Now that team signups have slowed down, there's one more topic to discuss for this year's Oracle Challenge - the scoring system. Last year there there was some support for changing the scoring system for times when you pick the wrong winner. Just as a reminder, you get 50-x points for picking the right winner, where x is how many % your prediction was off. These are three main options for what should happen when you pick the wrong winner:

1. 45-2x. This is the traditional system. When you're wrong, you lose 2 points for every % off you are instead of 1. There's also a flat 5 point penalty for being wrong. One of the big advantages of this system is that it keeps the contest exciting by allowing risk taking. If you pick a big upset and it pays off, you can gain a lot of points.

One of the big downsides is that you can lose a lot of points compared to someone who predicted almost the same outcome as you. For example, someone who predicted Master Chief with 50% gets 5 points less than someone who predicted Frog with 50%. Even more extreme, someone who predicted Samus with 50% gets nearly 15 points less than someone who predicted Mario with 50%, even though both of them were predicting a very close match and both were way off.

One other disadvantage is that this system arguably encourages people to hedge their picks in order to avoid the 2x penalty, rather than predicting what they actually think is most likely.

2. 50-x. Pure percentage scoring. In this system, your score would always be determined by how many percentage points off you were, regardless of whether you picked the right winner. Some might argue that this system is truest to the idea of the Oracle Challenge - it's all about predicting the percentage of the outcome, and the difference between 49% and 51% is the same as the difference between 51% and 53%. It would also reduce the role of luck in matches where the two competitors are basically equal.

One of the main downsides of this system is that without a penalty for being wrong, it's a lot harder to take a risk and benefit from an upset. People who fall behind may feel like they have no chance to catch up, and the contest may become less exciting.

3. 45-x. This is a cross between the other two ideas. Unlike the current system, there's no extreme punishment when you're on the wrong side of a 60-40 match. You'll still lose a lot of points, but you won't lose twice as many as someone who also predicted a 50-50 match but got the winner right. On the other hand, there's still a 5 point penalty for having the wrong winner, so picking an upset does get you somewhere if you're right.

On the other hand, a downside could be that there won't be as big a potential for shakeups as there is in the current system. It might make it less exciting, but it's hard to know without trying it. This is the option I support because I think it's the best balance between making the contest exciting, allowing risks to be rewarded, but not punishing people too much just for being unlucky.
---
Nominate Aeris for SC2k6
http://boards.gamefaqs.com/gfaqs/genmessage.php?board=8&topic=27379470