"Collusion" and De-Scoring

I have read a great deal of posts about “collusion” and de-scoring but I would like to open up a more general discussion.

Here are my thoughts:

  1. The game this year is highly score-dependent. I do not necessarily agree that teams should decide on a strategy with their opponents before a match. OFF-FIELD PRE-MATCH collusion is a shady practice. However, if both teams use good ON-FIELD strategy, the audience should think they had colluded because they would act in the best interests of each other. Thus, pre-match collusion should not even be needed to score high in qualification rounds becuase good, high-scoring strategy should be assumed.

My description of Qualification strategy follows:
If you and your opponent both have stacks of the same height, I do not see any reason to ever knock down either stack, as both teams will benefit equally from the multiplier.

To play this game with a limited scope (score many points for yourself and de-score your opponent) does not seem logical. It is my belief that the reason FIRST created a scoring system that factors in the score of the loser was to make matches more exciting, as both teams try to get the scores as close as possible. This type of on-field “collusion” should be used in every match, and both alliances should have an unspoken agreement to score as high as possible.

Again, this game is score-dependent, so I think whatever a team has to do to increase its score is not only within the rules and vision of FIRST, but is expected and embraced as sound strategy.

  1. On that note, I would like to discuss de-scoring in the elimination matches. First, the team that wins the first round of an elimination match is at a great advantage, because both teams are generally trying to score high, so the gap between alliance EPs is great. From what I have seen, it is often hard to make up EPs in the second match of an elimination round.

Did FIRST envision an elimination match strategy that involved destroying one’s own score to keep the other alliance from making up the EP difference? I don’t know. Does the strategy work? Absolutely.

Score is the only thing that matters in the eliminations. Again, I believe that whatever a team must do to keep their EPs higher than the other alliances’ is good strategy.

If a team is losing the first elimination match and there is no way to salvage it, there is no reason to go to the ramp. A losing robot on the HDPE increases the EP gap by 25 points, and makes it much, much harder to win the elimination series.

If Alliance A wins elimination match 1, then any boxes A leaves in its scoring zone at the end of the second match only brings the losing team closer to victory.

In the eliminations, winning is winning. In most cases (barring a few examples such as a DQ), any strategy that wins is a good strategy.

Elimination strategy is completely separate from qualifier strategy. De-scoring in a qualifier hurts both alliances, but de-scoring in an elimination after you have won your first match will help you win the series.


If you have other opinions, please post them here. I am sincerely interested in the ideas of others on this subject.

Jeff Alpert
Team 469

*Originally posted by JAlpert *
**If you and your opponent both have stacks of the same height, I do not see any reason to ever knock down either stack, as both teams will benefit equally from the multiplier.
Jeff Alpert
Team 469
**

How about if both teams have about half the bins, the other team is on top of the ramp and your team’s robots can’t make it up. Let’s say they have a 4 stack and you have a 4 stack. They are going to get 50 for 2 robots, and about 80 for bins. You will get 80 for bins. They wind up with 290 QP and you get 80 QP. Now if you knock down their stack, then they score 50 + 20 and you score 80. Your QP just went up to 220.

Lastly, I think this scoring system is ghastly. (Sorry about that rhyme.) My main complaint is that the scoring “celebrates” (rewards) collusion, thus encouraging teams to do it in their effort to win. Also in other sports, there is a defensive strategy as well as offensive. With this scoring, defense (reducing the other team’s score) is often non existent.

As to having to roll off the ramp to win, let’s do away with that for next year, please.

Why?

Apart from anything else, we would like to grow bigger, get more sponsors, and to do that we need to make the game audience friendly. I envision us on TV with lots of money for new teams, etc. However right now, the average person watching a match on TV would be totally confused by the scoring, when they see a team win 80 to 2 and then lose 40 to 50, but the other team moves on.

(The other thing that needs to happen in my opinion, is we need to get the game more focused. We have 4 robots all working away in different areas and it is a total nightmare to video and for an audience to watch on a screen. We have trouble making good videos to show sponsors. But in basketball, they have 10 people and they don’t have any trouble filming it, because there is only one ball, and they follow it.

Hey, lets try a game with a nice big ball, some goals AND hoops, learn to pass from robot to robot, throw the ball through the hoop for extra points or shoot along the ground for 1. We can even have a ramp in the middle with the ball on top for the autonomous period. If a robot can get the ball and shoot a hoop autonomously, that would be wild.)

I’m sorry if I was unclear about the qualification round strategy.

If both alliances have stacks of the same height, and one of those alliances cannot take the ramp, it is still in their best interest to leave the opponent’s stack up. If they knock down the opponent’s stack, the opponent will simply counter this move by knocking the final stack to the ground. So, the match will be back to the same place it was when the two stacks were up, except now the point totals are lower, and everyone recieves far lower QPs. The winning alliance gets 50 + 20 = 70, and the loser gets 20. Even if the winning alliance finishes with only one robot on the HDPE, they still win with 45. All you have done by knocking their stack down is lower both alliance QPs.

A team should only knock down a stack of equal height if the other team knocks thiers down first. If both teams play with everybody’s best interests in mind, no equal stack should ever be knocked down. Stack knocking is a last resort counter-attack to poor strategy.

Sorry for the confusion. I hope this clears up my earlier point.

Jeff Alpert
Team 469

JAlpert-

Thanks for finally stating what I’ve been thinking through all these “fixing” threads. If your team goal was to seed high, you would want to leave your opponent stacks alone if all possible in order to win with a greater number of points. Obviously, if you’re gonna lose, that would probably be the easiest way to get the win back. I just think of watching matches where there are still bins unclaimed at the top of the ramp yet within the first 30 seconds of the match, opponents stacks (or all stacks on the field) are knocked down… how would you know it is even worthwhile to do that yet? The smart strategy (in my opinion) would involve leaving all the stacks alone. Even if you do lose, if your stacks still stand you will have a much higher QP than if they are not. Everyone does better.

As for FIRST promoting de-scoring in the Elims. It’s funny. In NH we saw only a few times use that as a valid strategy in order to leave themselves at a better advantage for the second match. The smart thing if you win the first match is to often try and score zero in the second. If you are losing the first, you want to get rid of a reasonable number of points to lessen the gap between you and your opponent. You rarely saw a team that lost the first match come back to win and move on with the second (maybe never)

However, I went to watch UTC and before the elim’s starting, Woodie stepped up to the mike and explained this strategy to everyone… that ‘if you are losing the first match, it might be more beneficial to NOT get on the top of the ramp to keep from giving your opponents an extra 25 points in their cushion’

In UTC, I saw several first match losing teams win in the end. Maybe it was just a coincedence… or maybe it was after the advice of FIRST.

Interesting. We’ll see what this week’s regionals bring.

*Originally posted by colleen-t190 *
** it might be more beneficial to NOT get on the top of the ramp to keep from giving your opponents an extra 25 points in their cushion’

**

You actually would give them a 50pt advantage. (2x losers score (50))

Cory

boy what a slippery slope this is

if you stay off the ramp at the end, to keep your opponent from getting 2X your 25 points

how do you know from where you stand that your opponent really IS GOING TO WIN?!

they might be touching the ramp on the other side where you cant see it

or after control is taken away, a part of their bot might droop down, touching the top

or they might be DISQUALIFIED for something you didnt see during the match, and then you win and get TWICE YOUR OWN SCORE.

There is only one way to play this game.

PLAY TO WIN! PLAY TO WIN! PLAY TO WIN! PLAY TO WIN!

if your Bot is able to win, if your TEAM is able to win, then the team that wins MOST OF THE TIME will win the finals.

Get out there, give it all you got and GO FOR IT!

Ken Wittlief - sparkie engineer for team 578 “BLUE LIGHTNING” - Winner of the Canadian Regional SPORTSMANSHIP award! <<<HINT HINT HINT and 16th in the rankings <<< HINT HINT HINT

also first choice by the 5th ranked team for the playoffs

IT PAYS TO BE GOOD SPORTS AND TO ALWAYS PLAY TO THE BEST OF YOUR ABILITY - LET THE TOTES FALL WHERE THEY MAY!

You give the winners 50, but you get 25 for your team so the spread is just 25.

I agree with both Alperts. The descoring strategy in eliminations is the great equalizer. Let’s say your partner is shut off because they shot outside the field during autonomous period in your first match. What do you do? In elimination rounds 1 good team will, more than likely, not beat 2 good teams. You descore, descore, descore!! My rule of thumb is keep the game within 50 or so points and you can come back in the second match. There are many matches that the battle on top at the end determines the winner, but there are also many matches that are a slaughter within the first minute. For the latter, it is best to lower the margin so you may live to fight another day.

-Paul

*Originally posted by JAlpert *
**If both alliances have stacks of the same height, and one of those alliances cannot take the ramp, it is still in their best interest to leave the opponent’s stack up. If they knock down the opponent’s stack, the opponent will simply counter this move by knocking the final stack to the ground.

A team should only knock down a stack of equal height if the other team knocks theirs down first. If both teams play with everybody’s best interests in mind, no equal stack should ever be knocked down. Stack knocking is a last resort counter-attack to poor strategy.

Jeff Alpert
Team 469 **

I think that is true most of the time. However consider these scenarios:

Scenario #1:Both teams have 4 stacks and about 15 bins. 3 robots are on top but your robot can’t get up, so you go under the bar and come around the back to go up the ramp. You still can’t make it up the ramp, so you keep pushing until the last few seconds, and then you suddenly turn and clobber their stack, leaving them without enough time to nail your stack, and your team is now the winner. You go from having about 85 QP points to their 280 QP, to you getting 215 QP to their 65 QP.

And if they come off the ramp to stop you when you first come around the back, you have equalized the situation just by threatening their stack.

Scenario #2: All the robots are on the ramp but the other alliance is ahead in bins. You can wait until the last few seconds and then wip off the ramp and nail their stack to win.

This game has offence and defense (stopping the other team from scoring) and from my viewpoint, defense still has relevance.

(Anyway, I still want to get rid of this scoring system, so we don’t have to leave opponents stacks up to boost our score. I picture my aunt watching, and she just wouldn’t get it.)

I absolutely agree that in these situations, knocking down a stack can be beneficial.

However, these situations are also very risky. The odds of a robot being able to quickly and easily knock down a stack without leaving the other alliance time to respond is slim. The risk of a terribly low QP score is too great.

Your scenarios are great examples of when this strategy would break down, but I strongly agree with your point that a strategy that mantains the stacks is a good idea in the majority of qualifiers.

Jeff Alpert
Team 469

Doug–

I think a sport with a ball would be rather boring to watch; if you’d like that, you could go play 'bot soccer. While I agree that FIRST’s game is rather confusing as far as scoring this year, I certainly think that it is a blast to watch. Yes, it is true that you cannot keep track of all the robots at any given moment, but focusing on the robots that are doing the most action-oriented activities gives plenty for people to watch. A game with simpler scoring rules would make for a much more viewer-friendly competition. Also, to make the game more audience-friendly, the kickoff animation or something akin to it could be shown to those watching the competition, and the details could be left for the announcer to cover as they arise during matches.

Consider football: scoring and rules are not clear (just survey a bunch of girls in your classes), yet if it holds someone’s attention for just a quarter or two, they begin to understand the rules. Same idea here. As matches go on, more and more details are explained to the audience by the announcer. The basics, however, should be explained prior to the first match to clarify the game, as it changes year to year.

As far as descoring and collusions, I agree with JAlpert. The concept of descoring and on-field collusions is a valid strategy and completely fair; HOWEVER, I find off-field collusions made prior to the match not in the spirit of FIRST.

Here’s a novel idea. Instead of a game centered around balls, bins, donuts or floppies, we should just have and old fashioned drag race. FIRST should switch over to gas motors, Big Chevys of course, and let us duke it out. The guy to the finish line first, wins!! No silly multipliers, zones or x times the losers score to qualify! You cross the stripe first, you win. Easy for the spectators, easy for the players. Scoring is always so complicated in FIRST because there are so many ways to win and lose. As far as Descoring, it doesn’t make much sense to remove points during the qualifying matches. During that time you want as many points as you can get. Unfortunately this year, the bins have no safe zone. In 2000, once you put the balls in the trough, they were relativly safe. It was easier to keep track of what your score was and what you had to do to beat the other alliance. Once you start plaYing elimination, who wants a big score? You only need to win by a few points. Go in, keep everybody from scoring, play Battlebots then move on to the next round. Seems years past required a little more finesse and strategy to win than all out brute force. Hopefully next year, FIRST will be open to a game that as Jim Zondag from 33 says has 1 upmonship. You can score x way relativly easily, but you get more points if you can top that with something which requires more work. That would also promote more challenging robot designs, and still give the rookies a chance to compete. Drag racing would be a blast though…Spring is in the air, time to bring the Camaro out, put robots away and listen to the scream of 700 horses tearing up the ashphalt!! And win or lose based on my own merit… :cool:

At the SV Regional, 254 (cheesy poofs) used the minimize your own points strategy twice in the finals, i think. Once in the first round of the two (getting a perfect 0 points) and again in the very last round of the day, when they couldn’t get onto the ramp, they went into their own zone and cleared out all but 7 bins, and winning the regional. Very neat to watch :smiley:

*Originally posted by Vincent Chan *
**Doug–

I think a sport with a ball would be rather boring to watch; if you’d like that, you could go play 'bot soccer.
**

You may be right. I have never seen 'bot soccer, and the last thing I want is to make the FIRST Robotics Competition boring to watch. I want it to be really exciting. I will check out 'bot soccer when I get a chance so that I have a better idea of what you are saying.

Basically I have been trying to get a handle on a problem that has been discussed by Dean at the kickoffs and in threads on this board from way back: how to make the game suitable for a TV audience. (Some of those threads pre-date me on this board; I ran into one when doing a search.) Sure I don’t have the whole solution yet, but whenever I turn on the TV, there is always a ball zooming around with the camera following it, and apparently, people watch and there is lots of money for the franchises, etc. Maybe, just maybe, we could take a lesson from them and figure out a way to make a game with one ball, that is very exciting. I am far from claiming a complete solution, just an observation that all the really popular team sports on TV seem to involve one ball, sometimes a very small one as in baseball. The video person knows which way to point his camera. The audience knows where to look. It’s simple in that regard. The complexity and interest is built on a simplicity. Maybe we could do that too.

Why isn’t basketball boring? What is it about soccer and football that draws millions of people? Can we isolate those elements and imbed them in a robot competition? They are doing something right, as far as getting and keeping people’s attention. Maybe part of it is the passing of the ball from player to player. It’s like the old shell game, “Where is it now?” Then of course we have tackling in football, home runs in baseball, and the slam dunk in basketball. Maybe we could come up with a robotics version and combine it all into one game.

*Originally posted by Vincent Chan *
**
While I agree that FIRST’s game is rather confusing as far as scoring this year, I certainly think that it is a blast to watch. Yes, it is true that you cannot keep track of all the robots at any given moment, but focusing on the robots that are doing the most action-oriented activities gives plenty for people to watch.
**

The scoring is definitely part of the problem. Tonight, my son who is an engineer on our team, and I were watching a SOAP108 video of Semi Final 2 of the MidWest Regional. The Baxter Bomb Squad Team 16 cleared the stack at their opponents end which made sense, but then they proceeded to clear all the bins from their own side.We were confused. Finally I noticed that we had downloaded the 2nd match of the semi-finals rather than the 1st as intended. Then we could figure out what was happening, but I am darn sure a non-competing audience would have had trouble with it. I think that we have to learn to see the games through the eyes of people who aren’t competing. Sure we get a big kick out of it, because we are deep into it.

*Originally posted by Vincent Chan *
A game with simpler scoring rules would make for a much more viewer-friendly competition. Also, to make the game more audience-friendly, the kickoff animation or something akin to it could be shown to those watching the competition, and the details could be left for the announcer to cover as they arise during matches.

I agree. Both of those things are helpful. However, it seems like FIRST has been struggling with simpler scoring for years, and has never arrived at a solution that works for audiences.

*Originally posted by Vincent Chan *
Consider football: scoring and rules are not clear (just survey a bunch of girls in your classes), yet if it holds someone’s attention for just a quarter or two, they begin to understand the rules. Same idea here. As matches go on, more and more details are explained to the audience by the announcer. The basics, however, should be explained prior to the first match to clarify the game, as it changes year to year.

Yes but in football the focus is on one thing: the ball, and even though the rules are complex, the premise is simple, “Move the ball to the other teams end.” You can see that the players are trying to do that, even if no one has explained it to you. On a TV screen, the camera follows the ball threw the air or it follows the guy running with it, and always in the same direction.

*Originally posted by Vincent Chan *
As far as descoring and collusions, I agree with JAlpert. The concept of descoring and on-field collusions is a valid strategy and completely fair; HOWEVER, I find off-field collusions made prior to the match not in the spirit of FIRST.

I agree, given this year’s rules. I just want to get them changed, so when people are watching they don’t see odd ball things, like one team making a stack for their opponents. That might not seem to strange to us–we’ve been living and breathing the game for months, and most of us were involved in a somewhat similar game last year, but I am sure it is to most non-competitors.

On the subject of off-field collusions, Dean likes to bring up that a society gets what it “celebrates”. Well we got “off-field” collusion this year because the game “celebrated” it (ie rewarded it) through the rules. Solution: let’s not go there in future. Let’s just not bother lighting that particular brush fire and then spending weeks discussing how to get rid of it. Just change the scoring, so that collusion wouldn’t be helpful to a team. Perhaps a team’s qualifying points should be based on the gap in the score. Then you could increase the gap by knocking down the other team’s stack or by building one of your own: defense and offense both reward the team. And the losing team would be trying to narrow the gap. How would it work exactly? I don’t know, but I think it is worth discussing, so thanks for your viewpoints on the matter.

Keeping all stacks up by no touching them does not benifit anyone.

If you add more points to the system all you get is score inflation. The loser score is higher, but the winner score is proportionately higher.

Thus is makes little sense to gamble on untouched stacks when anyone can easily knock it over in the last few seconds of the game

Play your best, protect your stacks, and optimize your points when you can. (Let opponents get on top of ramp during qualifications, or let opponents stacks stand if they pose no threat)

*Originally posted by Patrick Wang *
**Keeping all stacks up by no touching them does not benefit anyone.
**

I beg to differ, in a game where QP’s and Ep’s are what is used to “advance” or “win” your strategy should be to score as many points as possible during the seeding rounds and the first elimination round. If you win the first elimination round then you use the reverse strategy to advance or win in the second match.

Two strategies for the two parts of the game.

I have something to say about collusion. It was quite rampant at the Central Florida regionals. As of them we had the top three qualifing point matches in the nation. We did not do this. We never tried to make a stack of eight. It was annoying that when we won, we got less qfs then many of the other teams did when they lossed but we were picked to go to elimination rounds because we proved that we we had a good bot not that we were good at cooperating.

We lossed in the semi-finals because we did not de-score. We were on the ramp at the end of the match and we lost. The rest of my team was cheering when the judge ruled that we had made it on the ramp and we got the extra twenty-five points. I guess this shows the confusion too because just about the whole team did not realize we would probably lose because of it. The result was that we needed to make up 94 points in the next match. Needless to say we didn’t.

When it comes down to it, the ones who do the ‘live and let live’ strategy of where all stacks are left alone, then they fall flat on their face on the elimination rounds since they were up there due to a loophole and not their bot working well.

I beg to differ joseph, if both teams let both stacks stay up, they have no outcome on the actual winner, so the better teams in that match will still win, if they are the only teams that want to optimize points by not needlessly knocking down stacks, then maybe they will be too high due to scores, but as long as i see an equal or smaller stack then mine, and i have scouted the robots and feel confident in my AI routine to get more boxes than the opposing robots, i will leave their stack untouched, until something goes wrong, Yes it is wrong to say not to knock down the stacks b4 the match, but the outcome makes sense, both teams get more points, if other teams feel the need to knock down the opposing stacks to make sure they win, then their scores will not be high and they may need to be picked as an alliance… although i dont think i would ever want to leave a stack of 8 standing, that’s just a litle high

There is one flaw in your reasoning.
If you are slightly behind and you leave your opponents stack intact, and loose as a result

you will get less quality points than if you knock it down and the last minute, and win

so you will have a HIGHER score - because you won, and you get your score PLUS twice the other teams score.

The teams that do the best overall have the ability to CHOOSE to leave the other teams stack intact, while pushing their own score up as high as possible, get up and off the ramp quickly

and have the abillity to swoop down off the ramp and take out a stack if they need to (to win) and make it back up on the ramp at T=120.

its really simple people! the more versatile your bot is, the more options you have during the match, and the better you are going to rank.