- Solve real problems with our hands-on interface
- Progress from basic puts and calls to advanced strategies
Interactive Options Course
Posted April 15, 2026 at 10:00 am
In my March 19, 2026, article, “Prediction markets might already be the best source for today’s weather forecast,” I presented preliminary evidence from the first week of formal tracking (March 11–17) that ForecastEx prediction markets appear to meaningfully improve daily high-temperature accuracy compared with National-Weather-Service-LAMP, the most directly comparable conventional forecast system.
I noted at the time that these were preliminary results and that I would provide regular updates as the sample grew. This article is the first update, now based on data through April 14, 2026.
Over the full period from March 11 through April 14, the dataset now includes 24 cities and 501 city-days. The lead-time summary figure (average absolute error vs. lead time) continues to show the same core structure as in the preliminary week, but with a much larger sample and thus greater confidence.
Using bins with at least N=30 city-days, ForecastEx’s weighted average absolute error is about 1.95°F, compared with 2.61°F for LAMP, a 25.0% lower error overall.
At long lead times, the two systems are closer (for example, around 24 hours out, they are near parity), but as the target day progresses, ForecastEx increasingly pulls ahead: roughly 19% lower error by 18 hours, ~25% by 12 hours, and about 55% lower error in the final 1–6 hours before end-of-day.
Figure produced with Python Matplotlib
The city-level map view (pooled over local 5:00 am–12:00 pm hours) remains broadly consistent with the first week’s data. ForecastEx now outperforms LAMP in 20 of 23 cities with valid morning-window samples, with a median improvement of about +17.2%. The strongest gains appear in cities such as Denver, Nashville, Seattle, San Francisco, and Houston, while only a small minority of cities are currently negative. As before, geographic variation remains an open question, but the dominant signal remains positive and widespread.
Figure produced with Python Matplotlib
So, what was initially a promising one-week result has held up over a larger out-of-sample period. We continue to see that ForecastEx is not merely echoing conventional forecasts but is able to aggregate and distill the best available information into more accurate same-day high-temperature forecasts than conventional methods. With nearly a month of data now in hand, the evidence that prediction markets can improve practical weather forecasting seems to be only getting stronger.
The analysis in this material is provided for information only and is not and should not be construed as an offer to sell or the solicitation of an offer to buy any security. To the extent that this material discusses general market activity, industry or sector trends or other broad-based economic or political conditions, it should not be construed as research or investment advice. To the extent that it includes references to specific securities, commodities, currencies, or other instruments, those references do not constitute a recommendation by IBKR to buy, sell or hold such investments. This material does not and is not intended to take into account the particular financial conditions, investment objectives or requirements of individual customers. Before acting on this material, you should consider whether it is suitable for your particular circumstances and, as necessary, seek professional advice.
The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Interactive Brokers, its affiliates, or its employees.
Interactive Brokers LLC is a CFTC-registered Futures Commission Merchant and a clearing member and affiliate of ForecastEx LLC (“ForecastEx”). ForecastEx is a CFTC-registered Designated Contract Market and Derivatives Clearing Organization. Interactive Brokers LLC provides access to ForecastEx forecast contracts for eligible customers. Interactive Brokers LLC does not make recommendations with respect to any products available on its platform, including those offered by ForecastEx.
Forecast Contracts are only available to eligible clients of Interactive Brokers LLC, Interactive Brokers Canada Inc., Interactive Brokers Hong Kong Limited, Interactive Brokers Ireland Limited and Interactive Brokers Singapore Pte. Ltd. Forecast Contracts on US election results are only available to eligible US residents.
Futures, event contracts and forecast contracts are not suitable for all investors. Before trading these products, please read the CFTC Risk Disclosure. For a copy visit our Warnings and Disclosures Page.
Displayed outcome information is based on current market sentiment from ForecastEx LLC, an affiliate of IB LLC. Current market sentiment for contracts may be viewed at ForecastEx at https://forecasttrader.interactivebrokers.com/en/home.php. Note: Real-time market sentiment updates are only active during exchange open trading hours. Updates to current market sentiment for overnight activity will be reflected at the open on the next trading day. This information is not intended by IBKR as an opinion or likelihood of a potential outcome.
This is commentary on economic, political and/or market conditions within the meaning of CFTC Regulation 1.71, and is not meant provide sufficient information upon which to base a decision to enter into a derivatives transaction.
Related Articles
Join The Conversation
For specific platform feedback and suggestions, please submit it directly to our team using these instructions.
If you have an account-specific question or concern, please reach out to Client Services.
We encourage you to look through our FAQs before posting. Your question may already be covered!