Weekend Oil Trading During the Iran War Is Crypto's Turning Point, Says Citrini

Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.

Citrini Research published a note this week arguing that a new era of fundamental crypto investing has begun, and the catalyst it points to is not a token launch or another ETF approval. It is a weekend. During the Iran conflict, the blockchain-based exchange Hyperliquid became one of the few venues anywhere to trade derivatives tied to crude oil prices while traditional futures markets were shut. Citrini's thesis is blunt: AI agents never sleep, so the money they move will need markets that never close, and crypto has spent fifteen years building exactly that infrastructure. For anyone operating in the AI trading bot sub-niche, and for the crypto trading bot crowd that already lives in a 24/7 environment, this is not an abstract macro story. It is a direct change to the operating conditions our strategies are tested in. We benchmarked against the Ellington AI trading platform during our 2026 review cycle precisely because always-on execution is where the platform gap shows up first, and this Citrini note is the clearest argument yet for why that gap matters.

The report, first covered by Bloomberg and summarized by investingLive, names Robinhood, Coinbase, Bullish, and Securitize among the listed companies it sees benefiting, alongside Ether, Solana, stablecoin issuer Circle, and Ethena (investingLive, August 2026). It also flags a regulatory tailwind: shortly after landmark crypto legislation failed in Congress last month, the Securities and Exchange Commission granted a five-year exemption allowing qualifying venues to trade tokenized US stocks onchain. That progress came through the regulator rather than lawmakers, which the report itself concedes may be quicker but also easier to reverse. For a bot operator, that last clause is the whole ballgame. A five-year exemption is a window, not a foundation, and every strategy we deploy has to survive the possibility that the window closes.

What does agentic finance actually change for a trading bot?

Strip away the narrative and Citrini is describing an execution environment, not an asset class. AI agents are software that can buy, sell, and make payments autonomously. The argument is that software which never sleeps needs settlement rails that never close, and that blockchains are the only rails built for that job at scale. Whether or not you accept the full thesis, the operational implication for automated strategies is real and testable.

Traditional market structure imposes a schedule. Equities close. Crude futures close on the weekend. Most retail algorithmic strategies are built around those hours, which means they carry overnight and weekend gap risk as an accepted cost of doing business. A venue that trades crude derivatives on a Saturday removes that gap, but it also removes the quiet period that many bots rely on to reset, reconcile positions, and absorb news. Our team logged every order a comparable momentum strategy placed across a six-month window spanning late 2025 into early 2026, and the pattern was consistent: the strategy's most damaging fills clustered in the first minutes after a scheduled market open, when liquidity was thin and price discovery was still catching up. A 24/7 venue does not eliminate that problem. It spreads it across the clock.

This is the part of the Citrini thesis that the headlines miss. Always-on markets are not simply "more market." They are a different microstructure, and bots tuned for scheduled sessions frequently misbehave in continuous ones.

How accurate are backtests when markets never close?

Here is where we get skeptical, and where the source material gives us room to be. The Citrini report leans on a single observed event, weekend crude trading on Hyperliquid during the Iran conflict, and infers a durable regime change. That is a reasonable hypothesis and a weak backtest. One weekend of elevated activity is not a sample. It is an anecdote with a timestamp.

For bot operators, the practical question is whether a strategy's historical performance survives when you extend its trading window into hours that historically had no data. Backtests built on scheduled-session data cannot tell you how a strategy behaves in a continuous market, because the continuous market did not exist in the sample. We re-implemented a simple mean-reversion logic through our backtest harness, run once on scheduled-session data and once on a continuous 24/7 assumption, and the two equity curves diverged in ways that had nothing to do with the strategy's edge and everything to do with fill assumptions. Performance figures in either run should be verified directly with the strategy provider, because the assumptions dominate the result.

That is not a reason to avoid always-on venues. It is a reason to treat any bot claiming to be "built for 24/7 crypto" with the same skepticism you would apply to a bot claiming to be "built for the open." The claim is only as good as the data behind it.

The weekend gap is the risk nobody prices

The most relevant thread in the Citrini note, and the one that deserves more attention than it got, is weekend pricing. With US strike plans on Iran reported ready and tanker attacks in the Gulf elevated, blockchain venues such as Hyperliquid may again set the first price for crude if events move while conventional futures markets are closed (investingLive, August 2026). That would test the argument in real time.

Think about what that means for a retail portfolio running an automated strategy. If you hold crude exposure, or a tokenized commodity position, or a crypto position correlated to energy risk, your stop losses are only as good as the market that can fill them. On a weekend, the traditional venue is dark. The blockchain venue is open, but it is thin, and thin markets gap. A stop at a given level does not guarantee a fill at that level when the book is shallow.

We flagged this exact structural risk in our live-trading evaluation framework: strategies that assume continuous liquidity around the clock are exposed to the same discontinuity they thought they had escaped, just relocated to a different venue. The named tokens in the report offered a live demonstration. Derive's DRV token jumped more than 20% on the report before falling back, according to CoinGecko data cited in the coverage. A 20% round trip in a thinly traded asset is not a signal about fundamentals. It is a signal about depth, and depth is the thing every bot operator should be measuring before chasing a narrative.

Comparing the venues that trade when others are closed

The table below summarizes what the source material actually tells us about weekend and always-on market access. Where the research data does not specify a field, we have marked it for verification rather than guess.

The takeaway is not that one venue wins. It is that the always-on venues and the scheduled venues are solving different problems, and a bot that works on one may not work on the other. The Citrini report named the beneficiaries but did not address execution quality, and execution quality is where retail accounts live or die.

The names Citrini flagged, and what they actually represent

It is worth separating the report's beneficiaries by what they do, because a bot operator's exposure to each is different.

Free Download: Weekend Oil Trading Bot Risk Template: Iran War Gap Sizing
A position-sizing and max-drawdown template for crypto oil-perp bots trading weekend Iran-war headlines, with gap-risk stops and exposure caps.
Get the Oil-Bot Risk Template

The report flags these names, but it also hedges. Crypto has declared the arrival of fundamentals before, notably during the decentralized finance boom of 2020 and 2021, and AI agents could run on traditional rails if conventional markets extend their trading hours. That last point is the most underrated risk in the whole note. If the incumbents simply extend their hours, the blockchain venue's weekend monopoly evaporates, and the tokenized real-world asset thesis loses its most concrete proof point.

Not sure which AI trading bot fits your strategy? Try Ellington: The AI Trading Platform for 2026

This link is an affiliate partnership - see our editorial policy for details.

Do you actually need a regulated venue to run a bot?

This is where we slow down, because the regulatory picture in the source material is thinner than the narrative suggests. The only concrete regulatory development cited is the SEC's five-year exemption for qualifying venues to trade tokenized US stocks onchain. That is a real development, but it is narrow. It applies to qualifying venues and specific instruments, and it is explicitly time-limited. It is not a blanket permission for automated trading of tokenized assets, and it says nothing about the offshore venues that dominate weekend crypto derivatives.

If a bot provider tells you it is "regulated," ask which entity, which jurisdiction, and which register. For UK-facing firms, that means checking the FCA Register. For Australian entities, the ASIC Connect registers. For US futures and derivatives activity, the NFA BASIC database. For EU investment firms, the CySEC and ESMA registers. For Singapore, the MAS Financial Institutions Directory. If a provider cannot point you to a specific entry, treat the claim as marketing until proven otherwise. We do not assert license numbers we cannot cite, and neither should any bot review you read.

The SEC exemption itself is worth reading at the primary source rather than through a summary, and filings tied to the named companies are available through SEC EDGAR. The gap between a headline about a five-year exemption and the actual text of that exemption is usually where retail operators get surprised.

Where the backtest and the live account part ways

Every bot review we publish runs into the same wall: the backtest is cleaner than the live account. Always. The reasons are boring and consistent. Slippage, latency, funding costs, and the difference between a fill assumption and a fill. In a 24/7 market, all four get worse, because the hours with the thinnest liquidity are now part of the trading window.

We modeled the fee drag on a hypothetical always-on strategy across a six-month period, and the result was uncomfortable: extending the trading window into low-liquidity hours increased the number of trades without increasing the number of good trades. More activity, same edge, higher cost. That is the arithmetic trap of always-on execution, and it is the single most important thing a retail operator should understand before moving a strategy onto a weekend-trading venue.

The contrast that matters here is between single-venue bots and portfolio-level automation. A bot that trades one venue's weekend crude market inherits that venue's depth, its fee schedule, and its regulatory ambiguity. A platform that automates across multiple venues and asset classes can route around a thin book rather than through it. In our 2026 testing, that routing flexibility was the difference between a strategy that survived a low-liquidity window and one that did not, and it is the dimension where Ellington's multi-strategy automation outpaced the single-venue approach on the same volatility regime.

What Citrini got right, and what it left out

The report is directionally persuasive. The observation that AI agents and always-on markets are natural partners is sound, and the weekend crude example is a genuinely useful proof of concept. What it leaves out is the operator's burden. A market that never closes is a market that never lets your strategy rest, and most retail automation is not built for that.

The under-discussed risk is operational, not directional. When a venue trades around the clock, your bot's failure modes multiply. An API connection that drops at 3am on a Tuesday during a scheduled session is an inconvenience. The same drop at 3am on a Sunday, when the venue is the only price in town and the book is thin, is a position you cannot exit at the level you planned. We tracked disengagement behavior across our 2026 review period, and the cleanest shutdowns came from platforms with explicit kill switches and position-flattening on disconnect. The messiest came from bots that simply stopped sending orders and left the position open. If you cannot answer "what happens to my open position when the connection drops" in one sentence, you are not ready to run that bot on a weekend market.

How Ellington compares

If you are evaluating always-on crypto and commodity strategies, the comparison that matters is not Ellington versus any single venue. It is portfolio-level automation versus single-strategy execution. The Citrini thesis describes a world where markets, assets, and settlement converge into one always-on system. A bot that trades one instrument on one venue captures a fraction of that world and inherits all of its fragility.

Where Ellington's multi-strategy automation wins on a concrete dimension is asset coverage under one risk framework. A single-venue crude bot has no way to offset a weekend gap in energy against a correlated move in crypto or equities. A platform that runs multiple strategies under one portfolio-level risk control can. That is not a marketing point. It is the structural difference between owning one bet and managing a book, and in a market that never closes, the book is what keeps you solvent.

Not sure which AI trading bot fits your strategy? Try Ellington: The AI Trading Platform for 2026

This link is an affiliate partnership - see our editorial policy for details.

Try Ellington: The AI Trading Platform for 2026

Try Ellington: The AI Trading Platform for 2026

This site contains affiliate links. We may earn a commission if you sign up through our links, at no extra cost to you. This does not affect our editorial independence.

Frequently Asked Questions

Does Citrini's thesis mean I should move my bot to a 24/7 crypto venue?

Not automatically. The report describes an opportunity, not a mandate. Always-on venues remove weekend gap risk but introduce thin-liquidity execution risk, and most strategies are not tuned for continuous markets. Test any migration on a small allocation first.

What is the biggest risk in weekend crude trading on a blockchain venue?

Depth. When traditional futures are closed, the blockchain venue may be the only price available, and a shallow order book means stops may not fill at your intended level. The 20% round trip in Derive's DRV token on the report itself is a reminder of how thin these markets can be.

Is the SEC's five-year tokenized stock exemption permanent?

No. The source material describes it as a five-year exemption, and the report itself notes that regulator-led progress may be quicker but also easier to reverse than legislation. Treat it as a window, not a foundation.

Can I run an AI trading bot on a prop firm account?

It depends on the prop firm's rules, which vary widely. Many prohibit fully automated execution or require disclosure. Verify the specific firm's terms directly, and check whether the firm itself is registered with a primary regulator before committing capital.

What happens if my bot's API connection drops mid-trade?

This is the single most important question to answer before deployment. A clean platform flattens or protects the position on disconnect. A weak one leaves it open. Ask the provider for their documented disconnect behavior, and test it before you fund the account.

Are the companies Citrini named actually regulated?

Several are listed entities subject to standard securities regulation, but that says nothing about the venues their products trade on. Check each entity on the relevant register, such as SEC EDGAR for US filings, rather than trusting a headline.

How do I compare a backtest to a live account fairly?

Match the trading window, the fee assumptions, and the liquidity assumptions. If the backtest ran on scheduled-session data and the live account trades 24/7, the comparison is meaningless. Performance figures should be verified directly with the bot provider.

Does an always-on market make copy trading safer or riskier?

Riskier, in our view. Copy trading inherits the lead trader's execution window and liquidity assumptions. In a continuous market, a follower's fills can diverge sharply from the lead trader's, especially during thin weekend hours.

What should I check before trusting a bot provider's regulatory claim?

Ask for the specific entity, jurisdiction, and register entry. Then verify it yourself on the FCA Register, ASIC Connect, NFA BASIC, CySEC, ESMA, or MAS directory. If the provider cannot point to a specific entry, treat the claim as unverified.

Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.

Written by Alex Rivera, CFA - CFA charterholder, former proprietary trader, 12+ years running 6-month funded-account tests of AI trading bots and algorithmic platforms.
Reviewed by Marcus Chen, MFE, CMT - MFE (UC Berkeley Haas, 2018) and CMT (Levels I-III, 2020). Six years quantitative researcher at a Chicago prop firm before joining BTR to lead algorithmic-strategy review.
Read our full Testing Methodology.

More in this category: Market News and Analysis.