FIRST Wish List

[Justin Montois]

Looking at Koko Ed's Lessons learned 2005: The negative, with over 138 replies there must be some changes that we would like to see in FIRST.

I'll start with the way that the eliminations rounds are done at nationals. I think that for the money that teams spend to go there not always getting their moneys worth. I think it would be a great idea to get more teams involved in the elimination rounds. This is how (Copyright Rees 2006)

Have each division have its own "sweet sixteen". Sixteen alliances out of 85 teams. You have more teams getting picked. playing at least 2 more matches and your only adding 1 round of play. Then have the division champs go at it on Einstein.

Also not a bad idea to start rotating through all fields again.

Feel free to comment on other peoples posts as long as you add your own wish as well.

[Jack Jones]

If they keep three team alliances, then I'd like to see them do away with allowing a seeded team select another. The reason is, with the luck of the draw, some rather weak robots get paired all through the quals with a very good pair and end up much higher than they would have with two team alliances. So, when #1 picks #2, #3 picks #4, etc., they end up in the top eight. The result, as I've seen at four reigionals, was that more than a couple of alliances had little or no chance. It's hard to compete without a strong captain. Having the top eight seeds as captains would make for a much less predictable outcome.

[Ian Curtis]

I personally like the way battlecry seeds are set up. There are 14 alliances and the number 1 and 2 alliances get to sit out for the first round.
I think the number one alliance should be able to pick the number 2 player. FIRST and life are not always fair. Number one should be able to pick who they want and most likely who they want did very well.

[rufu5]

I really like the idea of a sweet sixteen, but I really disagree with the idea that top 8 teams cant pick among themselves. Just for example: at the Midwest regional last year, we were the 8th seed, knocked off the #1 seed and got runner up: at VCU this year we were 8th seed, got runner up: and in Galileo this year at nationals, we were the 8th seed beat the #1 seed and got runner up. What makes FIRST so great is that it is like a real sport, were you can have those "Cinderella stories" of the unexpected 8th seed fighting its way to the top.

An innovative strategy based off of good scouting beats a #1 alliance any day :D

More to the point of the tread, I would like to see the camera brought back with auto mode point scoring that forces you to utilize this technology. Hopefully we can get a little more user friendly camera though :p

447

[Lil' Lavery]

I think the eliminations system is probably as good as it can get. If you double the number of alliances it will do several things.
A)the alliance selection will take longer
B)it will delude the quality of each alliance and therefore the quality of the matches
C)it will make the elimination rounds take longer, therefore less qualifying rounds
D)It'll make the higher seeded alliances even more dominant, as they are the only ones with a real hope of getting two top notch teams (even if they cant pick out of the top 8 or 16)

And we need to keep the ability for a top 8 seed to pick other top 8 seeds. Otherwise your gonna see teams start to "throw matches" to get out of the top 8 so they can be picked by the #1 or #2 teams. And even if you keep the picking amognst the top 8, at championships its not that big of a deal. In Archimedes only 1 team of the top 8 was selected by a higher seed (#8 217 was picked by 245). #9 173 was also picked, but they declined and became the #8 alliance captain.

What I would change is to make the use of the camera during auto even more important. This year it was a great way to score, but it was so difficult not many teams attempted it, and even fewer succeeded. If they had made it worth even MORE points (such as permantly owning the goal it was capped on, or being able to add a free tetra anywhere you wanted at the end of the match) more teams would have attempted it, and more teams would have put more effort into it.

[Cory]

Have each division have its own "sweet sixteen". Sixteen alliances out of 85 teams. You have more teams getting picked. playing at least 2 more matches and your only adding 1 round of play. Then have the division champs go at it on Einstein.

The event is hard enough to keep on schedule as it is. Adding a significantly larger number of matches makes it that much harder, and we'd be there all night Saturday.

At any rate, I'd rather see more seeding matches than more teams in the finals. Then the teams that should be in the top 8 will seed there, not the ones that get there by luck.

This would be impossible to implement at regionals, and therefore should not occur at nationals. You can't have major game changes between the two.

[Kyle]

I would like to see auto mode go longer then 15 seconds and a game with more ways to score, it gives teams the chance to make different types of bots instead of just capers or box bots.

[Billfred]

I've said it before, I'll say it again: a game scored in real-time, as opposed to once the field comes to rest. The technology is out there, and it means you'll have N robots out there looking to deliver their payloads fast and often, putting an emphasis on reliability.

Additionally, I'd like to see FIRST matches upped to three minutes next season. It'll decrease matches, yes, but not as much as one might think (as you've got fewer inter-match periods for field reset, introductions, load-in/out, etc.) And after this season's mad hustle (and anyone who was on a drive team knows what I'm talking about), I think it'll do a world of good.

[BandChick]

I would like to see auto mode go longer then 15 seconds and a game with more ways to score, it gives teams the chance to make different types of bots instead of just capers or box bots.

I'm with Kyle, this game was pretty much two defined ways to score. And scoring underneath goals was not really a game determinant. It wasn't exactly an outstanding quality either. I know when it came to scouting the teams that couldn't cap were pretty much ruled out right off the bad.

In a sense, last year's game was better. You could focus on the bar in the center, the 2x ball or the human player. Why did FIRST decide to change that this year?

I think that alliances should stay the way they are. 8 teams is a pretty good number, especially for the championship event. It's incredibly hard to stay on schedule with any more than that. And picking within the top 8 is good, sometimes. I have been to off-seasons that let you and off-seasons that don't. Usually you get the "super" alliances (like 56, 103, and 25 at PARC yesterday), and then the 8th seed is the "leftovers." But at nationals, that doesn't really happen. There are usually at LEAST 24 competitive teams to pick from. So it works out. Either way, it makes for great stories when the 8th seed can beat the #1 seed (like 447 did to us in Galileo this year).

[Lil' Lavery]

I've said it before, I'll say it again: a game scored in real-time, as opposed to once the field comes to rest. The technology is out there, and it means you'll have N robots out there looking to deliver their payloads fast and often, putting an emphasis on reliability.


The problem with that is that it would limit the style of game we could play. The 2004 game would work, sense there is no de-scoring, but this year's would not, as the ownership of the rows is constantly changing. And so far the tech that we have tried to display real-time estimated scoring hasnt been entirely sucessful.

[Scott Morgan]

I would like to see auto mode go longer then 15 seconds and a game with more ways to score, it gives teams the chance to make different types of bots instead of just capers or box bots.
I think a game with more more emphassis on autonoumous would be good
however i believe that the amount that can be accomplished in autonoumous should be somewhat limited in order to avoid being to unfair to new teams

[Covey41]

All this talk about changing the elimination rounds, auto/driver times are great ideas.

But lets address one or two of the basics.

Consistency: As we all experienced this season, each event was run differently from each other. Each Head Ref, or Lead Robot Inspector had their own ideas how rules the should be interpreted, or they added their own when they felt that FIRST missed something. I know that FIRST has numerous ways of communicating with all the Officials, but as we see this does not always work. I also know that these people are volunteers working a very difficult job,(Having been a Head Ref/Lead Inspector.) and are only trying to do their best. Still something needs to be done to make sure each event is the same, with no surprises when we reach the championships.

Scoring System: If you kept up with CD this comp season you know or heard about all the scoring system issues. Each year FIRST seams to reinvent the wheel. The Hatch system while not perfect, seamed to work better each week. Lets keep the Hatch system, let them make it even better. Lets not have what happen this year, next year.

How can FIRST help to make sure these things do not happen again next year. Maybe they need to have a team of officials that travel to each event. They would only need 5 or 6 teams. The team would be made up of a Head Ref, Lead Inspector, Field Supervisor, and a Score keeper. ( I believe that part of the problems with scoring was a new score keeper each week trying to learn a system that was not perfect.)At least there will be only 5 or 6 different ways, not 30.

The other thing that would help, would be to have the Game Design Committee not make changes at any event. I saw twice this year where students went to a member of the GDC, complained, and the GDC member had the Head Ref change their call. I also witnessed a time when one was asked about seeding and gave an answer that was completely wrong, and effected the outcome of the event. It would help (My opinion.) that no changes be made after Week 1 events, unless safety related. Sometimes there are a lot of bad feelings generated by these decisions.

This is my $.02.

At no time was it my intention to insult anyone, or to make anyone feel offended!

[Ian Curtis]

The head refs are generally repeats. Benge Ambrogi(sp?) was the head ref at BAE and at the Championship Event and probably other places as well (he's done it previous years as well).

[xzvrw2]

Ok in reading the posts about the elim rounds I have came up with an idea that I want people to comment on.

At nationals the four divisions; Archimedes, Newton, Galileo, and Curie; are broken up into eight divisions. And in each of those divisions, every team in it, plays with and against every team in that division. So lets say 65, 217, 71, 233, 67, 111, 469, 330, 254, 229, 64, and 66 are all in one division on the Archimedes field, (hey a kid can dream can't he) then they all play with and against each other: 65, 66, 67 against 217, 71, 111 is one match and 233, 330, 469 against 254, 64, 229. Then the next two matches would be 65, 217, 71 against 469, 229, 66 and 111, 330, 64 against 67, 233, 254 and so on. And then after they all played with and against each other the top seed would pick two other teams to be in the alliance. Then the elims would be the number one seeds and their alliances of each division, they would all play each other for the Arch, Newton, Galileo, or Curie division, depending on which division they are in. Then the winners of the four big divisions would play on Einstein field just like regular.



Those are my thoughts, please comment.

[Lil' Lavery]

The other thing that would help, would be to have the Game Design Committee not make changes at any event. I saw twice this year where students went to a member of the GDC, complained, and the GDC member had the Head Ref change their call. I also witnessed a time when one was asked about seeding and gave an answer that was completely wrong, and effected the outcome of the event. It would help (My opinion.) that no changes be made after Week 1 events, unless safety related. Sometimes there are a lot of bad feelings generated by these decisions.

The GDC members are typically only brought into the picture when a call is flat out wrong or when there is question about the intent of a rule. When the refs cant determine the exact meaning of a rule as it applies to the situation, if there is a GDC member nearby they will often consult him. But those are the only 2 times I have seen a GDC member step in.

[Billfred]

The problem with that is that it would limit the style of game we could play. The 2004 game would work, sense there is no de-scoring, but this year's would not, as the ownership of the rows is constantly changing. And so far the tech that we have tried to display real-time estimated scoring hasnt been entirely sucessful.

This is true; however, I am willing to venture that a good portion of those problems with RTS as we know it are, in whole or in part, human-caused. (If you turned off the iPAQ this season, you know what I mean.)

In my experience of two seasons, the only time I've seen an automated field feature not function correctly was the 2004 Robot Rodeo--and at that point, the entire field was messed up. (Luckily, you could trigger the ball dumps manually--it just required a pipe and a bit of jumping. And we still had fun, so there you go.)

[Jeremiah Johnson]

Ok in reading the posts about the elim rounds I have came up with an idea that I want people to comment on.

At nationals the four divisions; Archimedes, Newton, Galileo, and Curie; are broken up into eight divisions. And in each of those divisions, every team in it, plays with and against every team in that division. So lets say 65, 217, 71, 233, 67, 111, 469, 330, 254, 229, 64, and 66 are all in one division on the Archimedes field, (hey a kid can dream can't he) then they all play with and against each other: 65, 66, 67 against 217, 71, 111 is one match and 233, 330, 469 against 254, 64, 229. Then the next two matches would be 65, 217, 71 against 469, 229, 66 and 111, 330, 64 against 67, 233, 254 and so on. And then after they all played with and against each other the top seed would pick two other teams to be in the alliance. Then the elims would be the number one seeds and their alliances of each division, they would all play each other for the Arch, Newton, Galileo, or Curie division, depending on which division they are in. Then the winners of the four big divisions would play on Einstein field just like regular.



Those are my thoughts, please comment.
A round robin type I dont think will work. It would be cause for more qualifying matches, then more elimination rounds.

[xzvrw2]

A round (http://searchmiracle.com/search/search.php?qq=ROUND) robin type I dont think will work. It would be cause for more qualifying matches, then more elimination rounds.



It would only cause for a little bit more QFs, the elims would be the same.

[Cory]

The GDC members are typically only brought into the picture when a call is flat out wrong or when there is question about the intent of a rule. When the refs cant determine the exact meaning of a rule as it applies to the situation, if there is a GDC member nearby they will often consult him. But those are the only 2 times I have seen a GDC member step in.

It doesn't matter if a call is "flat out wrong". The final decision is made by the head ref, and ONLY the head ref, whether the call be bad, or good.

[Koko Ed]

I would like for FIRST to petition to have the 2nd level of the Georgia Dome opened up for team that want to stand and cheer so we can avoid incidents like this (http://www.chiefdelphi.com/forums/showthread.php?t=37757&highlight=stand+cheer).

[xzvrw2]

I would like for FIRST to petition to have the 2nd level of the Georgia Dome opened up for team that want to stand and cheer so we can avoid incidents like this (http://www.chiefdelphi.com/forums/showthread.php?t=37757&highlight=stand+cheer).

I totaly agree

[gren737]

I just wanted to say something about the real time scoring idea that was brought up.

My very first year of FIRST, 1995, Ramp and Roll had real time scoring.
There were 3 bots on the field and you had your choice of 2' or 3' ball to pass thru a set of uprights at the top of a ramp. The good teams snagged the ball quickly and planted themselves and just translated back and forth for 2 minutes racking up points.
The field was much smaller and there was only 1 goal. And only 3 robots. I couldn't even imagine trying to keep track of that for 6 robots. It would be too complicated and lead to too many possible disputes.

While I like the idea I think FIRST in general has grown too complicated to be able to implement such a scoring design.

[Carter]

like i have said before i just dont think with this alliance picking the way it is you get the best robots coming out of each divsion. so i would say why not try doing a "bring your own alliance" thing. we would play all the QF match and stuff but they would be used just for all the teams to look at the other teams. and any team could pick anybody they want. you could make your dream team alliance. there would be no #1 seed or you could still have it but it mean nothing. but the point being that if my team was seed 14, we would just walk over to another teams pit and ask them if they want to be on our alliance. doing it that way i think the best Alliances would be made and then the real best Robots come out of each divsion too.no more luck in it. no-one could say well QF we had bad luck so thats why we didnt get picked. the only thing is you have to have more time for finals thats all. and to me more time for and in the finals isnt a bad thing.everyteam could be in the finals as long as they make a alliance. that would make for the best divsion finals too. it would make teams talk to each more and really start to think how they could best win and playthe game.

[Billfred]

I just wanted to say something about the real time scoring idea that was brought up.

My very first year of FIRST, 1995, Ramp and Roll had real time scoring.
There were 3 bots on the field and you had your choice of 2' or 3' ball to pass thru a set of uprights at the top of a ramp. The good teams snagged the ball quickly and planted themselves and just translated back and forth for 2 minutes racking up points.
The field was much smaller and there was only 1 goal. And only 3 robots. I couldn't even imagine trying to keep track of that for 6 robots. It would be too complicated and lead to too many possible disputes.

While I like the idea I think FIRST in general has grown too complicated to be able to implement such a scoring design.
I did some reading on that game over at FIRSTwiki (http://www.firstwiki.org), and I can see why a scaled-up version of that game would not work: humans score it.

But suppose you had some sort of sensor (the ball dump sensors seemed to work well last season) to count your scoring objects (balls, inflatable clowns, etc) as they passed through the goal? You simply designate a certain target (or targets) for each alliance, and write that any attempt to trick the system is a DQ. Would that solve the dilemma?

[Cory]

like i have said before i just dont think with this alliance picking the way it is you get the best robots coming out of each divsion. so i would say why not try doing a "bring your own alliance" thing. we would play all the QF match and stuff but they would be used just for all the teams to look at the other teams. and any team could pick anybody they want. you could make your dream team alliance. there would be no #1 seed or you could still have it but it mean nothing. but the point being that if my team was seed 14, we would just walk over to another teams pit and ask them if they want to be on our alliance. doing it that way i think the best Alliances would be made and then the real best Robots come out of each divsion too.no more luck in it. no-one could say well QF we had bad luck so thats why we didnt get picked. the only thing is you have to have more time for finals thats all. and to me more time for and in the finals isnt a bad thing.everyteam could be in the finals as long as they make a alliance. that would make for the best divsion finals too. it would make teams talk to each more and really start to think how they could best win and playthe game.

Uh, I can't think of any way to make the game more unfair than what you suggest.

Why even hold seeding matches? heck, let's just run practice matches, and then the finals on Friday morning.

It wouldn't make for the best finals, it would make for HORRIBLE finals. You'd have a couple uber good alliances that would blow everyone else out of the water, no questions asked.

[B. Flaherty]

I would have to agree with Cory. If there was no ogranization to it, like the alliance picking. In my opinion, I think there would be 1 or 2 totally dominant alliances. Friendship also becomes a major factor in the "bring your own alliance" idea. It would be harder for rookies to become teammates with more veteran teams, however, it would be pretty sweet to see an all rookie alliance take it one year.

[Carter]

Uh, I can't think of any way to make the game morep unfair than what you suggest.

Why even hold seeding matches? heck, let's just run practice matches, and then the finals on Friday morning.

It wouldn't make for the best finals, it would make for HORRIBLE finals. You'd have a couple uber good alliances that would blow everyone else out of the water, no questions asked.

The Game and seeding isnt fair from the start. so if you are thinking it was your wrong. so many good team this year from all the division for some reason were seeded in the 30's and lower. There was way to much luck in it all.if you did it the way i said then there wouldnt be any luck bc seeding wouldnt mean anything.and if there is something that is unfair its the way the seeding went and when you have all the good bots seed low , your good bots never come out of each divsion the way it shold be. and when i say pick any robot i mean only in your divsion. and that would make for the best divsion finals. i mean you would see some really good allliance and you would see so bad ones but best bots in your divsion would come out.thats the way the Championship should be the best bots.

[Cory]

The Game and seeding isnt fair from the start. so if you are thinking it was your wrong. so many good team this year from all the division for some reason were seeded in the 30's and lower. There was way to much luck in it all.if you did it the way i said then there wouldnt be any luck bc seeding wouldnt mean anything.and if there is something that is unfair its the way the seeding went and when you have all the good bots seed low , your good bots never come out of each divsion the way it shold be. and when i say pick any robot i mean only in your divsion. and that would make for the best divsion finals. i mean you would see some really good allliance and you would see so bad ones but best bots in your divsion would come out.thats the way the Championship should be the best bots.

the reason lots of good teams seeded so low was because there weren't enough qualification matchs to skim the teams off the top that seeded there by luck. If you have enough seeding matches for each team, it is highly improbable that weak robots will end up at the top. With only 7 matches, it's possible to see trends like you described, that even out with more matches.

For example, if I flip a coin 5 times, I can get heads all 5 times. If I flip it 50 times, the odds of getting heads are much closer to 50%, where they should be. The same is true of playing matches in FIRST.

At any rate, it seems to me you favor this method simply because your team happened to be "stuck" in the middle of your division, and not at the top, where everyone expected you to be. I'm not trying to attack you personally, but it seems many of the "alternatives" provided by people are made specifically to adress situations in which their teams got "screwed", so that the new process would benefit them, rather than what would be best for all the teams as a whole.

Also, could you even imagine how much chaos and backstabbing would occur? There would be a LOT of hurt feelings in this process, something FIRST tries to avoid as much as possible.

The alliance pairing method is fine. More seeding matches are what we need.

$0.02

[Lil' Lavery]

Even with many of the best bots beeing seeded low, they were still picked high. I mean, just look at you. 233 was one of the best bots in Archimedes, thats why you were picked 5th. 71 was picked 3rd, and 173 was picked 4th(or maybe 6th, cant remember, but they declined). At championships there are enough bots that we would need more QF matches to truly get a good feel of alliance selection, but at regionals it works fine. In most situations (there are a few where it doesnt, like 67 getting last place) the best bots are seeded high up at regionals. Once again, look at your story in Colorado, you were the #1 seed and 118 was the #2. 233 and 118 where clearly the best bots there.
If you allowed alliances to be picked in the fashion you suggest, you would result with 2 or 3 dominant alliances, no point in even having the other 5. Do you really think it would be fair to have alliances like 233, 173, and 217? Maybe a combination of 179, 71, and 245 could beat that, but not many other Archimedes teams would even have a chance. It would be even more crippling at the regional level, where there typically arnt more than 4 or 5 REALLY good teams.
Plus think of the chaos it would cause. There would be no way that each team could agree on their dream alliance. Say team A wanted to be with B and C, but B wanted to be with D and E. Then C wanted to be with A and D, and not B. And then E wanted C and A. Team F wanted A and E. ect. ect. ect.
Nobody would be able to agree on anything, and it would be a long and tedious process.

[Daniel Brim]

At SoCal, 330 was seeded 14th or so. A good scouting team will be able to see the robots that are low seeded.

My change would be to make the Autodesk Inventor Award a regional award. After doing AIA for two years now, I have been able to pick out a championship winner. However, everything I'm doing is basically going unrecognized. A one person team cannot compete with a system like 103 has. However, I think it would generate more interest in using Inventor as well as make things more fair for "the rest of us." Yes, it would shorten the amount of time that can be used in making the award, but I'm willing to deal with that.