joey (Joseph) Krug. Augur’s ICO in 2015 allowed the organization to raise 5M USD and subsequently launch the mainnet in 2018. The goal of Augur is to provide a betting application (and an entire decentralized financial system, according to the Augur Master Plan published in 2017) that is censorship-resistant from platform providers and government bodies allowing any user to create a prediction market.

Based on its convoluted resolution model, Augur’s value proposition lies in forecasting and communicating the outcome of real-world events thanks to designated reporters who get rewarded in REP (Reputation, the cryptocurrency for the Augur ecosystem) for every correct reporting on actual events and the ensuing market resolution. This system aims at creating a universe where market odds should align with real-world sentiment towards a specific event by allowing traders to participate in these peer-to-peer prediction markets.

Part 1 — Business Case

1.1 Market Opportunity

The prediction market is a space ripe for disruption by blockchain-based platforms. Vitalik Buterin (an advisor to Augur) has been vocal in the past about the importance and necessity of these types of use cases.

In Augur, the prediction market is a collection of traders speculating on future events or outcomes. These events are varied in nature and can be related to anything, from sport results to elections outcome to stock performance. So these markets essentially perform as event derivatives. In a well-designed and efficient market, the value of such derivative should reflect the probability of an event occurring.

Now, thanks to the advances in blockchain technology, predictions can be far more accurate using the “wisdom of the crowd” by relying on a multitude of users around the world (thus theoretically diminishing the idiosyncratic bias effect on the overall prediction).

1.1.1 Total Addressable Market

The predictive analytics market size is expected to grow from USD 4.6 billion in 2017 to USD 12.4 billion by 2022, at a Compound Annual Growth Rate (CAGR) of 22.1%. This is obviously taking into account centralized as well as decentralized players.

But the predictive analytics market is really small compared to the potential of a cross-border betting platform: in the US alone, one report estimates that Americans bet on sport results more than 150B USD per year illegally, plus roughly 5B legally. That figure has been criticized and revised to be closer to 70B USD. Still, it is estimated that the global business of sport betting should be north of 250B USD, with Asia and the rise of eSports being the main drivers of growth in the future.

If we add to that the potential market for political betting (for example, about $277 million was traded on Betfair on the 2016 U.S. presidential elections, despite the fact that Betfair excludes Americans), we could reach a total addressable market of around 275B USD. But including the potential for a fully decentralized global financial system pushes the total addressable market well into the trillion dollar mark.

1.1.2 The Augur product

The product developed by the team at the Forecast Foundation has been celebrated at its launch for its novel use of the blockchain, but was quickly abandoned by users for its complicated UX and lack of prediction markets (more on that in the Ecosystem analysis below). Compared to its most direct centralized alternative in the political betting market, PredictIt.org, Augur appears as less organized, mostly because PreditIt only takes suggestions for a market from users, while Augur allows any user to create their market, leading to an uneven quality of question asked and markets categories.

A directly related impact on this feature allowing any user to create its own market is the potential malicious use of this capability: a question which is not straightforward and with a clear event resolution trigger (date, details, resolution source, etc.) can lead to controversial outcomes and settlements, which in turn slow down the integrality of the platform and can ultimately cause a fork in the system.

This is exactly what happened with the now famous question “Which party will control the House after 2018 U.S. Midterm Election?” to which the designated reporter (and also market creator) user Poyo-Poyo on Reddit replied with the Republicans. At the close of the market, on December 10th, 2018, the Republicans were still technically controlling the House, while the Democrats would only take control in January 2019. 90% of the traders lost on that occasion, with a combined sum of about 750M USD. That caused a stir in the Augur community, and opened the gate for malicious users to intentionally pose vague questions when creating a market.

Augur has since added a warning on its platform, but hasn’t really fixed the issue, which is in fact inherent to the model.

A validity bond (the amount staked by the market creator that is seized if the market created is marked invalid) was supposed to prevent that kind of behaviour, but due to a bug on the chain, it was set at a rate way too low to deter malicious market creators from creating bogus questions. According to Joey Krug, that issue should be fixed in the V2. If the validity bond actually increases with the number of invalid markets, at some point launching a “design flaw attack” will not be economically viable for the attacker. The validity bond has increased during the last 12 months, which is a positive sign for the ecosystem, but is still too low to create a clear incentive.

Another issue directly linked to the open crowd-sourcing of markets is the possibility for “assassination markets”: for example, by creating a market on whether politician X dies before the end of their incumbency and staking an important sum of ETH on “no,” someone could effectively put a bounty on the politician’s head. Wagering against the massive “no” market and then contributing in real life to a “yes” outcome could lead to a windfall.

In these cases, Augur doesn’t really offer practical solutions except trusting that its community will wage a war on malicious users trying to game the protocol and deteriorate the ecosystem by causing invalid markets, delayed resolutions, recurring disputes and potential forks.

1.1.3 Competitors Landscape

Augur is currently facing 2 types of competitors:

  1. centralized betting and prediction platforms
  2. decentralized betting and prediction platforms

On the centralized (or classic) betting and prediction platforms, since the May 2018 Supreme Court decision to overturn a 1992 ruling that had prohibited states from legalizing sports betting, 7 states have legalized it and 14 additional states have proposed legislation to legalize sports bets. From traditional casino like Caesar’s, the Sands, Crown Resorts to online sportsbooks like DraftKings, ZenSports, Betstar, and FanDuel, all the big players are taking position right now to take advantage of this new market. A huge number of international competitors focused on the sports betting business is also taking notice. To name the most relevant: Pinnacle, 5Dimes, William Hill, BetOnline, Bet365, Bovada, etc.

DraftKings alone rose more than 1.1B USD in venture capital, fueling the growth and expansion of the king of online sport betting, which has now more than 10M active users.

On a smaller scale and other market, the previously mentioned PredictIt is also a strong alternative to Augur, with more than 420k USD of VC funding, above 300 millions shares traded by PredictIt users in 2017 and around 80,000 active users.

On the decentralized betting platforms, Augur isn’t the only player trying to attract users. While some platforms are not decentralized, but do accept cryptocurrencies for the bets (Nitrogen Sports, SportsBet, FortuneJack, Betcoin, etc.), we now see new decentralized projects aiming at the same market.

Gnosis (GNO) is a good example of that, and is a real competitor for Augur. However, Gnosis still haven’t achieved a fully decentralized oracle, which Augur claims it has. As Gnosis, Stox, another competitor, is also based on the Ethereum blockchain, but aside from receiving an endorsement from Floyd Mayweather, they haven’t achieved much since its ICO in 2017.

Delphy and Bodhi are two other projects more focused on the Chinese market. Bodhi is built on the Qtum blockchain, which could give it an additional scalability compared to other projects built on the Ethereum blockchain.

Hivemind is another project which is more focused on the governance aspect. It is designed as a sidechain of Bitcoin. This project seeks to reduce issues in multi-factor decision making, and has already provided valuable insights into the capabilities of predictive markets in the realm of governance. In Hivemind the events must be resolved by voters and the decisions being voted on are either Boolean or scalar. Voters come to agreement on the decisions using the Hivemind VoteCoins, which is not unlike the process used by Augur. Also similar to Augur, voters are punished financially for inaccurate reporting, but rewarded for accurate results.

Although Augur is well positioned in the decentralized space, having raised the most important ICO, it is dwarfed by any usual player in the centralized space. If these players decide to invest in developing their own peer-to-peer decentralized protocols, Augur will need to find strong competitive advantages and/or clear feature differentiation if it wants to remain relevant.

1.2 Ecosystem Development

If Augur, despite its numerous shortcomings, has been able to continue growing, it’s mainly because of its active community, that is invested in product development as well as a source of innovation within the ecosystem.

1.2.1 dApps

Augur has attracted developers who built some value-adding dApps on top of Augur’s prediction protocol:

Overlays (public-facing UIs that let people trade in and interact with Augur markets)

Explorer & Analytics (tools to browse, explore, and visualize data on Augur activity)

Others

  • STLD (https://stld.exchange/) — instant settlement for Augur markets, letting traders cash out their positions as soon as markets enter reporting. Will be active in V2
  • PdotIndex (https://www.pdotindex.com/) — price-weighted indices that track the success of public figures like Ariana Grande and Lebron James. Each index is composed of Augur markets regarding those figures.

Although these projects can occasionally look good and give a sense of a strong community, the fact that Veil decided to call it quits speaks loud on the fundamental lack of users of the platform. Some projects like Guesser and STLD actually answer real issues of the system (respectively the bad UX and slow settlement process) and could bring in new users if they continue to innovate and respond to the users’ needs.

1.2.2 Users, Engagement and Community

This is where the Augur project shows its main weakness.

The number of users of Augur reached a peak at 265 users on July 11th, 2018, but quickly crashed to the levels where it sits right now: 18 users on the last 24h, and 86 during the last 7 days. Except from occasional active users spikes which are closely tied to real-world events such as the vote for the US House of Representatives, this user count is pretty low, even for a decentralized app. Considering that Augur has a market cap of around 120M USD, each daily active user is basically valued at 4.4M USD…

The lack of user has a direct effect on the platform: with low user count, the economic incentives of creating a market and reporting to resolve it are actually close to inexistant. This is the classic double-sided market problem, in which Augur needs prediction markets to be attractive to users, and needs users to create economically viable prediction markets.

Augur had a total of 2,625 markets created, of which only 26 are currently considered as liquid. When considering the Open Interest in Augur markets (Total Money at Stake), Predictions.Global gives us 3.2M USD, which is just not possible given the number of active markets, as the maximum Open Interest for a market is 31,048.70 USD, with a rapidly decreasing open interest for other markets. A quick calculation gives us an open interest in the range of 150,000 USD, which is more in line with what is reported by crystalball.be. This is a 90% fall from the peaks of open interest in 2018.

Another worrying sign of a low engagement of users is the evolution of the number of created markets. As can be observed on the graph below, the speed of creation of new markets is clearly decreasing over time.

To try and onboard more users, Augur just revealed its Affiliates program, where any user who brings new users to the platform (who end up trading on any Augur markets) earns a slice of fees. From the Blueprint posted last week, this looks like a classic affiliates structure, but as no precise fees are disclosed, we’ll need to wait until V2 to see the real impact of this new incentive for existing users.

On the developer’s community side, things are a bit more hopeful as the GitHub of Augur ranks 9th overall in the most active crypto projects according to coincodecap.com (a jump from just a month ago when it was sitting at the 20th place), with a total of 32,069 commits and 72 contributors. That being said, the number of commits per week is down to around 100 on average during the last month from a peak at 215 a year ago.

What is really positive about the Augur community is the activity in their Discord group where a lot of interactions can be observed between developers and users/traders.

On social media, the Augur ecosystem has a decent presence compared to other crypto project with a cumulative follower counts of more than 162k across Facebook, Twitter, Medium, LinkedIn and Reddit. The dynamism of these social accounts has clear benefits for the overall ecosystem, as it is there that most of the announcements and discussions about new features are made.

Speaking about announcements, although the Augur development team posts Weekly Updates on their Medium page, these updates are particularly dry (some metrics and articles) and don’t really help users get a hold on the compliance with the proposed roadmap.

As for the governance, the development of Augur platform is supported by bounties through The Forecast Foundation. This is a non-profit entity set up by the founders for research. The foundation retained 4% of the total 11M tokens.

1.3 Core Team

The core team from Augur is formed by the two co-founders Joey Krug and Jack Peterson. On LinkedIn, 21 people are linked to the company, with 17 indicating that they are still working on Augur and/or the Forecast Foundation.

Among the co-founders, only Joey Krug still commits to Augur on GitHub as he is one of the core developers of the protocol. He’s probably the most publicly active member of the Augur team. He joined Pantera Capital as a co-chief investment officer in June 2017, which apparently allowed him to invest some additional resources in Augur: “The #1 reason for startup failure is due to lack of product-market fit, which Augur will never achieve if it is not liquid. By partnering with Pantera, I can help make a big dent in this problem.” — Joey Krug on joining Pantera

Apart from this, Joey is also an investor in AngelList Syndicate and co-founder at Beam.io.

Jack Peterson is the co-founder and the lead developer at Augur. He has a PhD in Biophysics from the University of California.

Among their advisors is Vitalik Buterin, who said of Augur: “I think Augur is one of the most promising applications of blockchain technology out there; prediction markets have the potential to serve a very valuable purpose in society, and be a key ingredient in a very wide array of applications, and Augur is well-positioned to take that niche.”

The overall team seems pretty light and hasn’t grown a lot since the launch of the mainnet, but again, as the GitHub activity indicates, they can count on the support of the developer community to help them materialize their promises.

However, it has lost valuable members and isn’t exempt of instability: in April 2018, Matthew Liston, former CEO of Augur, filed a $152 million lawsuit against his three other team members, alleging that Jack Peterson, Joseph Charles (Joey) Krug, and Jeremy Gardner committed acts of fraud, oppression, and malice along with investor Joseph Ball Costello. According to Forbes, though the court records suggest that the case was dismissed and the parties have settled the dispute, many details remain unknown, and could potentially threaten the company in the future if they were to resurface.

Part 2 — Technology Case

2.1 Underlying Technology & Design Flaws

The Forecast Foundation team spent more than 5 years developing the decentralized oracle for Augur, but it appears a lot is still to be resolved, in particular on the risks and attacks mitigation part.

As seen before, Augur already suffered multiple “design flaws attacks” that look to take advantage of some mechanism inherent to the protocol.

  • Relationship with ETH

Being built on the ETH protocol allows Augur for more security, but makes it also dependant on the scalability of Ethereum. Unless Ethereum’s ability to handle concurrent transactions increases multi-fold, it will be difficult (and/or extremely costly) for Augur to become a consumer-facing app with massive user and transaction count.

Right now, ETH is preferred for the wager to take place on the Augur platform, and the REP token is not used as much. That is partly due to the fact that the reporting fee rate (which pays reporters in REP) has been fixed to the lowest value possible since November 2018 (0.01ETH). The goal of that was to lower the costs associated with a market resolution in order to get higher user engagement. As the engagement has not exploded during the last year, the reporting fee rate is still way too low for any user to be economically motivated to report on an event.

For example, in a simulation using Predictions.Global tool, we see that with 50 REP (or about 550 USD), the total profit is still negative.

Although having a rock-bottom reporting fee wards off parasitic markets, it doesn’t increase the velocity of the REP, and in fact favors the use of ETH across the platform.

  • Parasitic markets

Speaking about parasitic markets, right now Augur is not at risk, just because of the low level of users and total open interest. But if at some point in the future Augur’s prediction markets get enough engagement, parasitic markets (markets that don’t pay reporting fees to Augur, but still use Augur’s resolutions) will be able to offer lower fees as they don’t have to pay the reporting fees. With the current fees that low, Augur is actually not economically viable, and reporters are not economically incentivized to do the reporting, but at least parasitic markets are not an issue.

  • Open interest volatility

According to the whitepaper, large and sudden increases in open interest (for example during a popular sporting event) results in rapid increases in the market cap requirement for forking protocol integrity. If that market cap requirement surpasses the market cap of REP, attackers have an economic incentive to cause a fork to resolve incorrectly, sending the whole ecosystem into disarray.

That issue is not really solved in the protocol, and the ecosystem will have to rely on the reactivity of its traders who, seeing the increase in open interest, will buy REP, thus driving the market cap higher and “maybe” preventing the malicious fork to happen.

  • Uncertain fork participation

That point is one of the most overlooked parts in the whitepaper: forks have a huge disruptive power to the good behavior of the entire ecosystem. So they need to be the rarest possible, and in the case where one happens, Augur will need to count on the availability of every token holder to stack their token in the “true” universe. Augur developers hope that at least 20% of all REP will migrate to the “true” child universe during the forking period of a fork, but they have no way of guaranteeing it. If a token holder migrates its token to any universe other that the consensus universe, it may lose 100% of the value of its REP. But without knowing with absolute certainty which universe will be the “true” one, all users are exposed to risk of electing the wrong universe and losing their tokens. That risk will discourage participation in case of a dubious market where no outcome is clear. Of course, Augur offers a 5% REP premium to each user that stacks its token to the child universe during the first 60 days, but that may not be a high enough reward seeing the risk that this user faces.

In the case of a really low participation, the entire ecosystem is at risk of a malicious fork, which would place the ecosystem in a “false” universe, where tokens still have value despite being in observable contradiction with the real world.

2.2 Roadmap

Seeing all these shortcomings, the release of Augur V2 in September will be an important milestone for the ecosystem, as it should resolve some of the aforementioned problems.

  • Shorter resolution times & instant settlement

V2 will implement:

- 24 Hour Designated Reporting Window (as opposed to 72h in V1)
- 24 Hour Initial Dispute Window and Immediate Dispute (as opposed to having to wait up to 7 days as in V1)
- instant settlement using STLD for example

  • Less ambiguity in market question

In V2, markets should have stronger terms due to better guidance and a robust template system for market creators, along with stronger incentives to create valid markets, answering to one of the most important issues of the platform in the last year.

  • Malicious forking still an issue

The V2 doesn’t seem to address the problem of malicious forking.

Conclusion

While positioned in a really interesting market with a lot of growth potential, and supported by a strong and active community, Augur still isn’t offering a comprehensive and convincing solution. Thus, all hopes are set on the V2, which should address part of the limitations we identified.
Only a long-term growth in terms of users and activity (market creations and resolutions) will be able to bring a sustained demand for the REP token. In the case where we observe an important growth in the ecosystem, the resolution fee will also probably get an increase, which will also be beneficial for the demand in REP, fostering a positive feedback loop.

Blockchain
Token
Analysis

--

--

Cyprien

Written by Cyprien

199 Followers

cofounder · product x crypto Beem 🍿 // token engineering 🛠 // ultrarunning 🏔

Help

Status

About

Careers

Press

Blog

Privacy

Terms

Text to speech

Teams