Betting on diplomacy
How to predict the future and be ready for it
A warm welcome to the AAL Gazette. Also this week:
The Ambassador’s Guide to Risk (downloadable PDF)
Become a cyber diplomat: expert analysis and advice
5 open diplomacy learning opportunities
Bad Bunny and Olympic diplomacy
10 top diplomacy jobs, 5 internships
Betting on Diplomacy
Advice from a wise Ambassador I worked for, after I added a comment to a draft diplomatic telegram predicting an election result: “good diplomats don’t make predictions”.
As the saying goes, “never make forecasts, especially about the future”.
Instead: identify scenarios, assess risk, take mitigating action. Top 2026 risks are highlighted in the Global Risk Report, and the Gazette of 22 January. The micro-guide below incorporates steps from that edition on how to make, populate and use a risk matrix.
How to identify and assess headline risk? Whilst there’s no substitute for networks of locals, experts and national/international actors, now also available: the market.
The popularity and coverage of online prediction platforms has continued to grow since the 2024 US presidential election. Blockchain-based prediction market platform Polymarket allows users to trade on outcomes of real-world events.
Crowd-forecasting community Metaculus aggregates predictions, aiming to produce “forecasts for the public good”. In public and in real time, they invite thousands of participants to assign probabilities to election outcomes, ceasefires, sanctions, coups and diplomatic engagement.
Both have been supported by Good Judgment Project, a research initiative that aims to test and improve methods for predicting global events through crowd-sourced judgment.
By using algorithms that weight more accurate forecasters more heavily, they produce aggregate predictions that often outperform simple crowd averages.
Even talk of diplomacy can show up in data from Polymarket:
Image: will the US strike Iran before the end of February? The odds fell after Iran and the US confirmed diplomatic engagement.
At time of writing, predictions include:
A Ukraine/Russia ceasefire by end of 2026: 44%
The US Embassy in Caracas reopens by 31 March: 56%
North/South Korea direct talks before July: 25%
China invades Taiwan in 2026: 11%
None of this is to encourage betting on war or human suffering, or to ignore potential online financial manipulation.
An anonymous trader is reported to have made $400,000 betting on President Maduro’s capture “just hours” beforehand.
If you believe in the wisdom of crowds, the (fluctuating) numbers are worth taking seriously. A sudden spike in a platform’s forecast, say, for conflict escalation, can influence media framing before official statements issue. Governments may be judged by whether the market believes policies will succeed.
In some cases, prediction markets affect behaviour: actors may respond to what they see priced in.
Moreover, many of the markets give you a big enough (albeit skewed) sample to tell you something about sentiment. Will there be an Iran nuclear deal this year? Will China and Japan cut diplomatic relations?
How to use open-source probabilistic signals in risk management
Market sentiment can itself inform risk analysis and preemptive actions. The competitive advantage is not so much in the data as in integrating it strategically.
Use the 6-step PREDICT cycle, and the downloadable PDF diplomatic risk micro-guide.
