What Prediction Markets Are Forecasting

Prediction markets currently show a near-even split on whether London's high temperature will stay at or below 8°C on February 28. With about 47% of traders betting it will be 8°C or lower, this is essentially a coin flip. The market sees roughly equal odds between a chilly day and a slightly milder one. This tight spread reflects the inherent uncertainty in late-winter weather forecasting just days ahead.

Why the Market Sees It This Way

Two main factors explain the split forecast. First, late February in London is historically a transitional period. Average highs are around 9°C, but the city frequently experiences swings between lingering cold snaps and early hints of spring. The date sits right on this climatic borderline, making any single-day prediction difficult.

Second, while modern meteorological models are good, their precision decreases beyond a few days. Traders are likely weighing competing model runs and ensemble forecasts that show small variations in air mass and cloud cover. A difference of just one or two degrees, which could be swayed by a few hours of sunshine or a shift in wind direction, determines the outcome. The market's indecision mirrors the forecast models' own uncertainty at this range.

Key Dates and Events to Watch

The key period is the 72 hours leading up to February 28. Short-range weather models become significantly more reliable within this window. Specific forecasts from the UK Met Office and updates from the European Centre for Medium-Range Weather Forecasts will be the primary signals traders watch. Any consistent trend in these forecasts toward a colder or warmer air mass will likely move the market odds decisively in one direction before the day arrives.

How Reliable Are These Predictions?

For short-term weather events like this, prediction markets have a mixed but interesting record. They are excellent at aggregating available public forecast data and expert interpretation, often matching or slightly outperforming raw model outputs. However, their true strength is showing the confidence in a forecast. The current 47% probability doesn't mean meteorology is failing. It accurately communicates that, with the information available now, this specific threshold is too close to call. The market will become very reliable as the event nears and forecasts consolidate, but for now, it correctly signals genuine uncertainty.