Time to re-open an old can of worms

After some hotly disputed decisions at the recent Annihlator event (mainly by me being a loud-mouth ), it seems like a good idea to *publish* some "judging guidelines" that the judges at the next RoboWars event will be required to read and understand beforehand, so everyone knows why they won or lost a match.

I'll open it up for a bit of discussion here before finalising them and putting them on the RoboWars home page along with the Tech Regs and a "Tournament Proceedues document" (that I'll discuss in another thread).

We had this dicsussion once before on the old forum, and some good points were raised, but I'd like to see if anyone has any new insights now we've had a few events and before I finalise the judging guidelines for our next event.

The discussion at the old forum is here
http://www.robowars.org/abblforum/viewtopic.php?t=149&sid=39681d7e97c8479222684ea0e0130a45
so please have a browse over that to get a feel on what we talked about last time.

Heres an example of some modified judging-guidelines document that ours will be based on from Steel Conflict in the USA.

STEEL CONFLICT - TOURNAMENT JUDGING GUIDELINES
1. Tournament Judges

A panel of judges will determine the winner of matches in which time expires before one
combatant is defeated as defined in the Tournament Rules and Procedures. The number
of judges on the panel shall be an odd number to eliminate the possibility of ties.
Judges' decisions are final.

1.1. Qualifications
Judges must be completely familiar with the Official Rules governing the tournament.
Judges must be familiar with the scoring system and Judging Guidelines as defined here.
Judges must be reasonably conversant with combat robot design and construction.

1.1.1. Responsibilities
Each judge shall officiate in a given robotic combat Tournament with complete impartiality
and fairness, respecting and abiding by the rules that govern that tournament, in the true
spirit of sportsmanship.
Each judge is responsible for keeping track of the Combatants during the course of the
match. Many Combatants look similar, it is the responsibility of each judge to keep them
straight and award points correctly.
Each judge is expected to take careful note of existing damage when Combatants enter
the arena. Existing damage must not be counted against a Combatant in the event of a
judges' decision.
Judges must watch the entire match and award points accordingly. Judges are allowed
(and encouraged) to take notes during a match to assist in scoring.

1.1.2. Judge Foreman
One member of the judges panel will be designated the Judge Foreman. The Judge
Foreman will ensure that all other judges are conforming to the guidelines as set forth
herein. The Judge Foreman may or may not participate in scoring judges decisions,
depending on the number of judges available.
The Judge Foreman will ensure that all Combatants conform to the tournament rules.
Warnings and instructions from the Judge Foreman will be issued to the Combatants
verbally during the matches. Should a Combatant fail to comply, the Judge Foreman will
stop the match and the violating Combatant shall be deemed the loser.
The Judge Foreman will determine the point at which a knockout countdown is to begin
based on the strict interpretation of the rules. When a 10 second countdown is warranted
by the Judge Foreman, the non-responsive Combatant will be notified and the countdown
will begin. The arena announcer will start the countdown at 10 and count down to 0.
If the non-responsive robot has not displayed sufficient translational movement as described in
the rules, the Combatant will be declared the loser.

1.1.3. Conduct
Judges will clearly identify themselves as such.
Judges will not consult with each other or the audience while watching or scoring a
match.

2. Judges' Decisions: Scoring
When a match does not end in the elimination of one of the Combatants as defined by the
Tournament Rules and Procedures the winner shall be determined by a Judges'
Decision. In a Judges' Decision the points awarded to the Combatants by the panel of
judges are totaled and the winner with the majority of points is declared the winner.

2.1. Point Scoring System
Points are awarded in 4 categories:
· Damage
· Aggression
· Control
· Style

All points must be awarded - each judge will determine how many points to award each
Combatant in each category, according to the Judging Guidelines (see below).

2.2. Judging Guidelines
2.2.1. Scoring Aggression
Aggression scoring will be based on the relative amount of time each robot spends
attacking the other.
Attacks do not have to be successful to count for aggression points, but a distinction will
be made between chasing a fleeing opponent and randomly crashing around the arena.
Points will not be awarded for aggression if a robot is completely uncontrollable or unable
to do more than turn in place, even if it is trying to attack.
Sitting still and waiting for your opponent to drive into your weapon does not count for
aggression points, even if it is an amazingly destructive weapon.

Awarding Aggression Points
o A Combatant who attacks a full-body spinner (e.g. intentionally drives
within the perimeter of the spinning weapon) is automatically considered
the aggressor and awarded a 3-2 score in the case where both robots
consistently attack, or both robots consistently avoid each other.
Note: a Combatant is considered a "full body spinner" if the robot cannot be attacked
without moving within the perimeter of the spinning weapon.

2.2.2. Scoring Damage
Judges should be knowledgeable about how different materials are damaged. Some
materials such as Titanium will send off bright sparks when hit but are still very strong and
may be largely undamaged. Other materials such as Aluminum will not send off bright
sparks when hit. Judges should not be influenced by things like sparks, but rather how
deep or incapacitating a "wound" is.

Judges should be knowledgeable about the different materials used in Bot construction
and how damage to these materials can reduce a Bots functionality. Judges should not to
be unduly influenced by highly visual damage that doesnt affect a Combatant's
functionality effectiveness or defensibility. For example, a gash in a Combatants armor
may be very visible but only minimally reduce the armor's functionality.

Judges should look for damage that may not be visually striking but affects the
functionality of a Combatant. For example:
· a small bend in a lifting arm or spinner weapon may dramatically affect its
functionality by preventing it from having its full range of motion
· bent armor or skirts can prevent the Combatant from resting squarely on the floor,
reducing the effectiveness of the drive train
· A wobbly wheel indicates that it is bent and will not get as much traction.
· Cuts or holes through armor may mean there is more damage inside.
Damage suffered to robots can be grouped into the following classifications:

Trivial
· Flip over (or being propelled onto bumper, ramp, or other obstacle) causing no
loss of mobility or loss of weapon functionality.
· Direct impacts which do not leave a visible dent or scratch.
· Sparks resulting from strike of opponent's weapon
· Being lifted in the air with no damage and no lasting loss of traction.

Cosmetic
· Visible scratches to armor.
· Non-penetrating cut or dent or slight bending of armor or exposed frame.
· Removal of non-structural, non-functional cosmetic pieces (dolls, foliage, foam, or
"ablative" armor).
· Damage to wheel, spinning blade, or other exposed moving part not resulting in
loss of functionality or mobility.

Minor
Flip over (or being propelled onto bumper or other obstacle) causing some loss of mobility
or control or making it impossible to use a weapon.
· Intermittent smoke not associated with noticeable power drop.
· Penetrating dent or small hole.
· Removal of most or all of a wheel, or saw blade, spike, tooth, or other weapon
component, which does not result in a loss of functionality or mobility.
· Slightly warped frame not resulting in loss of mobility or weapon function.

Significant:
· Continuous smoke, or smoke associated with partial loss of power of drive or
weapons.
· Torn, ripped, or badly warped armor or large hole punched in armor.
· Damage or removal of wheels resulting in impaired mobility
· damage to rotary weapon resulting in loss of weapon speed or severe vibration
· damage to arm, hammer, or other moving part resulting in partial loss of weapon
functionality.
· Visibly bent or warped frame.
· Major: Smoke and visible fire.
· Armor section completely removed exposing interior components.
· Removal of wheels, spinning blade, saw, hammer, or lifting arm, or other

major
component resulting in total loss of weapon functionality or mobility.
· Frame warping causing partial loss of mobility or complete loss of functionality of
weapon system.
· Internal components (batteries, speed controller, radio, motor) broken free from
mounts and resting or dragging on the arena floor.
· Significant leak of hydraulic fluid.
· Obvious leaks of pneumatic gases.

Massive
· Armor shell completely torn off frame.
· Major subassemblies torn free from frame.
· Loss of structural integrity - major frame or armor sections dragging or resting on
floor.
· Total loss of power.

Post-Match Inspection
Judges may request the combatants to demonstrate operability of their robots drive train
and/or weapon following the end of the match, before the arena doors are opened.
Judges may inspect the Combatants robot after a match to determine how best to award
damage points. The judges will not handle the Combatants robot. The driver or a
designated team member will handle the Combatants robot. A member of the opponent's
team may be present during any such inspection.

Damage self-inflicted by a robot's own systems and not directly or indirectly caused by
contact with the other robot or an active arena hazard will not be counted for scoring
purposes.

---- END OF Steel Conflict Guidelines ----

Add to these our additional categories of

** Control ***
---------------

Control

Each robot is awarded a point in the Control category based on its demonstration of control of itself, its opponent, and the match. Length of time during which control is demonstrated, and the degree to which control is demonstrated, shall both be taken into account in the final determination of which robot showed superior control and is to be awarded the control point.

The degrees of possible control examples are...

Helpless. The robot is unable to demonstrate control, is barely mobile, or moves randomly or intermittently.

Occasional control. The robot may achieve a few shoves or wedge lifts on the opponent, or brings its weapon to bear a few times. It may dodge or escape a few of its opponent’s attacks. It shows some control of its position relative to its opponent. It often has to correct its aim while driving.

Adequate control. The robot achieves several attempted attacks, and some effective attacks. It avoids some of the opponent’s attacks. it only occasionally has to correct its driving aim

Good control. The robot executes many attempted attacks or several effective attacks. It dodges or escapes regularly. It generally drives to its target without correcting its aim. It can often compensate for its opponent’s motion.

Excellent control. The robot executes many quality attacks. It rarely misses its target. It dodges or escapes regularly and recovers quickly. It is rarely out of position relative to its opponent.

Transcendent control. The robot attacks at will and can stay on its opponent. It dodges most blows or escapes almost all attempts to lift, pin, or grapple. It rarely misses and does not need to stop to take aim or correct aim.

and the ever-controversial

*** Style ***
-------------

This is the curly one. Style can mean totally different things to different people, and in this circumstance, this is exactly as it should be. What we would like this category of points to indicate is how "cool" the judge (and hopefuly the audience) thought the robot was in its battling.

Style can incorporate an effective strategy, where an obviously outpowered bot reliably executes a manever designed to lessen the impacts of the other bot. It can incorporate using a tactic that is particular effective against that particular style of opponenet (high-siding a wedge for example, or keeping a spinner rammed into the wall to prevent it from spinning up)

In most cases style could be as simple as a team having a bot that "looks cool", whether through innovative mechanical engineering, or a novel themed paint job. More typically, style points would be given to the bot that shows the greater level of creativity, engineering, uniqueness, articulation, or motion. a raw-steel box on wheels would not be likely to win many style points.

the style category can be summed up in one word.. Mechadon. an unquestionably stylish, cool robot, that nonetheless didnot do very well in its fights. everyone remembers and loves it simply because it looks and moves so damn cool.

This is what style is supposed to encourage. If someoe chooses to front an arena full of wedges, lifters, bar-spinners, flippers and other "conventional" desgns, with a bot that defies description, then they will quite likely get the style points. Think WarHead,

Quite often, "Stylish" robots sacrifice some "effectiveness" in order to look cool. Cosmetic additions use weight, and experimental designs are hard to stick with with when going up against a well-proven wedge or rammer. By awarding a point for style, we hope to encourage people to work around these limitations and build something that will cause spectators to keep coming back for more..

---

One of the best recent examples of "Style" that I can recall, was Pinscher's jaws grabbing onto Arachnophobia in mid-spin with magnificent timing. You can hear the crowd roar its approval in the video's, which highlights what the "Style" point is all about to me. something indefineable that nevertheless is "cool".

Remember style is a *tiebreaker* point - so looking cool and having a snazzy paintjob will not win you a match if you perform poorly in the other areas, but it may help to swing a close match one way or another.

OK, thats enough for a mega-post to start the brow-beating. If anyone has any suggestions, improvments, or criticisms.. go for it.

regards

Brett

Ill let others sort it out but Brett please. SHORTER messages please. i can read alright but geesgh. .
Seriously though. I dont really mind either way how its done as long as its fair and theres reasoning behind each decision in which i know why it happened.
_________________
Andrew Welch, Team Unconventional Robotics

sorry its a long message, but I didnt think there would be much difference between one long message and a few short ones, and I wanted to posted the entire judging guidelines for discussion..

i 100% agree with all of the criteria above especially control and damage, bar style still.

i think strategy would be a more fitting basis for judging and as you wrote i think this would be a perfect example of strategy -

quote:

---

Style can incorporate an effective strategy, where an obviously outpowered bot reliably executes a manever designed to lessen the impacts of the other bot. It can incorporate using a tactic that is particular effective against that particular style of opponenet (high-siding a wedge for example, or keeping a spinner rammed into the wall to prevent it from spinning up)

---

quote:

---

One of the best recent examples of "Style" that I can recall, was Pinscher's jaws grabbing onto Arachnophobia in mid-spin with magnificent timing. You can hear the crowd roar its approval in the video's, which highlights what the "Style" point is all about to me. something indefineable that nevertheless is "cool".

---

quote:

---

More typically, style points would be given to the bot that shows the greater level of creativity, engineering, uniqueness, articulation, or motion. a raw-steel box on wheels would not be likely to win many style points.

---

quote:

---

so looking cool and having a snazzy paintjob will not win you a match if you perform poorly in the other areas

---

Sorry Glen, I know you dont like the style point much, and you make your case fairly well, but I still think its a valuable addition for the reasons below..

quote:

---

i think strategy would be a more fitting basis for judging

---

quote:

---

"One of the best recent examples of "Style" that I can recall, was Pinscher's jaws grabbing onto Arachnophobia in mid-spin"

that yells strategy to me.

---

quote:

---

the other factors of style to me come back to good strategy and luck

---

quote:

---

well you have the best looking award at robowars, is not that encouragement enough to build good looking bots?

---

quote:

---

take for example a robot such as maybe, the original version of the subordinate, a metal wedge with nothing pretty on it, if that went into the fight with scoopy and viper say and wedges them around rather convincingly well why would the good looks and engineering of the scoopy and viper get them points if they where getting "owned" and really if it beat on them that much would style points really matter? example...

---

OK, Back on topic guys..

I've moved the interesting discussions on what people are doing with the visual styling of their robot over to a new thread. by all means keep discussing what robots you think look good and ideas there, but here, its back to comments on the judging guidelines if anyone has any other thoughts please.

A very good comprehensive judging guidelines, similar in some respects to ones I read pre-Marayong.

My only comment is...

Good luck to whoever gets that onerous task.

I stumbled across this as a bit of a side effect of the bun fight the poms are having on there forum as a result of our decision to match the Americans feather weight limit.





In the arena the robots face each other in a timed fight to the death. If within the allotted time one of the robots becomes immobilised, the other robot will be declared the winner. If neither of the robots have been immobilised our panel of expert judges will declare the winner on a points system using four judging criteria as follows:
1. DAMAGE. (Weight of 4) 1-5 points x 4 = Score for Damage.
2. AGGRESSION. (Weight of 3) 1-5 points x 3 = Score for Aggression.
3. CONTROL. (Weight of 2) 1-5 points x 2 = Score for Control.
4. STYLE. (Weight of 1) 1-5 points x 1 = Score for Style.
The winner will be the robot who has scored the highest number of points if there has been no clear winners. This scoring system gives a fair and unbiased chance for every robot- irrespective of size, weight or power- to win!
(Please refer to Rules and Guidelines for more details on the judging of combat situations.)


This appears to be the system used in the Mentorn Robot Wars series and it seems to me to be a good way of quantifying and deciding who won, giving more weight to the damage and just a tiny reward for the style point without dropping it all together.