2015 OPR After Week One Events

[Ed Law]

The OPR/CCWM numbers have been posted, please see

http://www.chiefdelphi.com/media/papers/2174

There are a number of points I would like to make:

1) All week 1 events are included.

2) If the negative numbers at some of the events mean surrogate teams, I still included them in the calculations. This is because I am calculating World OPR. It would not be fair to teams attending that event to not have their OPRs included in the calculation because some matches have surrogate teams. Hopefully we will not have too many surrogate teams this season.

3) I checked my component OPRs against Ether's. I am 99% confident my numbers are correct.

4) World OPR is published. Since there is no interaction of teams in different events, the numbers should be the same as the OPR numbers at their event.

5) CCWM is meaningless this year. I am including it because the program automatically calculates and report it.

6) This spreadsheet is all set up to run update of week 2 to week 7 events. During the events, you just have to do Cntr-Shft-P in each tab and it will update everything for you. However, because of changes FIRST made to their website, it may take up to a minute to update each event instead of seconds. I also avoid updating Playoff Rounds so it will not take as long each time during Qualification Matches. You will need to modify the macro to turn that back on.

7) Because of problem with data in week one, Cntr-Shft-P does not calculate OPR/CCWM after updating data. You need to do Cntr-Shft-O to calculate OPR/CCWM

8) I am now able to get the alliance selection directly from the FIRST website. There is no need to update that Google doc.

If you find any error or have any questions, please let me know.

[Ether]

http://www.chiefdelphi.com/forums/sh...7&postcount=10

[Mark Sheridan]

Great work.

[Ed Law]

I think I misunderstood what the negative numbers mean. Now my understanding is these are regular surrogate teams just like in past years. FIRST is just using to minus sign to signal they are surrogate teams.

I have always included surrogate teams in calculating OPR. I assume all teams always try their best in every match whether they are a surrogate team or not. In the past, based on Match Results only, we don't know which team is a surrogate team unless you look at the match schedule. Now that we know which teams are surrogate, we can potentially not include them in the calculation. However, I propose for the sake of consistency to previous years, we will continue to calculate OPR the same way including surrogate teams in both the A matrix and b vector when solving A x = b.

For component OPR, I will need to study it some more to see if there is a benefit to not include surrogate teams. I already checked my program. It can be done relatively easily. It may improve the accuracy of the component OPRs. However the scale and shift operations I do to take care of surrogate and DQ teams seem to work well enough. By not including the surrogate teams, I still have to do the scale and shift for DQ teams. I have to do a little more testing before making a decision whether to change it or not.

[Jacob Plicque]

Thanks Ed for your wonderful spirit of sharing. This OPR spreadsheet is a valuable asset for teams with little or no scouting capability.

[Kevin Leonard]

Thank you for your yearly work, Ed. It's always appreciated.

[stuart2054]

Ed,
Thank You for this valuable resource. I know it must take a ton of time to put together and any mentor already has his hands full. If the mentor that recruited me in 2010 had been truthful about the time it takes I probably would not have done it and missed out on all the fun I have had since!!!

THANK YOU!!!!!