Subscribe to newsletter
Paying for Predictions
STRATEGY COMPETITION: RESULTS OF THE "PAYING FOR PREDICTION" ONLINE GAME
We have the winners! We received 71 submissions for the strategy competition on the game "Paying for Predictions". Given limited resources (ten beans) and some information about the risk of likely floods (determined by the roll of two dice), each player indicated how much they would invest in a bidding process to get an early warning system (which allows to see one of the dice before the potential flood), and proposed a standard operating procedure for when to take early action based on the early warning.
The game challenge was similar to what humanitarian workers confront every season in the real world: Should one spend a bean for flood preparedness? (risking "acting in vain" and wasting resources if no flood). Or is it better to wait and see? (risking "failing to act" and having to spend four beans for disaster response). A humanitarian crisis occurs if not enough beans are available to pay for disaster response.
Depending on a player's own decisions, other players' decisions, and the luck of the rains, each strategy gets good or bad outcomes. Climate change increases the risk of flood after the sixth round. The player with the most beans at the end of ten years wins one game. A computer simulation allowed us to compare the performance of the various strategies through thousands of games, capturing their total number of wins as well as their average number of humanitarian crises and remaining beans.
The 71 strategies were submitted by competitors from Argentina, Botswana, China, Finland, Guatemala, Kenya, Italy, Malawi, Namibia, The Netherlands, Nicaragua, Rwanda, South Africa, Spain, United States, and Zambia.
This is the list of top performers:
1. Javier Gómez Serrano (Instituto de Ciencias Matemáticas Madrid, Spain)
2. Mark Voorhies (Department of Microbiology and Immunology, UCSF, USA) & Satrajit Ghosh (McGovern Institute for Brain Research, MIT, USA)
3. Jezabel Curbelo (Instituto de Ciencias Matemáticas Madrid, Spain)
4. Joseph Benias (Ministry of Education, Malawi)
5. Mario Saguí (Guatemala Red Cross)
6. Jake Buzaid (Northeastern University)
7. Newton Chirambo (Department of Meteorology, Malawi)
8. José Luis Chen (Guatemala Red Cross)
9. Denis Argeñal (CARE Nicaragua)
10. Erick Chavajay (CARE Guatemala)
11. Selvin Jarquin and Nery Pérez (Caritas Guatemala)
12. Franklin Yonamu (Malawi Red Cross)
As explained in the rules of the competition, prizes will be given to the winning strategies submitted before the deadline, based on three distinct categories: Academia, Red Cross, and "other".
So these are the names of the winners:
- Javier Gómez Serrano (Instituto de Ciencias Matemáticas Madrid, Spain)
- Jezabel Curbelo (Instituto de Ciencias Matemáticas Madrid, Spain)
- Jake Buzaid (Northeastern University)
(Note: the submission by M. Voorhies and S. Ghosh was second best performer overall, but arrived after the deadline... So we hereby recognize their accomplishment - but no prize for them :-)
- Mario Saguí (Guatemala Red Cross)
- José Luis Chen (Guatemala Red Cross)
- Franklin Yonamu (Malawi Red Cross)
- Joseph Benias (Ministry of Education - Malawi)
- Newton Chirambo (Department of Meteorology - Malawi)
- Denis Argeñal (CARE Nicaragua)
There was a wide range of strategies, from extremely risk averse to remarkably risk taking, and from very simple to highly detailed. Not surprisingly, the top three strategies were based on a finely tuned risk management approach.
- Javier's strategy involved different thresholds for whether to take early action or not, based both on value of the forecast and whether or not climate change has materialized.
- Mark's strategy was derived from an online tool for collaborative problem solving built by Satrajit Ghosh: it analyzes the game outcomes based on simulations of the competition itself, thus helping any user to test and refine possible standard operating procedures, based on results. This crowdsourcing approach to linking early warnings with early actions, inspired by the book "Reinventing Discovery", can certainly help the humanitarian sector improve decision making by mobilizing the brainpower of humanity!
If you are interested in learning more about the winning strategies, please email Pablo Suarez.
On behalf of the Climate Centre, a big "Thank You" to all participants for a fun and fruitful experience. Our gratitude also to Marcos Donnantuoni for creating the computer simulation tool that enabled this competition (see his Blog for more strategy competitions), and for processing the results.
Do you like this playful approach to experiencing the complexity of future risks?
Read our serious yet fun book, "Games for a New Climate " (downloadable for free).