Re: A novel strategy: Always score for your opponents
 

If a cooperative game was intended where 6 robots work as a team and share the score/result of their combined efforts, then such a game should have been designed and played that way. It might be great idea. But to pretend that a game is a traditional competition between two teams when in fact the scoring system rewards you for not competing, that is just confusing and discouraging. Regardless of which side of this issue you are on, the turmoil it is creating is not good for the health of FRC, and that makes it a bad idea no matter what.

Re: A novel strategy: Always score for your opponents
 

The best teams will be able to take advantage of the system to get themselves into one of the top 8 spots. They will score for themselves as well as their opponents, but win the match so they get 2X their opponent's score. As long as you are winning (and you don't penalize yourself to death) scoring for your opponent will get you more points.

As far as scouting, Team 343 will be looking at good scorers and hangers, and teams that understand strategy . Teams that get tons of penalties will not be picked by 343.

As far as audiences watching, if you explain the winning alliance gets their score plus 2X the losers score, then they understand.

Or it could all just be a war where we make sure we kill the same number of people on both sides.

Re: A novel strategy: Always score for your opponents
 

Many people are saying that because a 6 v 0 qualifying match is better for both alliances than a 3 v 3, teams will always do it or those that 'cheat' will do better overall. They will realise that in the playoffs, the 6v0 format is unviable and will obviously be scrapped. Thus, teams that played good defense (even if that's bad for seeding points) will be picked because that is a strategy that can win games, as opposed to leading to higher scores. Also, watching a 6v0 match, nothing seriously stands out (well, maybe, but moving on..). When you watch one team play great defense on another, it looks good, and you'll check out that team more to see if you'll want to pick them later. Furthermore, all teams that played 'for real' in qualifying will have a strategic advantage going into eliminations, because they have played 'real' matches and know what they have to do to win. So basically, this 'cheating' may be good for some teams because they seed higher, but the real winners are those that win elimination matches, which will more likely be the teams that play 3 v 3, even in qualifiers.

In this way, it may come down to a balance of two types of teams:
1. Teams that play 6v0 in order to get higher seeds and hopefully be able to select the best teams to their alliances and
2. Teams that play 3v3 in order to look good, play better, and get picked on alliances. Which team you want to be is the same decision as 'do you want to go over the bump?' or 'do you want to elevate?' It's a decision that is merely strategic, and is designed for you to have the greatest chance of reaching Atlanta and winning the World Championship. It all comes down to a balance of specializations. Will your robot get the #1 seed and pick the best teams, being the figurehead of a winning alliance? Or will your robot play hard and strong on opposing teams, leading to you getting picked by the top alliance, and actually leading your alliance to victory?

Ideally you want to be a type-2 robot that gets seeded in the top 8, so that you'll have 3 robots that play the real elimination game. However, that really looks highly unlikely. It's actually a bit like playing defense: any robot can play defense or play 6v0 decently well, but it's the offense or the 3v3 that makes your robot stand out. I'm not trying to say that any robot can or will absolutely specialize one way or another, because that's impossible, because either way lots of your points will be scored by scoring balls in a goal, yours or the other alliance's.

I hope you enjoyed that 'brief' post. I thought it was something a lot of people were missing. And 2337 will be a type-2 robot.

Re: A novel strategy: Always score for your opponents
 

I am not saying I necessarily agree, or with or like the system we have been handed. However, I personally don't see any moral dilemma in playing a match as a 6v0 (not arranged with the opponents) or scoring for our opponents to maximize our seeding points. I have not yet talked with my team about this, and will certainly be doing so before implementing either strategy to make sure everyone feels the same way.

We (assuming my team agrees with me) will not be using 6v0 in every match, or even in most matches, as strategically I don't think it is the best way to maximize seeding score and I don't think it is the best way to prepare for eliminations. We will be playing at least most of our matches to win, but to win close if possible.

I feel that there is no such thing as "running up the score" in sports. If you don't want the opponent to score so many points, then you should stop them. I have been on a basketball team that lost a game by 100 points. While that experience sucked pretty bad, it wouldn't have been made any better if the opponent had stopped scoring at some number like a 30 point lead or a 50 point lead. Personally, I use experiences like this to push me to work harder.

I feel the same way about self-scoring in FIRST. If you want to stop your opponent from doing this then you better bring the defense to stop it or the offense to make it unwise. If I don't give them a reason not to, I fully expect my opponent to play the game in a way that will maximize their Seeding Points. This is the game we have been given, and I will not fault my opponents for playing it.

Re: A novel strategy: Always score for your opponents
 

I wish FIRST did something more simple for Coopertition, like "if you put this ball on a special goal both teams double their SP! YEAH COOPERTITION" rather than something that changes the whole game like this. Or even "hey I'm going to stop playing defensive tackle in the middle zone so you guys can get some balls too!".

Right now, I have no idea what my team's going to do regarding 6v0 or whatever. All I know is that I hate that I have to even think about it.

Re: A novel strategy: Always score for your opponents
 

Right. In both cases, it was more of a 3v0 situation where the weaker alliance essentially refused to play against a stronger alliance.

-----------

I want to be clear here that there is no bad blood between me and anyone. My team did great, and I'm tremendously proud of them. 22 hanging points is the highest in week one, and in addition to regional champion and top seed, they won the judge's award and the coopertition award. This is by far the best we have ever done, and we did it all through 3v3 play.

-----------

My goal in posting in this thread is to make people realize that there are absolutely situations where teams are *hurt* by 6v0 or 3v0 play, and it may very well hurt FIRST itself.

On 3v0:
1. It's not Coopertition. Strong alliances are denied coopertition bonuses and higher QPs by 3v0 strategies, and I think it is not in the spirit of FIRST to deliberately lose by refusing to play.
2. It's counter to the spirit and mission of FIRST. 3v0 games are *boring* for the players and for the audience, and *boring* techno-geek-stuff is exactly what For Inspiration and Recognition of Science and Technology was designed to _not_ be. It is neither inspiring, nor is it a good demonstration of science or technology.
3. It's counter to the goal of making the game more audience-accessible. 'Nuff said there.
4. It might be harmful to your team in terms of sponsor and media support. I cannot imagine that such decisions sit well with sponsors, who want video, still shots, and articles written about how successful the teams that they have poured time, money, and energy into are doing. Newspapers want win/loss ratios and whiz-bang robo-tricks, not strategies that involve doing nothing.

On 6v0:
1. 6v0, fully-cooperative-and-agreed-upon-beforehand play may well be in the spirit of first, but it has several negative impacts.
A. The long-term benefits are low. Used excessively, all it does is maintain the status quo. Compared to winning medium-scoring 3v3 games, it is not as good at generating QPs.
B. It benefits a pre-determined alliance more. The alliance chosen to be the 'loser' will likely get more points than the 'winner', due to inadvertent penalties.
C.i. It hurts your chances of getting chosen if you do not get into the top 8. 6v0 play removes opportunities for teams to show off their robot -- to the audience, to the fans, to their sponsors, and to the scouts of the other teams (who are looking out for which robots will *win*.)
C.ii. It hurts young teams more. I think this is especially damaging to rookie and second-year teams, who may go along with the strategy due to peer pressure from more established and successful teams.
D. It confuses and annoys the audience. (See above).

It's bad strategy for teams, and it's bad strategy for FIRST.

Patrick

Re: A novel strategy: Always score for your opponents
 

I will disagree that 3v0 games are necessarily boring. They can provide the opportunity for robots to showcase inspiring potential offensive capabilities. I'll point to '07 GTR Q15 (it was a 3v1 match, but the alliance with 1 wasn't playing defence)

Re: A novel strategy: Always score for your opponents
 

Don't be fooled. 6v0 is not a viable strategy. Below is the competition record of team 1515.

The first number is the match number followed by the red teams then the blue teams, next is the match score, red, blue. Then the winning alliance followed by the match seed points and coopretition points. The totals are last.
My total is off by one point, but I do not have access to unpenalized scores so there must have been a penalty

I do not know that 1515 actively pursued 6v0 but 4 of their matches are the desired result of this strategy, 3 of those would be considered very successful.

Their final seed was 55 of 61. Their highest seed gain was match 79 which was a well played match NOT a 6v0 game.

Looking at the results of our regional all of the top seeded teams had a boat load of coopretition points. A successful 6v0 match results in NO coopretion points. Looks to me like this is the fatal flaw. If you want seeding points play hard, with just enough defence to maintain a narrow lead.


3 2865 3188 2898 948 8 1515 0 5 blue 5 0
13 2374 8 956 1778 2951 1515 1 0 red 1 0
23 1425 1515 2471 1571 2411 2002 1 1 blue 1 2
31 753 2922 2521 3024 1515 368 6 3 red 6 0
46 2915 948 1318 1700 1515 3070 2 1 red 2 0
57 3131 1823 1515 2130 3192 2192 0 5 blue 5 0
62 1595 1515 2046 3024 3223 1540 0 7 blue 7 0
79 2517 3210 3131 3145 2522 1515 2 4 blue 4 4
86 1571 957 1515 2635 3213 2147 0 3 blue 3 0
34 6 40


BTW concrats to team 1515, in spite of their bad seed 1515 was selected as an ally of the #1 seed and earned a trip to Atlanta.

Why is seed so important?

Re: A novel strategy: Always score for your opponents
 

If you don't make it to the top 8 seed does not matter at all. Seed matters to try and get into the top 8 to both guarantee you will make elims and give you the control over assembling your alliance.

Re: A novel strategy: Always score for your opponents
 

Clearly, and the best way to get into the top 10 is to Drive Well, Play Well , Fix Well.

This misbegotten strategy is simply not the answer.

Re: A novel strategy: Always score for your opponents
 

I don't think many are arguing that this is a strategy that one should always play with. I agree that in general, a team will not score as much as if they had won a competitive game, but there are situations where it may be beneficial. The 6v0 strategy (at least amongst two alliances aiming to seed high) reduces risk of a low score and thus may be helpful if teams in the match are trying to protect leads in QP.

One must also be aware of in match circumstances and they may suggest "switching sides". For example, if your alliance finds itself down by an insurmountable number of points and your alliance is interested in receiving QPs (I realize that may not be the case), the alliance should begin to score on itself. If you feel it is insulting to score on your own goals, realize since the '00 season, FIRST has always had an element to make it more advantageous to score for your opponent. If you still have a mental block, it may help to think of the goals as "near goals" and "far goals" instead of "our goals" and "opponent's goals" and of the QPs you would receive as your score.

Re: A novel strategy: Always score for your opponents
 

I'm led to believe that the very clear stance that many people have against the adaptive manner in which teams are playing this game may have led to them to unfairly making assertions about the nature of said play.

I personally don't believe that the initial suggestion made in this thread of "always" scoring for your opponent is the way to play it, but there are plenty of situations in which it makes perfectly good sense to do so.

What is "Coopertition," really? I can see it's place in the confines of a regional and even throughout build season. The notion of helping those that we compete against right up until we go on to the field is clearly what separates FIRST from so many other competitions and programs that run with similar goals and motivations.
But in terms of on the field play, what exactly does it mean? To me, it seems like a wholly artificial concept that really has no concrete meaning in terms of match play; it's really just a term given given to a scoring system, isn't it?
I was particularly puzzled that an award was handed out to the team that "scored" the most Coopertition points. If the game is played in the organic manner in which many people would like to see it played (you score for yourself, not against yourself), the winning team should never have any impact whatsoever on the Coopertition bonus. Isn't it funny to award a team for something they never played a part in? Perhaps FIRST really did intend for teams to play the bonus manipulation game.
Denying other teams bonus points while attempting to score them for yourself is how teams climb up the standings this year. If you're the "stronger alliance," find ways to adapt to the strategy being played against you, don't call the strategy unfair, it's merely a product of the game.

I don't necessarily agree that the manner in which 3 vs. 0 matches are played is boring - but I'll leave that to the eye of the beholder. I do, however, think that the responsibility of making a game interesting lies not in the hands of teams, but instead in the hands of the GDC. Do teams go out on to the field with the intent of entertaining the audience or setting out to accomplish what they designed and built their robot to do? These actions are, again, merely a product of the game and its rules.

I think this goes back to the idea of "always." If you're always playing this game, you're right in saying that your sponsors and supporters will never be able to see the results of their inputs.
However, is a team's ultimate success ever really defined by win-loss record - even in years past? I would think newspapers, other media outlets, and sponsors (those not completely attuned to all the nuances of FIRST games) would most want to hear about the ultimate results of the team's exploits not. Being able to tell them that you won the regional because you seeded first certainly sounds a lot better than saying that you finished 5-3-1 and made it to the quarter finals.

Agreed. Don't do it all the time, pick the right moment when it to everyone's advantage and carry it out then.

The fact that penalties may derail a strategy is really not dependent on the strategy being implemented at all. Penalties hurt you in any scenario - read the rules carefully, build your robot appropriately, and don't take penalties.

I'm not sure how playing a match in which everyone is scoring on one team is going to prevent you from showcasing your robot's capabilities.
Winning in the short-term is very different from winning overall, and I think that's where one has to be careful in evaluating this strategy. A couple of well thought out implementations of this strategy may be short-term "losses" (what does a loss mean this year?), but wins in the long-term. The point about scouts is moot - they shouldn't really be looking for wins but rather a robot's performance.

Perhaps this is true, but it is no different than any other strategy alliances may choose to pursue - it's not really a flaw with the strategy.

Again, a product of the game.

Patrick, I hope you don't take this as a personal affront. I think you bring up some valid points to consider in the implementation of this strategy. You counter the assertion at the beginning of this post that teams should always score on their opponents very well.

However, I think you, like many other people who have taken a bit of a stand against this approach to the game, have unfairly written it off. Some teams will choose not to implement the strategy (we ran into this a couple of times at FLR), and of course no one team has the right to question the values and beliefs that another has. That said, it is equally unfair for a team who chooses not to implement the strategy to call out teams that are choosing to put it into action.

Whether it is effective or not remains to be seen. We're but a week into competitions. Let's see where the next few weeks takes us and then determine whether it is truly an effective strategy. For the moment, my money is on yes.

Re: A novel strategy: Always score for your opponents
 

Here's a way of thinking of the game that is not nearly as analytical as the previous post, but perhaps simple enough to actually think about in the heat of a match. It assumes you have some offense, and the confidence to play to win.

Don't think of the opponent's goals as opponent's goals, think of them as your far or bonus goals, and think of your own as your near goals. If you can win, all balls scored in all four of your goals contribute to your score. Balls in your far goals count double, and your opponent will help you with those! You simply want a few more balls to go into your near goals than your far ones. Play to win. Score as much as you can in your near goals, and don't impede scoring in the far goals. If you can build a comfortable lead (you'll have to decide what that is), take any opportunity you have left to score in your far goals rather than your near ones, because they count double. If your opponent is playing a strategy of scoring in your goals, fine, take the goals. This only gets you to the tipping point sooner where you can begin scoring in your far goals. As the winner, you will come out ahead. (Hmmmm......could it be that this is the type of play that was intended?)

If you are absolutely certain that your alliance has no chance of winning, then you should score everything into your far goals, because they are the only ones that will contribute to your score.

Re: A novel strategy: Always score for your opponents
 

As I read this thread, I’m surprised of how many people think that the 6vs0 strategy is not with the ideas of FIRST. Perhaps, this game was not intended to be 6vs game, but I feel that that scenario actually is quite FIRST like. Every year, we are pressed to further promote gracious professionalism. What is more professional than working with your alliance and your opponents to make the most of a 6vs0 game? This would benefit a lot of people allowing rookies to work along side with veteran teams, as well as good teams working with other good teams. If the 6vs0 game was pushed to its potential, it would utilize every robot in a game that makes them all work together.

I understand that there are problems with a 6vs0 game, but believe that they can be solved. One issue a senior (taking economics) mentioned was that in a event of mutual rewards, cheaters of the system would always emerge. Also, the issue of showing off the robot to scouts is potential problem for many teams.

Perhaps I am just out there dreaming, and I understand that this is supposed to be a competitive sport, but wouldn’t it be interesting if the games instead where to see how well 6 robots could work together? In the finals, alliances would be chosen by teams that can help the most and in the right positions. Then, the matches would be which 3 robots can ultimately work the best together, which team can fully utilize their resources.

I am from a 3rd year team with a robot that isn’t one that is guaranteed to dominate and that might be why I sort of like this strategy, but really, isn’t FIRST more than just win or lose…and more than just thinking about that one more bonus score?

Re: A novel strategy: Always score for your opponents
 

But if your opponent starts to score on you in response, then they will get 2 bonus points for every goal into your goal plus the already high score you gave them. This might work for a while but i think teams will quickly figure out how to beat this strategy.