The 5 Invisible Hands That Move Oil Prices Every Day
A refresher on what drives crude oil prices in the short term.
7 minute read
Summary
Understanding and then forecasting crude oil prices in the short term (up to 12 months ahead) requires integrating multiple market signals spanning supply-demand fundamentals, financial drivers, and sentiment indicators. Short-term oil price movements are highly sensitive to even small imbalances in supply and demand, due to the low elasticity of both – meaning minor shocks can trigger outsized volatility. Our approach combines quantitative data (e.g. production levels, consumption rates, inventory changes, currency strength) with qualitative inputs (e.g. news sentiment, geopolitical risk) to capture these dependencies and correlations. By weighting each factor according to its historical impact on prices, we obtain a formula that adjusts the baseline price for fundamentals and market sentiment. In essence, the model adds up contributions from supply tightness, demand strength, inventory swings, macro-financial conditions, and qualitative news signals to predict short-term oil price changes. The full equation below summarizes this multifactor approach, with each term detailed for its rationale and weight.
The 5 Key Indicators To Predict Price Movements
At least five core factors influence crude oil markets: demand (from OECD and non-OECD countries), supply (from OPEC and non-OPEC producers), the balance between them (often reflected in inventories), financial market conditions, geopolitical conditions and events; and current spot prices (and various types of price arbitrage e.g. between regions), among others. Building on these categories and additional research, we identify the following key factors and data sources that drive short-term oil price fluctuations:
1) Supply (Production Levels & OPEC Policy)
Global crude production is ~102 million barrels per day as of mid-2025, with OPEC+ controlling about 40% of output. OPEC’s production decisions have an immediate impact on prices, as history shows that cuts in OPEC output put upwards pressure on prices. OPEC members account for ~40% of world oil production and about 60% of internationally traded petroleum, so their coordinated supply changes strongly influence the market. Non-OPEC supply (e.g. U.S. shale) also matters, though it tends to respond more slowly to price changes. We monitor weekly production data (e.g. from U.S. EIA for U.S. output, OPEC monthly reports for member quotas) and any unexpected outages or increases. A sudden supply drop (e.g. an unplanned outage or an OPEC cut announcement) creates a tight market and upward price pressure, whereas a surge in production or lifting of export bans can push prices down. Because oil supply is fairly inelastic in the short run (it’s costly and slow to ramp production), small supply shocks can require large price moves to restore equilibrium. In energy, supply is far less predictable than demand which tends to change gradually with the one major exception in modern history being COVID lockdowns.
2) Demand (Global Consumption & Economic Activity)
World oil demand is roughly 100–103 million barrels per day in 2025, and demand growth, especially in emerging markets, directly bolsters prices. We incorporate indicators of economic activity that proxy oil demand: for example, global GDP growth rates, industrial production indices, and transportation indices. Strong economic data (high manufacturing PMI, increasing vehicle miles traveled, etc.) signal rising oil consumption, which tends to lift prices. Conversely, signs of economic slowdown or fuel substitution efforts temper demand. In the short term, demand changes are subtle day-to-day, but trend indicators and high-frequency data (like power generation fuel usage or mobility statistics) can flag shifts. Because demand for oil is also relatively inelastic (people and industries cannot quickly cut usage when price rises), a given demand shock (e.g. an unexpectedly cold winter increasing heating oil and natural gas use) can cause a sharp price response. We weight demand-side factors in our models as a crucial price driver.
3) Inventory Levels and Supply-Demand Balance
Oil inventories act as the buffer between supply and demand, so changes in stockpiles are a critical short-term indicator. A drawdown in inventories (stocks falling) means demand exceeded supply in that period – a bullish signal for prices – whereas inventory builds indicate oversupply and are bearish. Weekly inventory reports, notably the U.S. EIA Weekly Petroleum Status Report, “function as the oil market’s heartbeat monitor” by providing real-time feedback on the supply-demand balance. These reports (e.g. crude stock levels at Cushing, Oklahoma) often trigger immediate price moves: a larger-than-expected draw has an outsized positive impact on price (as seen when a recent stockdraw sent Brent prices surging to ~$79.8), while a surprise build can send prices down. Market participants closely watch and even trade ahead of inventory releases, amplifying price volatility beyond the fundamental change. Our models use inventory changes (particularly U.S. crude and gasoline stocks) as a quantitative proxy for the global supply-demand imbalance. Because short-term supply/demand data for the whole world can be lagged, inventory data serves as the high-frequency measure of balance. A term for inventory change captures this: if inventories drop, the model will lift the price forecast (and vice versa). We also account for spare production capacity (especially OPEC’s spare capacity) as a related metric – low spare capacity means the market has less cushion, so any demand uptick or outage causes a bigger price spike.
4) Financial Markets and Currency (Dollar Index)
Oil does not trade in isolation; broader financial conditions influence it. One major factor is the U.S. dollar’s strength. Crude oil is typically priced in USD globally, so a stronger dollar often pressures oil prices downward (making oil more expensive in other currencies, which can dampen demand). Empirical studies confirm that the U.S. Dollar Index has significant predictive power for oil prices. We include the inverse relationship: if USD appreciates, our formula will tend to predict a slight drop in oil price (and vice versa for USD weakening leading to higher oil price). Other financial variables include interest rates (higher rates increase storage financing costs and can reduce speculative demand for commodities) and investor flows. For instance, we monitor the net speculative length in oil futures (Commitment of Traders data) as an indicator of market sentiment and liquidity – extreme speculative long positions might signal overbought conditions, whereas an influx of investor buying can boost prices. Additionally, oil often correlates with global equity and risk sentiment; a booming stock market and low volatility environment (risk-on mood) can coincide with higher oil demand expectations, whereas a financial crisis or risk-off crash can pull oil prices down due to anticipated demand collapse. These financial market linkages are incorporated via variables like the trade-weighted USD index, interest rate trends, and even volatility indices (e.g. VIX) as needed.
5) Foreign Policy & Geopolitical Events + Qualitative Sentiment
Some of the most dramatic short-term oil price swings come from geopolitical shocks and market sentiment. Wars, conflicts, and political crises in key oil-producing regions can remove supply or create fear of future disruptions, adding a risk premium to prices. For example, recent Middle East tensions (such as the Israel–Iran flare-up) triggered a ~7% spike in Brent crude price virtually overnight. A significant chokepoint risk (e.g. threats to the Strait of Hormuz, through which ~20% of global oil passes) can add several dollars per barrel in insurance and rerouting costs – analysis shows that rerouting tankers due to Hormuz risk effectively adds a $3–$5 per barrel premium to oil prices. Our model captures geopolitics via a risk indicator. In addition to actual disruptions, news and sentiment around oil play a key role. Market expectations are often shaped by news headlines, analyst commentary, and social media. We incorporate a sentiment index derived from textual data – e.g. sentiment scores from oil-related news articles and tweets. Studies have reached a consensus that sentiment analysis improves oil price predictions, as it helps quantify the otherwise unpredictable factors. For instance, bullish news about future demand or tensions can drive prices up in advance, while bearish news about new supply or economic troubles can push prices down. By using natural language processing on news feeds and Google Trends data, one can gauge the market’s qualitative mood. Our model weights this sentiment factor such that strong positive sentiment (many bullish headlines, high investor optimism) shifts the forecast upward (but only slightly as this is a weaker indicator), whereas negative sentiment or fear will drag it downward. This qualitative component ensures the formula is not purely reactive to hard data, but can also “sense” the market psychology, which is vital in short-term forecasting. For example, unexpected policy announcements or tweets by officials can whipsaw prices even if physical balances haven’t yet changed.
There are of course many other factors that affect oil prices such as demographics, technology developments and the those found in the refining value chain, but the above are fundamental for the day-to-day. Stay tuned for how this fits in to a more comprehensive forecasting model and the added layers to navigate refined products and other commodity markets.