During last month's newsletter, we announced that we are going to begin quarterly reviews of all our contest rules and submission guidelines.

Based on the valuable feedback you guys provided to us during our Modular Fuel Cell Energy Management Contest, we learned a lot of great lessons that we wanted to implement before we release our next Evidence of Use Contest.

One of those lessons was that while we want to reward the user who submits the largest number of high scoring products, we also want to make sure that users who submits one high scoring product will be rewarded as well.

In our new rules, we wanted to recognize both users proportional to their contributions. To that end, we will be implementing a new prize distribution model for Evidence of Use Contests.

Let’s take an example. Say we run an Evidence of Use Contest for a prize of $10,000. As the Contest comes to a close, four verified products have risen to the top as the highest scoring products across the entire contest. We’ll call them Products A, B, C, and D. For these products, we’ll set aside 90% of our overall prize pool to be split between these four products. (We’ll get to that 10% in a second.)

Prize Distribution for Evidence of Use Contests
Figure 1. Distribution of the Overall Prize Pool

Now, we have four users who participated in the Contest and have submitted at least one of these highest scoring products. Let’s call them Users Alice, Bob, and Charlie.

For each of these products (A, B, C, and D), they’ll be ranked against each user’s submission independently. Think of it like every product having its own mini-Contest. The highest scoring submissions for each product will be marked as Winner or Runner Up and be allocated a prize in our familiar split (80% between winners, 20% between runner ups).

So for example, in Figure 2:

For Product A: Alice is the winner at 100 and gets $1,800. Charlie is the runner up at 80 and gets $450. Bob doesn’t rank since he’s below the median.
For Product B: Bob is the winner at 90, which gets scaled to 100, and gets $1,800. Charlier is the runner up at 60, which gets scaled to 67, and gets $450. Alice didn’t submit this product, so she sits this one out.
For Product C: Alice is the winner at 70, which gets scaled to 100, and gets $1,800. Charlier is the runner up at 40, which gets scaled to 57, and gets $450. Bob didn’t submit this product, so he sits this one out.
For Product D: Alice is the winner at 70, which gets scaled to 100, and gets $1,800. Bob is the runner up at 50, which gets scaled to 71, and gets $450. Charlie didn’t submit this product, so he sits this one out.

Finally, remember that remaining 10% of that overall prize pool? That be given to the submission with the highest number of winners and runner ups as a final bonus to the user with the greatest overall contribution.

So for example, in Figure 2:

Alice and Charlie got the most winner/runner up products each at 3, so they’ll evenly split the $1,000 for $500 each.

Figure 2. Hypothetical Scoring of an Evidence of Use Contest

Now that everything is out of the way, it’s time to tally the results. Users will inherit the title and score of their highest scoring product, and will receive the cumulative amount of their prizes. So in that case:

Alice will be marked as a “Winner,” receive $5,900, and receive a final score of 100/100.
Bob will be marked as a “Winner,” receive $2,250, and receive a final score of 100/100.
Charlie will be marked as a “Runner Up,” receive $1,850, and receive a final score of 80/100.

We’re hoping these changes can help us reward every user who successfully contributes to our Contest. These rules will take immediate effect to all contests posted on and after May 7th, 2014.

We always welcome any feedback or questions you might have to our rules and policies. Please feel free to comment below or email us directly at contests@patexia.com.

Thanks for being a valuable member of the Patexia Community and the best of luck on your next Contest!

Cheers,
Andy Chu
Manager, Case Research

Evidence of Use Contest Rules (ver. 2.0.0)

Changed submission’s score to be “calculated based on the maximum of the verified relevancy score of all products.”
Clarified definition of verified product to include products that are not disqualified and that received verified “Yes” answers for all mandatory criteria.
Added to standard for verification that explanations will verified based on “a confirmation of the technical workings of the feature described in the question.”
Added new rule to specify that "any issues with interpretation of a contest questions will be resolved using the original text of the patent(s) from the contest design."
Clarified tie-breakers to include that "the final score may be modified while maintaining overall ranking."
Changed Product Distribution rules to focus on awarding per product quality and overall number of successful submissions.