Judges' Notes

Rubrics
The judges will use these rubrics to evaluate your projects.


 * Finalist Judges Rubric
 * Expo Judges Rubric

Single Person Teams
This is covered in the rubrics. Single person teams get 5 out 10 points, dysfunctional multiperson teams get less than 5 points. Single person teams have to score higher in other areas to get the same results as successful multiperson teams.

Some judges want to reconsider this and give single person teams a lower score. But, we realized we had to use this year's rubric for this year's judging.

Teamwork is strongly encouraged, but not a requirement.

The quality of the research produced, is much more important.

Multi Year Projects
Teams are judged on the basis of what's new in this year's project, not on the work they did in prior year(s).

For example: Team 72 improved its model with dark matter distributed around the galaxy instead of at a point mass at the center of the galaxy. Instead of using this model to reconcile theoretical predictions of red/blue shifts in other galaxies with experimental observations(last year), the model was used to study the collision of two galaxies(this year). The new work won 2nd Prize.

On the other hand: Team 66 did little more than parallelizing their model to improve its performance. They won a prize for the best use of the Python programming language. Only a few teams competed for this prize.

Original Research, Applications, and Replications
Successfully doing original research is much more difficult than replicating existing research. This year Team 82 took 1st Prize for original work image processing Alzheimer's MRI scans.

Team 72 is replicating the work of professional scientists, and won 2nd Prize this year. Last year's work on this project took 1st prize. Team 72's work is using garden variety consumer grade computers. The professional researchers routinely use much more powerful computers for their original research.

We encourage teams to apply existing research to a specific problem, this is usually the most straightforward path to a successful project. We need to look through the archives for examples of teams who won prizes for applying existing research to specific problems.

This is only one of many aspects of each team's project considered by the judges.

Mentors and Parents
Yes, teams with close relatives as mentors have advantages. But, a parent is not always the best mentor.

A team with a strong mentor obviously has advantages.

However, It is my observation that the best mentors are the mentors the team finds on their own. Consult attempts to find mentors for teams who request help. If a team feels they need a better mentor, it is the teams responsibility to get off their butts and go look for one. Just like it is their responsibility to go the library, and internet to look for previous work on their topic. Teams may not have experts in their local community, but they can find internet mail lists and forums used by professionals and ask for help finding a mentor.

The judges often consider whether the team really did the work they are claiming. Parents, teachers, and mentors are not permitted to participate in the team's presentations. Teams that do not demonstrate a clear understanding of "their own work" are not judged highly.

We do not have, or want handicapped scoring (except for teamwork). For example, elementary and mid school teams do not get extra points to make them more competitive with high scool teams. And, they do amazingly well on a level playing field.

General vs. Specific
Good (and poor) work can be done on both specific and general problems. The quality of the work is more important than the scope of the project.

General projects are almost always more difficult than specific projects. Very general projects are discouraged because they are often beyond the scope of a single year project. Teams that work on more general problems are less likely to succeed. Teams with general projects are encouraged to choose a specific part of their general project for this year's project. Successful teams often begin the year with overly ambitious elaborate approaches to general problems. As the year progresses and they monitor their progress, they often narrow their focus to a simpler more specific problem with excellent results.

Computational Depth Requirement
This is because this is a computational science competition. And, the broader the scope of the project the greater the scope, scale, and complexity of the computation required.