Re: A novel strategy: Always score for your opponents
 

Wow I took my eyes of this thread for a day and we get but through the ringer.

A couple of point of clarification.

1. For all of those who think Beverly Hills is well funded, I am sorry but you are WRONG. We have to fund raise every penny on our own, our school gives us nothing, our school district gives us nothing, our boar of education gives us nothing. We have to pay full price for our trip to the regional hotel, air fair, lunch at competition, everything. We build our bot with 2 cordless drills, a band saw, and a hot saw (metal version of a chop saw) and that is it. We don't have CNC's, mills, or anything really precise for that matter. This is not the Beverly Hills you see on TV.

2. As of 11 pm on Monday night 1515 is NOT going to Atlanta because we don't have the money for the competition fee or the money to ship the robot there. Keep in mind that we have to tell FIRST by noon Tuesday 13 hours from now. So for of those who say we are "well funded" I take offense.

3. A 6v0 situation is in the highest spirit of FIRST in my eyes. It is the definition of Coopertition and allows all 6 teams to work together to achieve a common goal. Helping each other to succeseed has always seemed like a large goal of FIRST.

4. Those of you who say a 6v0 will not work in the long run, I completely agree. It is a last match of the day strategy to try and help everyone bump up a bit. As I said in my first post you bump up 10% of the field with you at a large regional and even more a a smaller one.

5. I consider the seeding points as much a part of the same this year as the soccer balls we play with. We must take into account what is a W or a L and what goes into determining that. The truth is the W L T record this year means nothing during qualls. I really wish they would show the seeding score on the big screen and not the points in my opinion.

6.There was also multiple Rule clarifications sent to the GDC on this set of rules and they knew about the possibility of this situation and by not changing the rules inherently supported it. This is an intended result of the rules, there are no loop holes, blurry lines, or invisible walls, the GDC knew this would happen.

I think at the end of the day it all comes down to the perspective you take on the situation and what you use to determine a teams W/L record. Persioanlly I don't look at the match score I look at the seeding score for each match.

Here is a good example:
In the traditional since scorning on yourself is a bad thing, but in this game it is a good thing b/c you are advancing your score so now in the "translated definition" we are scoring on the other team by scoring on ourselves. Now by the "translated definition" of scoring on ourselves we are scoring on the other team. So now when I say score on ourselves are we talking about the "translated definition" which is scorning on the other team or the traditional since which is scoring for us. And then what do you use as a standard. The traditional or the "translated". Amongst our team we still dont have a standard we still have to spend 45 seconds clarifying which direction we are talking about every time scoring direction is brought up.

Society has drilled into us that "trowing" a game by "scoring on yourself" is a bad thing, which I agree with but as anyone can see from the complicated paragraph above it is not so black and white with Breakaway and we must re-evaluate theses sosital norms and apply modifications for the context that they are being applied within.

Matthew Forman

Re: A novel strategy: Always score for your opponents
 

Lol these forums are sometimes a terrible place for a valuable discussion because of so many ignorant people involved. I just read the best discussion about 6v0 on someone's facebook. How sad.

Re: A novel strategy: Always score for your opponents
 

I am the one who made that comment. I apologize.
I made that assumption after talking to one of your teammates, when answering the question, "Why the heck are up here?" His response was that you were attending 3 or 4 different regional. To me that is well funded, perhaps I heard wrong. The pits are pretty noisy and I don not have a good noise rejection system.

Unfortunately it is not beneficial for either team, the winning alliance suffers the most. A 0 losing score means no coopertician points. If you look at the seedings the top teams all have lots of coopritician points. Really the strategy best for seeding is for neither team play defence. If each team scored as many points as they can with no effort on anybodies part to prevent scoring then both team would get more seed points then in a 6v0 game.

Sigh. Unfortunately you are wrong. Unless you are in the top 15 seed going into the last match 6v0 will not help. You can at best gain 5 to 10 seed points. Late in the elimination that is not enough to lift only a small handful of teams into the top 9. It will never help, once again the most seeding points are earned in a close high scoring match. Remember the victor gets 2x the lower score in cooptitician points.

Lop sided victories are discouraged by the seeding system, this is why your team was seeded 55 out of 61. Frankly 6v0 is stupid, It is certainly not a fun match to watch, it is unethical and it hurts your seeding.

Did you see the post I made earlier tonight analyzing your teams performance? It is clear the lack of coopritician points hurt you a great deal.

Really coopritician points are the key to seeding, high ranked teams all have lots and they are minimized by lop sided victories.

I would really liked to have seen a no defense match, each team scoring with no interference and bots in zone 2 scrambling to claim balls. If a team gets too far ahead then it would make sense to score some for the opponent. But not enough to loose the match.

I apologized, now it's your turn. :)

Team 957 Mentor and ally of team 1515 in Portland match 86

Re: A novel strategy: Always score for your opponents
 

What about lifters chosen to be on the '0' side? They should not lift at the end, else they will erroneously score for their alliance. Ditto for robots with autonomous scoring capability (like 217). Ditto robots with great defensive capabilities.

These are robot capabilities that IMO should be showcased as much as humanly possible, but in a 6v0 game, they will not be.

It is not moot, as my previous point should illustrate. We were 13th going in to day two, but we were positive that whoever the #1 seeded alliance was going to be, they were going to pick us -- because we were an awesome robot playing the game.

Likewise. The internet poker face makes most people come across as more aggressive than they actually are. Please know that I will never, ever be offended by people disagreeing with me, and that I never, ever intend offense -- even when I say that something is a violation of GP. I say it because I think it's true, and I don't have it in me to mince words.

Five years ago, our rookie year, our robot was deliberately tipped and damaged by another team (who will remain nameless) -- it was not called by the ref in spite of high hitting being illegal. (This was in the tetra game). They hit us high, knocked us off balance, and then hit us again to knock us over. Then they rammed our drivetrain while we were down...

...and they were *laughing* about it after the match.

I think that it is perfectly fair for teams that choose not to implement strategies that they believe are not in the spirit of gracious professionalism to call out teams that do choose to implement them.

You can disagree that it's unfair, but we'll have to agree to disagree.

Just look at the seeding data for the teams that won the Coopertition award:

Regional Team Seed
San Diego 1266 2
DC 339 3
Peachtree 1466 No data, but they were regional finalists
Bayou 1912 1
KC 2345 1
Granite State 1073 2
FLR 1551 1 (woo!)
Oregon 753 + 847 2 and 3 (this was a tie at 44 CPs)
NJ 25 1

The data doesn't lie. Want to be a top-seeded robot? Play 3v3, and be good at it.

Patrick

Re: A novel strategy: Always score for your opponents
 

Excellent data points. Clearly high CP pts imply good seed. With this in mind it is clear that 6v0 is a fools folly.

The clear best strategy for any team at any time is NO DEFENSE every ball scored helps BOTH teams it is shooting yourself in the foot to prevent a ball from scoring in either goal. If one alliance out paces the other in scoring it is a big advantage to score some points for them with the goal of maximizing CP. This means that a well played match will be close enough at the end that EITHER team, through clever play, has a chance at claiming the CP.

I believe that this is the game the game designers had in mind. All teams would get to exercise their scoring capabilities.
and it would be a ball to watch.

It is to bad that nobody seemed to understand this at the Oregon regionals.

Re: A novel strategy: Always score for your opponents
 

While I may agree with your conclusion, I do not believe those data points show that 6v0 don't work. That only shows that those teams that had high cooperation bonus seeded high. Any other conclusions from that data are not inappropriate and not supported.

Until I see a team use a 6v0 strategy in all or nearly all their matches and NOT seed high, I cannot say that 6v0 strategies do not work.

Re: A novel strategy: Always score for your opponents
 

Since high CP is 2/3's of the ranking formula, it makes sense that high CP is strongly correlated to high rank.

Re: A novel strategy: Always score for your opponents
 

Some people want to argue that everyone understood this 6v0 strategy before week 1 started. I watched webcasts from 5 or 6 events and I saw defense being played all day on Friday at all of them. I don't think will happen nearly as much this week (if at all). Which tells me that data from last week isn't useful in determining if it works. Also before week 1 we didn't anticipate the amount of penalties that happened every match which also lends to the 6v0. I think the teams that seeded high as good scorers were benefiting win or lose from their own capabilities on average. But when you consistently put 6 bots working together (no defense also) everyone's points will go up. I'm sure you're going to see a significant data difference between last week and this week. But I do think the best offensive teams will still shake out on top if everyone plays a similar game.

Re: A novel strategy: Always score for your opponents
 

Team update #16 has verified that 6v0 and 3v0 strategies were *not* what the GDC intended, and has added +5 QPs to the winners' score to discourage (and/or completely prevent) this strategy and promote the *competitive* nature of the game.

Re: A novel strategy: Always score for your opponents
 

6v0 isn't necessarily dead. If I am sure beyond a reasonable doubt that my alliance will lose a match, it is still in my interest to pad my ranking points by scoring for the other team. Now the difference is the "winning" team isn't totally screwed in the rankings, especially if they incur a penalty.

Re: A novel strategy: Always score for your opponents
 

I didn't say it was "dead", I said it's clearly not what the GDC intended. It is clear from the update that they intended the game to still be competitive, and for alliances to play to win *each game*.

Re: A novel strategy: Always score for your opponents
 

Ok, it is time for me to weigh in here...

First, I will be up front and say that our team was one of the original "collusion" teams in 2003. We saw nothing in the rules that stated that we couldn't agree with our opponents to not knock stacks down. Since ranking was determined by total scores (not win-loss-tie) it simply made more sense to increase the total scores in the match by not knocking down stacks.

Here is the important part: That was the extent of the "agreement". It was in essence more of a statement: "We will not knock your stacks down unless you knock ours down."

Both alliances went at all other aspects of the competition with the full intention of winning. We knocked the wall, we climbed the ramp and played "king of the hill" - it was still a competition. In the end, all teams that agreed to play this way earned high ranking points, and the matches were still exciting. No matches were "thrown". In fact, in one match, when the opposing alliance thought they could pull off a win by knocking down our stacks with 10 seconds remaining in the match, we drove off the ramp, knocked their stacks down and returned to the ramp to win.

Now to 2010 -

I believe this current ranking system is... problematic. In an attempt to have matches be close (this is what those Coopertition points reward), the GDC have created a system where numerous bizzare strategies are rewarded. For example, 6 teams score for one alliance with the hope of getting a high x-0 score and everyone gets that score, or all three robots on one alliance don't move at all so as to not get any penalties and get the "winners" non-penalized score.

The problem here is that these strategies ARE VALID. They are within the rules. Some here have argued that they are possibly encouraged by the GDC. So we will see them in future regionals. I have no problem with any teams playing within the rules.

The issue is this: This new ranking system has the possibility to really hurt FIRST.

FIRST advertises itself as a "robotics competition". As Dean said at the beginning - FIRST borrowed a page from the world of sports - winners and losers. We have "teams" (teams compete). We award "Finalists" and "Highest Rookie Seed" and "Champions".

Competition is something that can motivate people. After returning from our regional in San Diego, the most common question is "how did you do?" In the past, we could say "we ranked 5th out of 48 teams" or "we were finalists" (it has always been difficult to explain what winning the "Gracious Professionalism Award" means.) Now the ranking system can be very deceptive.

With the 6 vs. 0 strategy or the do nothing strategy, we're not rewarding creative or innovative solutions to the game challenge (as has been demonstrated, you don't need a robot at all to get a high ranking!). We're rewarding creative and innovative solutions to the ranking challenge. This is a much harder sell to young people, teachers, schools, mentors, sponsors and the public at large. ("We built a great robot, but it doesn't matter - we don't drive it and score higher than our opponents because they get penalties.")

One thing to consider that doesn't involve any sort of rule change at all is to change how the game is presented to the audience. Currently, at the end of each match, we announce the scores and winners, but with this qualifying system, this information is useless. What we should announce and celebrate is the qualifying points that each alliance gets.

This way all of these strategies (6 vs. 0, 3 vs. no movement, 3 vs. 3 or whatever new strategy someone might come up with) will at least be somewhat transparent to the audience and the teams.

Example:

A Red 3, Blue 5 score becomes Red 5, Blue 11. (Blue "wins")
A Red 5, Blue 0 becomes Red 5, Blue 5. (Tie!)
A Red 4-2 penalties, Blue 3 score becomes Red 3, Blue 11 (Blue "wins")

In one of those 6 vs. 0 matches, you might see this:
A Red 13-2 penalties, Blue 0 score becomes Red 11, Blue 13 (Blue "wins")?!?

But it doesn't explain this:

A Red 5-3 penalties, Blue 1 score becomes Red 4, Blue 5 (Blue "wins")?!?


I'm at a loss as to how to explain this to anyone - especially those outside of FIRST.

I can not imagine that the GDC thought that scores would be so low and penalties would be so high as to cause this problem. Why would you design a system where you could be assured a "tie or win" by blocking your own goals and not moving?

San Diego:

It seems like no one in SD figured out the bizzare strategies at all. In fact it seems as if most alliances were unaware that playing defense during the qualifications only served to lower your points. They just played the game. Defense, offense and everything else.

AND IT WAS GREAT. BREAKAWAY is the best game from FIRST since 2006. Scoring is easy. The game is easy to follow. It can be exciting.

The ranking system on the other hand, is the worst I've seen - with 2002 a close second. (In both of these games, elimination rounds are different than the qualification rounds.)

We survived 2002 (and the tape-measure rule change - mid-season rule changes have happened before!) and we survived the "collusion" arguments of 2003. We'll survive 2010's ranking system. The question is will FIRST continue to survive if we can't explain to people who we invite to the events (and sponsors and possible future team members) why teams do what they do on the field and why winning is losing.

Final note: I do not know exactly how our team will handle these strategy questions in Los Angeles. I do know I very, very much like Travis's definition of "coopertition": Cooperate behind the curtain, compete in front of it.

-Mr. Van
Coach, Robodox

Re: A novel strategy: Always score for your opponents
 

Mr. Van,

Long post, hopefully accompanied with a giant sigh of relief from Update #16...

Re: A novel strategy: Always score for your opponents
 

YES! In the time it took me to actually write and put up the post, Update #16 was released. It makes most of my post moot.

Thanks GDC.

-Mr. Van