Quote:
Originally Posted by Chris Hibner
Because of the fact that coop balancing will probably occur in 90% of the matches at IRI, coop bridge points need to be increased in order to increase it's affect on seeding. Make the coop bridge worth 4 points for a double balance in order to really penalize a failed attempt.
|
I can see what you're saying, but I don't think a single element of the game (the ability to balance the bridge) should be able to overwhelm the several elements of the game and various intangibles that go into winning a match. Several things concern me about more than 2 points being awarded for a standard co-op balance:
- When two robots in a match failed to get co-op points for a "silly" reason... perhaps a robot was flipped on the bridge, they got a wheel stuck on the siderail, etc.... you'd have the failure of a single robot at the last second dragging down six teams in the rankings. This wouldn't be a single point, this would be equivalent to winning two matches! So, your un-defeated, super-awesome team whose only flaw was that they trusted their capable partner to balance the co-op bridge for them is suddenly set back anywhere from a 1 or 2 to 10 or more places in the rankings.
- If a 2-robot co-op balance were worth more than a win, you'd be radically skewing the ranking system toward a single capability: balancing. There's a lot more to this game than balancing though... hybrid scoring, ball harvesting, accurate shooting all in addition to the intangibles like strategy, driver skill, etc. By elevating any single element so dramatically the rankings would skew dramatically too.
Taking Troy as an example, as it had many co-op balances, I was interested in seeing how applying 4 points for each balance instead of 2 would affect the rankings... Attached are two plots side-by-side for comparison. I also attached the spreadsheet from which I made the plots.
The plots indicate a noticeably higher correlation between Rank and Points Scored and Rank and Wins when the co-op balances are worth 2 points. They also have fewer outliers. One could say the graphs for 4 point balances are almost characterized by outliers, producing a loose correlation.
Interestingly, the top 12 teams remained the top 12 teams, but the order jumbled around a fair bit... instead of proceeding "1->12", it went, "1, 8, 2, 3, 5, 4..." The greater changes seemed to be in the mid-tier teams though...
This graphing really didn't provide any overwhelming change in correlations, but it certainly looks like it'd put at least one more team in the top 8 that would make you scratch your head.