FRC Spyder 2014?

[Dave Scheck]

Originally Posted by Banderoonies

Anyone know why the FRC Tracker has no data at all for New England Events? Is it just me?

Do you mean FRC Spyder or FRC Tracker? We only do Spyder. Groton is giving a 404 on the FIRST website http://www2.usfirst.org/2014comp/Eve...chresults.html. UNH has a results and the data shows up for me when I click refresh in the event view.

Remember, if there's nothing published, there's nothing that Spyder (or Tracker) can do about that. If you notice a case like that your best bet is to talk to someone at the event and make them aware that their website isn't showing correct information.

[plnyyanks]

Originally Posted by Dave Scheck

Remember, if there's nothing published, there's nothing that Spyder (or Tracker) can do about that. If you notice a case like that your best bet is to talk to someone at the event and make them aware that their website isn't showing correct information.

I'm scorekeeper at Groton - I'll take a look into it when I'm at the venue tonight

[Mike Soukup]

I just posted Android version 2.0.5 that adds two new tabs to the event screen:
- "team stats" has OPR, DPR, CCWM calculations for each team
- "predictions" has score predictions of each unplayed match using each team's OPR

I'll leave it up to the community to debate how useful the OPR statistic is for this year's game, but we wanted to provide it as another easily accessible data point available at your fingertips.

[Hoover]

Hi Mike,

I like the new tabs. What I'd like to see in the predictions would remain after the match so we could compare. I know that subsequent games alter the OPR. It would have to consider the OPR as it was before the match so that the prediction could remain as it was just right before played. This might be a lot of work but if it could be done then an indicator could show if the prediction held. Perhaps even a percentage accuracy could be known.

This could even be done post event. If it took a lot of CPU it might be better that way. Any possibilities?

[Mike Soukup]

Originally Posted by Hoover

Hi Mike,

I like the new tabs. What I'd like to see in the predictions would remain after the match so we could compare. I know that subsequent games alter the OPR. It would have to consider the OPR as it was before the match so that the prediction could remain as it was just right before played. This might be a lot of work but if it could be done then an indicator could show if the prediction held. Perhaps even a percentage accuracy could be known.

This could even be done post event. If it took a lot of CPU it might be better that way. Any possibilities?

I was wondering how long it would take for someone to ask for this. I guess the answer is ~7 hours. It's trivial to do it the wrong way and just apply the current OPR to previous matches, but that's just silly. It's a bit more work to do it the right way, as you described. I like the idea though, it's something we can look into, but it's low on the list.