Quote:
Originally Posted by SteveGPage
We will probably, at BAE, still have our scouting done via web forms on smartphone/mobile devices, and then processing the data locally on my laptop. (darn job getting in the way of me spending the 100 hours or so I need to finish this the way I want!) Graphs and analysis will be displayed using DHTML- or Flash-based dashboards. My plan is, however, by the time we get to our week 3 regional, the database and all the calculations will be done on a Linux server and the dashboards will all be iOS based. We have 2 iPads now, and several people have iPhones and iTouchs as well.
I messed around with an algorithm to calculate a team's contribution by match during Lunacy, but could never get it to work consistently. I had a huge matrix to do the calculations, based on a modified BCS algorithm, but it would only give me good data when every team had been in the same number of matches - so I gave it up! Last year was a whole lot easier to calculate! I would love to see your approach to that calculation sometime.
See you at BAE!
|
Basically I only look at qualification matches, and to account for teams playing in different # of matches, I take the number of matches of the team your looking at -1 and set that as the max # of matches. Then I go through all the teams you play with. and calculate their total score over those n matches. based on this I get a total score for the alliance at that event, and then based on each teams ratio of that total score I find their contribution to that match. Averaging a team's contribution to each match, I estimate their expected contribution to the next match.
I also do the same for the opposite alliance, to calculate the teams expected spread and points allowed