WTI crude oil pushed higher on Friday, trading near $103.95 a barrel, up about 1.9% on the day. That is its highest level since mid-September, when it touched a four-month peak of $105.83. Oil is on course for a gain of roughly 15% in October.
The latest rise followed a Wall Street Journal report that the top US commander in the Middle East had briefed President Trump on military options against Iran, as Washington considers restarting major combat operations. Trump has warned that strikes on Tehran could resume after next week's US midterm elections.
There were other worries for supply too. Houthi forces in Yemen fired a wave of missiles at Saudi oil company Aramco on Thursday, and two separate tanker incidents raised tension around the Strait of Hormuz, a key route for Gulf oil. Ukraine was also reported to have hit a Russian refinery in the Samara region.
US crude oil (WTI): live daily chartDaily candles, last 3 months
Live prices from TradingView. They keep moving, so they can differ from the figures in the story, which were correct when it was written.
Signs of diplomacy have not calmed the market. Iran confirmed it received a letter from US Vice President JD Vance aimed at restarting peace talks, but said it would not soften its position. A day earlier, oil had briefly slipped below $100 on hopes for those contacts.
With this the last weekend before the US vote, some traders appear to be buying protection against a possible attack before markets close.
What it means for traders
Weekend news on Iran could cause sharp gaps when oil markets reopen on Monday, in either direction. Higher oil also feeds into inflation worries, which can move energy-linked currencies such as the Canadian dollar and Norwegian krone, and bond yields.
Risk warning: Trading foreign exchange and CFDs on margin carries a high level of risk and may not be suitable for all investors. Leverage can work against you. Content on FXROLL is for information and education only and is not investment advice. Prices and data may be delayed or contain errors.