paper: Stop the Stop Build

[marshall]

Originally Posted by Joe Johnson

To me the idea of helping another team is completely okay. Help away. But suppose that Team A helped Team B before an upcoming competition but Team B could only use the improvements IF they were on alliances that included Team A. Doesn't seem right. Also, before a competition, Team B seems to be in the driver's seat. They can accept the changes or not. It is up to them. But once an alliance is formed, Team B is under much more pressure to accept the cheesecake proposals of their alliance captain whether they like them or not.

I know, I know, Zebracorns feel that they were not taken advantage of. I hear you. And I don't care. Well that is too strong of a statement. I care, in fact, I am happy for Team 900. It was a good experience for you. But I STILL think that such excessive cheesecaking was bad for the sport. The typical team in the future will not have a great experience having their hard work (for 6 weeks ;-) being pushed to the side so that a top team can cheesecake the snot out of them.

Dr. Joe J.

So prior to an alliance being formed, it's acceptable to you to offer to help make a team as competitive as possible but once the alliance is formed it isn't?

Tell you what, forget I brought it up. I'm good without the public display of mental gymnastics that is someone coming to terms with their own cognitive dissonance.

[marshall]

Originally Posted by Michael Corsetto

I think a lot of FRC's issues boil down to program cost actually. Hmmm...

-Mike

Ding ding ding.

Edit: Makig clear what I was highlighting.

[FrankJ]

Originally Posted by marshall

So prior to an alliance being formed, it's acceptable to you to offer to help make a team as competitive as possible but once the alliance is formed it isn't?

It is a question of degree. Picking a team and making them a ramp anchor and telling not to move isn't inspiring. At least not to me. Admittedly that is an extreme. The Zebracorn collaboration of 2015 is the other extreme. (For those coming late to the party read this thread) Rules generally do a bad job of corner cases. Does that help with your confusion?

[Oblarg]

I don't see the need for half-measures here. Don't give everyone an unbagging time slot, just get rid of the bag entirely. The current policy is regressive and unfair, and lessened version of it is still going to be regressive and unfair, only somewhat less-so.

Originally Posted by Michael Corsetto

Random thought on cheesecaking.

Would teams cheesecake less if they were allowed to enter multiple robots for less-than-ridiculous costs? We would probably enter 3-4 robots if it didn't cost an arm and a leg.

This might just be me, but I can't envision this doing anything but further exacerbating the gap between high- and low-resource teams and breeding a lot of ill-will. For instance, I don't think many people would take kindly to seeing an elims bracket at district championships consisting of multiple robots from only a handful of "elite" teams. That might be a more accurate reflection of the distribution of resources in FRC (both monetary and human), but I doubt it's what's best for the program.

[Joe G.]

Originally Posted by Tim Sharp

I agree. One of the strongest motivational factors that sustains a team (IMO) is the sense of ownership the students have in their machine. Win or lose, being able to watch your robot on the field and knowing that part of it exists due to your hard work and effort is a powerful thing.

Completely agreed. And it is interesting to bring up cheesecake in the context of the bag discussion, because of a hypothetical scenario that has been bothering me for a while.

Most years, "cheesecaking" consists of relatively minor, relatively low tech additions dreamed up on the spot, in a collaborative effort between teams. Alliances are largely selected on the base competency of the robot a team showed up at the event with, and cheesecake provides minor enhancements. In 2015, we had a game uniquely suited to cheesecaking, largely because of just how completely the vital canburgling task could be completed via a self-contained, sub 30 pound mechanism, developed and brought in entirely by another team, and how few less than elite teams put any effort whatsoever into this task, or developed systems with a prayer of being competitive at it. As a result, we started to see some teams picked more on their willingness to abandon elements of the robot they brought in. And at the end of the season, cheesecake and its possibilities being in people's minds lead to us seeing a new robot built from the ground up at an event, based largely on design work done by another team prior to the event.

The debates over whether or not this was a positive thing have been beaten to death. But one indisputible fact about the experience remains: the harpoon build was a monumentally difficult feat for all teams involved. It required an unprecedented level of coordination, pre-planning, and engineering skill. We've only seen it once, and I don't know that we'll ever see anything quite like it again. If we do, it'll be hard not to be in awe of the teams that pull it off, and the amazing accomplishment will once again overshadow any sourness about the ethics of attempting it. One can say similar things about teams that manage to pull off mid-season full-bot rebuilds under the bag system, and arguments about design convergence.

However, the difficulty of this feat was almost 100% artificially generated, through the bag rules and withholding rules. Getting rid of bag and tag would presumably also erase poundage limits on fabricated items that a team can bring to competition with them.

Which brings up the logical questions: What stops teams, many of whom are already building multiple robots, from bringing in pre-built "Cake-bots," ready to roll as-is with different team numbers slapped on (or Cake-tops that can bolt on top of a kitbot, if FIRST adopts VRC-style definition of a robot)? Would the hypothetical gains in performance of the average team be enough to erase an elite team's motivation to do this? Would the sense of collaboration and involvement by all teams so often quoted regarding past extreme cheesecake endeavours always be maintained? Would we want to stop this at all, or would it be a positive thing to a degree?

EDIT: To be clear, I don't anticipate this ever becoming a widespread thing, nor do I mean to suggest that certain teams are ready and waiting to do this, only held back by the current ruleset. But the door does open up if we aren't careful.

[Joe Johnson]

Originally Posted by marshall

So prior to an alliance being formed, it's acceptable to you to offer to help make a team as competitive as possible but once the alliance is formed it isn't?

Tell you what, forget I brought it up. I'm good without the public display of mental gymnastics that is someone coming to terms with their own cognitive dissonance.

Marshall, Marshall, Marshall...

I have obviously offended you (and all of Team 900?). That wasn't my intent. Sorry for that.

AND... I have a views on how the world should work. I do my best to come up with a consistent set of values which you graciously call mental gymnastics and talk of cognitive dissonance.

From my point of view, our differences boil down to this: regarding excessive cheesecaking, I come down against it while you come down on the other side.

Can we disagree without insulting each other? Maybe?


Dr. Joe J.

[Cory]

Originally Posted by Michael Corsetto

Random thought on cheesecaking.

Would teams cheesecake less if they were allowed to enter multiple robots for less-than-ridiculous costs? We would probably enter 3-4 robots if it didn't cost an arm and a leg. The amount of time we could commit to cheesecaking would definitely taper off at that point.

I think a lot of FRC's issues boil down to program cost actually. Hmmm...

-Mike

I wouldn't want to compete in FRC if teams were allowed to enter multiple robots. That would be the very arms race that everyone wants to avoid. Poor teams would have no chance. Competition would be far more boring, strife between students on a given team would probably increase, as well as interaction between other teams becoming more cutthroat. Guaranteed teams would sandbag to try and get all of their robots on the same alliance. To keep up with the Joneses elite teams will be forced to dump more money and time into building more robots, training more drivers, and going to more events.

I know this has basically been done with 494 and 70, but that is a sort of unique situation that had a large opportunity cost.

[Oblarg]

Originally Posted by Joe G.

Which brings up the logical questions: What stops teams, many of whom are already building multiple robots, from bringing in pre-built "Cake-bots," ready to roll as-is with different team numbers slapped on (or Cake-tops that can bolt on top of a kitbot, if FIRST adopts VRC-style definition of a robot)? Would the hypothetical gains in performance of the average team be enough to erase an elite team's motivation to do this? Would we want to stop this at all, or would it be a positive thing to a degree?

Hopefully, ethical sense on the part of both hypothetical teams?

I don't see this as ever becoming widespread as I don't think there are that many teams who would ever consider this as an acceptable way to participate in FRC, especially on the part of the receiving team. Then again, I'm not from an ultra-competitive district, so perhaps the mentality really is that different there. I know our students would be pretty offended if someone suggested that we do that.

[nuclearnerd]

Originally Posted by Oblarg

Originally Posted by Joe G. View Post
Which brings up the logical questions: What stops teams, many of whom are already building multiple robots, from bringing in pre-built "Cake-bots," ready to roll as-is with different team numbers slapped on (or Cake-tops that can bolt on top of a kitbot, if FIRST adopts VRC-style definition of a robot)? ...

Hopefully, ethical sense on the part of both hypothetical teams?

I don't see this as ever becoming widespread as I don't think there are that many teams who would ever consider this as an acceptable way to participate in FRC, especially on the part of the receiving team.

From my own competition experience:

• In 2014 your alliance would be seriously improved if your third pick robot could be fitted with a trampoline for quick inbound bounce passes.
• In 2015 it would help to add a ramp or canburgler
• In 2016 (and 2013) a flip out defensive wall could make a huge difference.

Each of these examples were used in competition with great success (I could link to some, but given the "ethical" controversy surrounding cheesecake I won't). When the withholding allowance is removed, I can't see why more teams won't do the same.

That said, cheesecake limits are a separate issue from Bag/Witholding, and should be addressed with separate rules. The GDC tried to put in some rules last year, but they probably went too far. I think it's possible to strike the right balance with something like a separate weight limit, but that's a different conversation.

[Greg Woelki]

Great read, thanks for posting!

Minor stats comment: Fig (5) is a little concerning because each curve represents a different population of teams, so it isn't a very clear way to show the trend of increasing performance as teams have attended more and more events. It is unclear to what degree teams from the left-hand side are moving rightwards as they attend more events or if most of those teams simply aren't included in the next curve. While it can be surmised by looking at the right-hand bounds of the distributions there are some performance increases, the graph would more directly support your point if a single population of teams (perhaps the 304 that competed at 4 events or the 765 that competed at 3) were tracked across their multiple events instead.

Edit: Please disregard the second part, I had misinterpreted the next figure

[natejo99]

That was fantastic. The paper did an excellent job of pointing out the flaws with the current system and proposing new changes to rectify some of those. I think many teams would benefit from a weekly 8 hours of unbag time and I would love to see FIRST make this change. Thanks for writing this, Jim!

[Nate Laverdure]

Originally Posted by Greg Woelki

...the graph would more directly support your point if a single population of teams (perhaps the 304 that competed at 4 events or the 765 that competed at 3) were tracked across their multiple events instead.

Isn't this what is shown in the following Fig 6?

[IKE]

Jim,
It is really neat to see these thoughts put into a cohesive piece. I know you have been working on and off on this for at least since 2010 talking about ideas and discussing possible formats.

I am a big advocate of the "transition" method. One piece missing from Jim's paper is the propensity of procrastination from some teams. If you have a stop build day, it sets a deadline and the procrastinators will miss that. If you get rid of stop build, the procrastinators will just procrastinate until the event, which can be incredibly detrimental to the week of their first competition.
The "transition" model of a stop build, but weekly test/train/tune/repair sessions give teams some development experience without completing loosing a lot of the intended meaning from the stop build.

As long as there is a stop build day and some limit to access, teams with the drive and resources will continue to build a second robot. Even with no more bag day, many of the highest performers will still build two robots so that one can be used for programming team, and one for training/testing.

My only ask out of this would be that every team get that every week. Please do not give 6 hours for competition unbag week vs. 8 hours for "other" weeks as that would get very confusing.

8 hours each week will be very beneficial though will be a bit of B&T nightmare. If FRC keeps the 2 hour blocks, that would be 4 sessions per week by 6 regional weeks or an additional 24 potential sessions not including displays. We may want to re-think the tag portion of the B&T.

[Jim Zondag]

Originally Posted by Greg Woelki

Great read, thanks for posting!

Minor stats comment: Fig (5) is a little concerning because each curve represents a different population of teams, so it isn't a very clear way to show the trend of increasing performance as teams have attended more and more events. It is unclear to what degree teams from the left-hand side are moving rightwards as they attend more events or if most of those teams simply aren't included in the next curve. While it can be surmised by looking at the right-hand bounds of the distributions there are some performance increases, the graph would more directly support your point if a single population of teams (perhaps the 304 that competed at 4 events or the 765 that competed at 3) were tracked across their multiple events instead.

To answer: Fig(5) is this graph:

Each population of teams is a subset of the previous group.
3114 teams played this year, they all played at least one event (Blue)
Of these teams, 1928 teams played at least 2 events (Red)
Of these teams, 765 teams played at least 3 events (Green)
Of these teams, 304 teams played at least 4 events (Orange)
Of these teams, 58 teams played 5 or more events (Black)
The chart shows the progression of skill improvement by the population with each consecutive event played.

This trend is basically the same every year, regardless of the game, the only change is the magnitude of the vertical axis, which is a function of the annual game design and how many points are available to be scored.

To see the trend more clearly, the dotted black line in Fig (6) shows how the averages of each of these group subsets increases through the season.


So, in a nutshell, if you choose to play late, odds are there are more experienced teams in the house who have progressed in skill while you have been waiting.

[efoote868]

Originally Posted by Jim Zondag

So, in a nutshell, if you choose to play late, odds are there are more experienced teams in the house who have progressed in skill while you have been waiting.

It's interesting to see the 4 event teams every so slightly above the 5 event teams in events 1-3. Does this hold for prior years? Would you think it is a function of event spacing (4 event team plays weeks 2, 4 and 6 and champ while 5 event team plays weeks 1, 3, 5, 7 and champs) while the game is learned, or teams in districts more likely to be a 4 event team?

I like your suggestion for unbag time for each week, I hope it gets implemented for next season.