2017 FRC Events API

[Caleb Sykes]

Does anyone recall when FIRST first published documentation for Score Details requests to their API for the 2016 game?

I am particularly interested in if the API this year will provide detailed information on gears on the airships, or if we will have to determine this information indirectly based on scored rotors. That would be difficult to do properly because gears and rotors don't have a linear relationship.

[Jon Stratis]

I would guess it will only provide scored rotors. From the field tour videos, the sensors in place detect the first gear (nearest the rotor) rotating, not which ones are placed. So I doubt the field will "know" how many are placed.

[Caleb Sykes]

Originally Posted by Jon Stratis

I would guess it will only provide scored rotors. From the field tour videos, the sensors in place detect the first gear (nearest the rotor) rotating, not which ones are placed. So I doubt the field will "know" how many are placed.

I hope you are wrong, but I agree that this is the most likely scenario.

[Ty Tremblay]

The API usually only contains details on things entered in the scoring tablets and FMS during the match. Since the FMS doesn't really care how many gears are delivered, you likely won't see that on the API.

[Eugene Fang]

Originally Posted by Ty Tremblay

The API usually only contains details on things entered in the scoring tablets and FMS during the match. Since the FMS doesn't really care how many gears are delivered, you likely won't see that on the API.

True, but they included which face of the tower was Challenged in 2016 and how many totes were on the near and far platforms in 2015 even though they weren't necessary for scoring, so one can hope.

"towerFaceB": "Challenged",
"towerFaceC": "None",
"towerFaceA": "Challenged",

"tote_count_far": 12,
"tote_count_near": 27,

Though I'm not optimistic since it would be hard to keep track of when gears are delivered.

[Doug Frisk]

Originally Posted by Eugene Fang

True, but they included which face of the tower was Challenged in 2016 and how many totes were on the near and far platforms in 2015 even though they weren't necessary for scoring, so one can hope.

"towerFaceB": "Challenged",
"towerFaceC": "None",
"towerFaceA": "Challenged",

"tote_count_far": 12,
"tote_count_near": 27,

Though I'm not optimistic since it would be hard to keep track of when gears are delivered.

Different referees could potentially enter that data, so it was necessary.

[Ty Tremblay]

They're quite responsive. I'm sure they would at least entertain a request for this.

[Jon Stratis]

Originally Posted by Eugene Fang

True, but they included which face of the tower was Challenged in 2016 and how many totes were on the near and far platforms in 2015 even though they weren't necessary for scoring, so one can hope.

"towerFaceB": "Challenged",
"towerFaceC": "None",
"towerFaceA": "Challenged",

"tote_count_far": 12,
"tote_count_near": 27,

Though I'm not optimistic since it would be hard to keep track of when gears are delivered.

Yes, but that's because that information was entered into the tablets by the refs. That way the same information could be displayed/updated on multiple tablets at the same time to help ensure the counts were correct. There doesn't appear to be such manual entry involved with the gears.

[The Doctor]

What would be really nice is to get an accurate count of boiler goals. Clearly, as other threads have shown, there is no accurate way of manually scouting those, so an automated solution would be really handy.

[Ty Tremblay]

Originally Posted by The Doctor

What would be really nice is to get an accurate count of boiler goals. Clearly, as other threads have shown, there is no accurate way of manually scouting those, so an automated solution would be really handy.

That's what the counters in the goals are for.

[The Doctor]

Originally Posted by Ty Tremblay

That's what the counters in the goals are for.

I'm saying that scouting team members can't get an accurate count of the number of fuel scored. If FIRST exposes this data, we can use it to aid in collecting scouting information. I did realize that the balls weren't counted manually by humans.

[Eugene Fang]

Originally Posted by The Doctor

I'm saying that scouting team members can't get an accurate count of the number of fuel scored. If FIRST exposes this data, we can use it to aid in collecting scouting information. I did realize that the balls weren't counted manually by humans.

The number of balls scored in boilers will almost certainly be exposed by the API. Of course, this wouldn't show which team scored the ball.

[ngreen]

Originally Posted by Eugene Fang

The number of balls scored in boilers will almost certainly be exposed by the API. Of course, this wouldn't show which team scored the ball.

The indexer in the boiler was much more than I expected, but the game wouldn't work without a reasonably accurate count, which of course will be in the API.

Filling in the blank between what an alliance scores and each team does are where some have realized limits of what scout can reasonably observe. I felt reasonably successful applying an OPR calc last year to alliance scores. This year I wonder if scout-generated weights applied to the matches might improve the calculation results. For instance, if only Blue.1 has a high goal scorer, then 100% of that score will be given to them and 0% to Blue.2 and Blue.3.

For gears, it would probably help to have the end number of gears scored to remove ambiguity between 7 and 13. They may like having that number too when they are deciding whether to up the number of gears for later competitions.