Where does Pyth fit in?

For my very first story, I examined Broadridge’s bullish use of blockchain in the repo market. Then at the start of this year, my colleague Rebecca Natale and I set out to revisit the initial blockchain mania (2015–2018: gone but not forgotten) and answer the question: Where have all the blockchain startups gone?

Our answer came in a mixed bag. Some firms had found the technology didn’t suit their use cases or regulatory mandates, and they had moved back to more traditional tech. Others were acquired by larger firms, shifted to crypto or retail business lines, or in one case, entirely disappeared.

Although prominent examples of blockchain viability suffered a major blow recently when the Australian Securities Exchange (ASX) would take a pre-tax write-off of AUS$245 million to AUS$255 million ($165 million to $170 million) after halting its Chess replacement project, blockchain persists, even if those halcyon days of the late teens are dissipating.

It’s in this environment that saw the launch last summer of the Pyth Network, a peer-to-peer network and decentralized oracle—a device or entity that connects a blockchain with off-chain data—that currently has more than 70 data contributors supplying data across multiple asset classes. The majority of the network’s contributors or publishers are in the crypto space only, but it has also managed to entice some familiar capital markets names, such as Jane Street, Jump Trading, Susquehanna, and Cboe Global Markets.

In October, Cboe joined Pyth, becoming the largest exchange group to join the network. Catherine Clay, executive vice president and global head of data and access solutions at Cboe, today, says the exchange’s participation fits in well with its ethos of data accessibility. “Pyth is just another extension of our distribution of market data, albeit in a decentralized marketplace,” Clay tells WatersTechnology.

In our view, action is information. This is almost an R&D effort for us to really explore what decentralized finance looks like, because we know exactly what it means to be a global powerhouse in traditional finance

Catherine Clay, Cboe Global Markets

Following Cboe’s acquisitions in Canada, Japan, and Australia, Clay says the exchange found itself with a huge breadth of market data, particularly in equities. “We’re confident that there are users and consumers of Pyth who aren’t necessarily accessing traditional markets but want to have high-integrity market data to view,” she says.

But why would a highly regulated, centralized exchange wade into decentralized waters? Clay says participation in Pyth falls in line with the exchange’s mandate to bring its market data to more viewers and users. Data accessed through Pyth is free.

Cboe joins Investors Exchange (IEX), Members Exchange (Memx), and Bermuda Stock Exchange (BSX), a subsidiary of Miami International Securities Exchange (Miax), in supplying equities market data to Pyth. Memx and Miax declined to comment on their participation in Pyth, and IEX did not return requests for comment.

Last summer, my colleagues Joanna Wright and Wei-Shen Wong wrote about efforts among exchanges to seek tighter relationships with their customers to understand how their market data was being used. Exchanges have long relied on intermediary vendors like Bloomberg and Refinitiv to facilitate these relationships, but the tables appear to be turning. Pyth may be the newest iteration of that same spirit, says Tim Baker, founder of consultancy Blue Sky Thinking and the former head of IEX Cloud until late 2021.

“What’s interesting about this is that it’s a mechanism for exchanges and other data sources to deliver data in a much more federated model, so they don’t have to go through an aggregator,” Baker says. “Right now, most exchange data goes through aggregators or specialized firms that aggregate the data together and then redistribute it out to consumers. I think there’s general interest in the exchange community to be able to bypass those intermediaries and deliver data directly to consumers.”

Stephen Kaminsky, who works in special projects and strategic initiatives for Jump Crypto, a Pyth contributor, says traditional exchange participation in Pyth is valuable to the network.

“These exchanges, like Cboe, Memx, IEX, and Miax in the equities space, along with LMAX in FX and crypto, have been super valuable to the strength of the network,” he says. “These exchanges have unique and first-party data that is critically important to the robustness of Pyth. As the gap between traditional finance and on-chain development continues to lessen, we certainly expect more traditional exchanges to participate on Pyth in various capacities.”

The prices listed on Pyth and on-chain are in aggregate, Kaminsky says, so it’s just as important that data sources are many as it is that they’re varied and diversified. For example, 20 publishers may be contributing a price for AAPL (Apple Inc.) directly on-chain, and that data would be informed and input by hedge funds and prop trading shops such as Jane Street, Jump Trading, and Two Sigma, but also exchanges like Memx and IEX.

As Pyth continues to scale across symbols and asset classes, having consistent and additional data provider participation will be critically important to the success of the network, Kaminsky adds.

And because time is a continuous loop, there’s always fear of missing out (Fomo), which fueled the original blockchain mania. The Fomo could entice more large exchanges into publishing on Pyth, Baker says. And as data democratization and accessibility become more relevant issues to the professionals and the public, exchanges may end up on Pyth or experimenting with other oracles—even if such projects aren’t necessarily put into production.

“With Pyth, we are talking about the decentralized world of smart contracts, right? This is a very unknown space; there are more questions about decentralized finance (DeFi) than there are answers. But in our view, action is information,” says Cboe’s Clay. “This is almost an R&D effort for us to really explore what decentralized finance looks like, because we know exactly what it means to be a global powerhouse in traditional finance.”

Going to the (data) marketplace

Interestingly, the Pyth Network and its on-chain data distribution model represents an alternative to the young but established data marketplace model, through which providers offer (usually cloud-based) “storefronts” for users to browse and purchase datasets from throughout their own large organizations and often from smaller, third-party vendors looking to get more exposure.

S&P Global Market Intelligence unveiled its simply titled Marketplace in 2020. Last year, Amazon Web Services (AWS) launched Data Exchange in a similar vein, with FactSet signing up its data to the platform. And fintech newcomer Snowflake has recruited more than 240 third-party data and data service providers to its own Snowflake Data Marketplace.

“I think the exchanges will continue to invest in their own data distribution capabilities, into the cloud, into Snowflake, into all these other different mechanisms and marketplaces, which allow them to deliver data directly to customers,” says Blue Sky’s Baker. “The vast majority of consumers of that data are not using blockchain; their institutions aren’t running their business on chain. This is really a subset of a subset of the market.”

It is still early in the Pyth journey and the broader development of crypto technology, so there could be second-order effects; for example, certain off-chain applications like data aggregators or analytics platforms may find interesting use cases for Pyth

Stephen Kaminsky, Jump Crypto

But as exchanges continue to improve APIs, market direct feeds and build out data businesses, these advancements could naturally lead into more unconventional fields like data oracles—if only for the sake of trying, such as in Cboe’s case.

Kaminsky says Pyth doesn’t see itself competing with names like Snowflake, Databricks, and other marketplaces directly, but that it may offer interesting use-cases for data purposes outside of low-latency market data. “It is still early in the Pyth journey and the broader development of crypto technology, so there could be second-order effects; for example, certain off-chain applications like data aggregators or analytics platforms may find interesting use cases for Pyth,” he says.

And that could also be tied to speed. A Pyth official, speaking on background last December, told Max Bowie that the Solana blockchain platform that underlies Pyth was fast enough for many data needs but may be too slow to supplant existing data platforms entirely.

“Solana updates every 400 milliseconds or so, which is fast for a blockchain, but slow for equities markets, for example. So I don’t think it will ever completely replace the types of platforms that our participant firms currently use,” the Pyth official said.

Risk mitigation

Unfortunately for blockchain, even when it seems promising, its steps forward are often simultaneous with steps back. November proved to be a tumultuous month for the blockchain field with the implosion of cryptocurrency exchange FTX. The halting of ASX’s Chess replacement project also doesn’t bode well, and our editor Anthony Malakian has written about what lessons can be learned from the Chess fallout.

Pyth has the recent news of FTX’s downfall on its radar (FTX was a publisher on the network, as well as Alameda Research, the crypto hedge fund founded by FTX CEO Sam Bankman-Fried) but it remains optimistic. Kaminsky acknowledges recent setbacks, but says innovation and allocation are still ongoing and initiatives like Pyth help fuel the technology’s growth.

Philipe Redaelli, managing director of on-chain data at Kaiko, a cryptocurrency market data provider for institutions and a publisher on Pyth, isn’t ready to make any predictions as to how this decentralized movement fits into a still centralized world. “There are thousands of questions: How do you bring it in? Is it scalable? Are you going to replace every other computer with blockchain? I don’t have the answers for that,” he says.

Others have been consistently skeptical of blockchain and what it can provide. Virginie O’Shea, founder of Firebrand Research, points to issues around processing power and scalability in current blockchain technology as to reasons why the technology should get passed over. “You don’t actually have to have a blockchain to have DeFi right; they’re not completely interlinked,” she says. “There are decentralized database structures that don’t use blockchains, that are more efficient, [and] which I imagine may eventually get picked over blockchain. They’ll be more scalable, easier to implement, and at a lower cost.”

The flip side of the same coin is that Pyth participation represents a safer foray into blockchain. After all, it is merely an alternative form of data publishing—does it matter whether users look at it there or somewhere else?

“This is quite a good way of dipping your toe in and saying ‘We are doing blockchain, tick the box,’” Baker says, “and very often that’s the right strategy with an emerging technology like this.”

Update, December 9, 2022: This story was updated as it is BSX, not Miax directly, supplying the market data to the Pyth network.