paper: The Penalties will continue until Morale Improves

[Swan217]

Thread created automatically to discuss a document in CD-Media.

The Penalties will continue until Morale Improves by Swan217

[JesseK]

Nice paper with interesting analyses. I didn't realize there were issues with '05-'07 though. Ha.

[Chris Hibner]

That was a fun read.

I wish we could go back to the days where real-time scoring wasn't needed. There were numerous games where you could look at the field and within 2 seconds know who was winning, unless it was REEEEAAAALLLY close (i.e. which goal had more balls in it - that team is winning).

[Garvs72]

Great article, though I disagree with you about defense. So long as a good defense doesn't completely ruin the flow of a game, I think it adds a lot of depth, makes the game more exciting, and forces teams to build robots that can do more than just repetitive cycling.

[Swan217]

Originally Posted by Garvs72

Great article, though I disagree with you about defense. So long as a good defense doesn't completely ruin the flow of a game, I think it adds a lot of depth, makes the game more exciting, and forces teams to build robots that can do more than just repetitive cycling.

I'm not suggesting that defense doesn't have its place. In fact, defense was very influential in Florida during Ultimate Ascent. I agree with you - I'm saying that the best games are when the defense doesn't ruin game flow.

[cadandcookies]

Originally Posted by Swan217

I'm not suggesting that defense doesn't have its place. In fact, defense was very influential in Florida during Ultimate Ascent. I agree with you - I'm saying that the best games are when the defense doesn't ruin game flow.

Ruin might be describing it a little too lightly. I've been seeing matches where defense took the game flow out back and beat the snot out of game flow with baseball bats and bricks.

[Jim Zondag]

Wow!.....Wow!!
Great job on this paper.
It is great to hear this from the referee perspective and you have a tremendous amount of excellent details on the long history of the progression of this problem. I agree that 2008 was the beginning of the dark times for rules. The good core game design of the past 3 years kind of mask this, since the penalties were more of a sidebar, hiding how bad the penalty rules actually were, since they were more avoidable.

I have always felt that the core problem with this entire topic is a volume thing. FIRST keeps adding rules in an attempt to control gameplay. As a result, the refs are overburdened watching trivial things with black and white definitions and not properly policing the grey areas of robot interaction with their full attention. Your perspective seems to reflect this same observation.

This is what happens when Engineers try to design a sport. Engineering is all about strict rules and controls, sports are all about fair play, motivation, balance. There are lots of grey areas in sports, and this is why we need refs. Not for black and white, we need them most for the grey.

If you are a runner in baseball and you get hit with a ball, are you out or are you safe? It depends where the ball came from. Refs decide.

If your opponent's ball lands in your machine in Aerial Assist, do you get a penalty? Yes, always, even with the rules modifications. Fail. It should depend on where it came from. If an opponents rebound lands in your robot, why is this your team's fault?

All rules in an interactive game MUST have situational dependency. This is what the refs should watch, not the HPs finger tips.

FIRST likes rules. They have lots of rules about how to build robots, lots of rules about how to make bumpers, lots of rules about when you can work on your robot, lots of rules about how to get penalties on the field; rules, rules, rules, rules, rules. I think on this topic, less is more in every category. Most of these rules add little actual value and just make everything more difficult for all of us.

Quote of the day from your paper:
"most teams would rather have chaotic good rules rather than lawful evil rules."
Amen brother!

In about 4 weeks, the new VEX game will be released in Anaheim. I would bet $1000 that there will not be any 50 point tech fouls in their game. The VEX GDC has ACTUAL competitors on the team, so their rules make sense. Just sayin'

[DampRobot]

Originally Posted by Jim Zondag

Wow!.....Wow!!
Great job on this paper.
It is great to hear this from the referee perspective and you have a tremendous amount of excellent details on the long history of the progression of this problem. I agree that 2008 was the beginning of the dark times for rules. The good core game design of the past 3 years kind of mask this, since the penalties were more of a sidebar, hiding how bad the penalty rules actually were, since they were more avoidable.

I have always felt that the core problem with this entire topic is a volume thing. FIRST keeps adding rules in an attempt to control gameplay. As a result, the refs are overburdened watching trivial things with black and white definitions and not properly policing the grey areas of robot interaction with their full attention. Your perspective seems to reflect this same observation.

This is what happens when Engineers try to design a sport. Engineering is all about strict rules and controls, sports are all about fair play, motivation, balance. There are lots of grey areas in sports, and this is why we need refs. Not for black and white, we need them most for the grey.

If you are a runner in baseball and you get hit with a ball, are you out or are you safe? It depends where the ball came from. Refs decide.

If your opponent's ball lands in your machine in Aerial Assist, do you get a penalty? Yes, always, even with the rules modifications. Fail. It should depend on where it came from. If an opponents rebound lands in your robot, why is this your team's fault?

All rules in an interactive game MUST have situational dependency. This is what the refs should watch, not the HPs finger tips.

FIRST likes rules. They have lots of rules about how to build robots, lots of rules about how to make bumpers, lots of rules about when you can work on your robot, lots of rules about how to get penalties on the field; rules, rules, rules, rules, rules. I think on this topic, less is more in every category. Most of these rules add little actual value and just make everything more difficult for all of us.

Quote of the day from your paper:
"most teams would rather have chaotic good rules rather than lawful evil rules."
Amen brother!

In about 4 weeks, the new VEX game will be released in Anaheim. I would bet $1000 that there will not be any 50 point tech fouls in their game. The VEX GDC has ACTUAL competitors on the team, so their rules make sense. Just sayin'

Jim, if you were to design this years game with "chaotic good rules" rather than "lawful mean rules," what kinds of rules would you make?

[Basel A]

Originally Posted by DampRobot

Jim, if you were to design this years game with "chaotic good rules" rather than "lawful mean rules," what kinds of rules would you make?

Perhaps for possession to be called, the robot would have to complete the process of the catch.

[BJC]

Originally Posted by DampRobot

Jim, if you were to design this years game with "chaotic good rules" rather than "lawful mean rules," what kinds of rules would you make?

I'm not Jim but I have thought about this enough that I can give a non-comprehensive answer.



1. It shouldn't matter what zone a robot is in for an assist to count.

- When a robot first acquires a ball that is currently considered a 0 point assist. This should simply be nothing. As such only 2 assists would be possible per cycle.

-For simplicity, each assist should be the same (10 or 15) number of points.

-For simplicity, each assist should be scored when it happens.

--This new assisting method is much more intuitive for teams, spectators, and refs. Trying to control where the assists happen via field lines is dumb and a non-obvious thing to be looking out for. Did the ball go between those two robot? Now its an assist, no questions asked.

2. Increase catching points to 20-25 points.

--Catching is one of the most exciting things to watch in this game, however, it is undercosted versus it difficulty to execute.

3. Eliminate hot goals in auto, but allow assists.

--Aside from not relying on software to properly display lights, taking the idea of assists in teleop and reusing it for auto rather than introducing new scoring methods (hot goal) keeps the scoring simpler to understand and keep track of. Plus, seeing all three robots pass their balls around to each other twice before scoring would be awesome.

4. Human players should be allowed to hold balls near the side of the field rather than waiting by the pedestal.

-Human players would use their eyes and common sense to watch when a cycle ends and determine if they can start a new one.

-A foul would be called for entering a third ball into play and repeatedly doing so would be a red card.

--This will decrease the amount of time robots spend waiting for balls.

5. Two balls should be allowed on the field at a time.

--Currently only one robot can play offense at a time. Two balls should allow at least two of an alliance's robots to stay occupied attempting to score points. This, in turn, will decrease the amount of defense played and make everything faster paced more enjoyable to watch.

6. Move human player areas further from the field or make use of a polycarboate wall.

-Their should be no foul for a robot extending 20" (legal amount) outside the field.

--Safety related penalties can be completely avoided through good game design. Almost all collecting designs force robots to extend outside their bumper zone. If we don't want robots to be able to touch humans then no humans should be allowed within 2ft of the field, simple as that. Especially when considering that the field is 3ft narrower than usual this year this should not be difficult to implement.

7. A robot with a ball cannot be given a robot interaction penalty while attempting to score.

--Teams should never be given penalties for attempting to complete the game challenge.


I think my most controversial change would be adding a second ball to gameplay which could make keeping track of assists and such tricky. However, allowing more than one robot to be playing offense at a time is absolutely critical to having a successful game and naturally mitigates defense without additional rules. If necessary, I would sacrifice other aspects of the game in order to achieve this.

Again, not comprehensive.
Cheers, Bryan

[Tristan Lall]

Originally Posted by Jim Zondag

I have always felt that the core problem with this entire topic is a volume thing. FIRST keeps adding rules in an attempt to control gameplay. As a result, the refs are overburdened watching trivial things with black and white definitions and not properly policing the grey areas of robot interaction with their full attention. Your perspective seems to reflect this same observation.

This is what happens when Engineers try to design a sport. Engineering is all about strict rules and controls, sports are all about fair play, motivation, balance. There are lots of grey areas in sports, and this is why we need refs. Not for black and white, we need them most for the grey.

If you are a runner in baseball and you get hit with a ball, are you out or are you safe? It depends where the ball came from. Refs decide.

If your opponent's ball lands in your machine in Aerial Assist, do you get a penalty? Yes, always, even with the rules modifications. Fail. It should depend on where it came from. If an opponents rebound lands in your robot, why is this your team's fault?

All rules in an interactive game MUST have situational dependency. This is what the refs should watch, not the HPs finger tips.

That's true to a point, but I think it's possible to swing it too far in the direction you propose. Add too many situational dependencies, too many judgments of intent, and you end up with something like a boxing score, where it's often impossible to discern what the judges were thinking. Every difficult judgment presents an opportunity for inconsistency, and FIRST competitors are certainly quick to decry inconsistent officiation.

Maybe the real issue with the ball possession penalty is that it could have been disincentivized rather than penalized. Penalties imply an infraction, and the conversation is naturally about equity: who was wronged? Disincentives don't have to be about that at all. It could simply have been a feature of the game that if an opponent's ball lands in your robot, for any reason at all, you lose some points—and get to control that ball for a while.¹ Teams will have no reason to feel wronged (as they do now), and will instead develop designs and strategies to avoid that situation. What's more, it would be easy for the referees to judge.

You're certainly right that too much of the referees' attention is devoted to trivialities.

Originally Posted by Jim Zondag

FIRST likes rules. They have lots of rules about how to build robots, lots of rules about how to make bumpers, lots of rules about when you can work on your robot, lots of rules about how to get penalties on the field; rules, rules, rules, rules, rules. I think on this topic, less is more in every category. Most of these rules add little actual value and just make everything more difficult for all of us.

Definitely true, and definitely something that they need to address. The rules should be as exact as possible where precision adds value, and as lenient as possible where a general constraint would satisfy the rulemakers' intent.

The rules for bumpers and pneumatics have awful return on investment, and are long overdue for an overhaul.

¹ Obviously this would have to be studied in the context of the game as a whole, because it might lead to certain strategies dominating. I don't propose it as a hypothetical remedy for Aerial Assist, but merely offer it as an example of a different way of managing gameplay behaviour.

[fb39ca4]

Originally Posted by Tristan Lall

The rules for bumpers and pneumatics have awful return on investment, and are long overdue for an overhaul.

I have no experience with pneumatics, but why do you say bumper rules are subpar?

[Tristan Lall]

Originally Posted by fb39ca4

I have no experience with pneumatics, but why do you say bumper rules are subpar?

Don't get me wrong: the bumper rules have improved markedly each of the last couple years. 5 years ago they were basically a travesty. (Here are some of my suggestions from that era; I would still stand by most of that. I grant that there has been a lot more robot violence this year, which increases the utility of a bumper.)

Despite the improvements, there are still too many vague bumper rules. Bumpers are fundamentally overspecified for the limited purpose they serve, and yet those specifications are often underdefined or (more so historically) contradictory. This is hard to understand and hard to enforce fairly and with a straight face¹ and leads to a lot of effort (on the part of teams and officials), for very little benefit.

It has long been my experience that bumpers and pneumatics take up a disproportionate amount of a lead inspector's time, because those rule sets are the ones most likely to require complex interpretations that turn on very fine details. They are consequently the most controversial, and therefore induce the most argumentation and the most animosity.

¹ As an inspector, I've been criticized, understandably, for rigourously enforcing useless bumper requirements like frame perimeter support. Some inspectors didn't enforce that rule—again understandably, because it was stupid—but this led to inconsistency between events.

[GaryVoshol]

Dan, how did you forget minibots? Those sensors worked flawlessly too.

I love your "Do I look like an idiot?" rule.

[Swan217]

Originally Posted by GaryVoshol

Dan, how did you forget minibots? Those sensors worked flawlessly too.

I love your "Do I look like an idiot?" rule.

I thought about that, but it didn't neatly fit into my narrative, & I didn't pay enough attention that year to be able to analyse it carefully.

There need to be at least "DILLaI" rule in FIRST each year.

Originally Posted by cglrcng

Hope to see you back officiating in the future

Doubt it. The Return on Investment is definitely not worth it this year, & I'm having much more fun doing www.TheRoboShow.net. Out of the problems in FIRST I have the abilities to fix, I rank them as 1) New Media, 2) Regional Planning, 3) Inspecting, 4) DJs, 5) Refs.

Originally Posted by Tristan Lall

Despite the improvements, there are still too many vague bumper rules.

I think the bumper situation can be fixed with one little statement:
"Any deviations to these bumper rules that are deemed by the LRI to be more rigid or a higher quality than the written rules can be allowed on a regional to regional basis."

This would mean a team that has a superior bumper configuration doesn't have to scramble to recreate bumpers in a lower quality, but still indicates this is a temporary solution & protects LRI's from the criticism "The last regional let me do it"

Originally Posted by Jim Zondag

This is what happens when Engineers try to design a sport. Engineering is all about strict rules and controls, sports are all about fair play, motivation, balance. There are lots of grey areas in sports, and this is why we need refs. Not for black and white, we need them most for the grey.

This either makes me a really good engineer or a really bad one, because I don't try to control what I design, I try to work around and anticipate alternatives. I design with the assumption that I will have to change the design quickly at some time, and so try to make things easy to adjust to.

The inherent reason FIRST has so many rules is to take the inconsistency out of the officiating. When issues are black & white, there's no room for interpretation, and therefore you have a more consistent result, regardless of the quality of that result.

The reason why the officiating is that the officials are inconsistent. In Major League sports, you have an official who's job is to critique the other officials, insure they're doing their job correctly, offer improvements. As the head ref gets more & more responsibilities lately, you don't have enough time or eyes to do enough critiquing of the referees under your command. From the regionals I have seen around the country, I've seen certain referees consistently calling penalties wrong, or sometimes even blatantly calling things in favor of their own teams. These usually concern defensive strategies.

But instead of improving volunteer quality, or designing a game that discourages defense (such as the past few years, IMHO), they go with more rules to micromanage teams & referee calls.