# Overall Evaluation Criteria **URL:** \ **Category:** Water Supply Forecast Rodeo\ **Created:** [February 17, 2024, 7:46am UTC](https://community.drivendata.org/t/overall-evaluation-criteria/10327 "2024-02-17T07:46:16Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![kmande](https://avatars.discourse-cdn.com/v4/letter/k/e47c2d/32.png) [@kmande](https://community.drivendata.org/u/kmande)\ **Post date:** [February 17, 2024, 7:46am UTC](https://community.drivendata.org/t/overall-evaluation-criteria/10327/1 "2024-02-17T07:46:16Z")
Since it is allowed to redevelop/improve/rethink a modelling approach at this stage, latest model and reports can contest for 90% of overall evaluation criteria. However rest 10%_**(Forecast Skill (Forecast) (10%))**_ will take into account evaluation from a model which has already been submitted into forecast stage .So there is no scope to improve upon this criterion. Is it correct? If yes, is it possible to revisit and change criteria for that rest 10% somehow so that contestants can contest in this stage for 100% of evaluation criteria on the basis of only fresh/updated work? Thank you! ---
**Author:** ![jayqi](https://yyz2.discourse-cdn.com/flex028/user_avatar/community.drivendata.org/jayqi/32/924_2.png) [@jayqi](https://community.drivendata.org/u/jayqi)\ **Post date:** [February 20, 2024, 4:24pm UTC](https://community.drivendata.org/t/overall-evaluation-criteria/10327/2 "2024-02-20T16:24:05Z")
Hi @kmande, That is correct. The 10% Forecast Skill (Forecast) is based on the performance of your frozen model in the Forecast Stage. Your submissions to the Final Prize Stage do not affect the outcome for that portion of the evaluation. The work you submit to the Final Prize Stage will constitute 90% of the evaluation score. This has been chosen to balance a variety of considerations and goals of the challenge. There are no plans to change the evaluation criteria for the Final Prize Stage.