A novel strategy: Always score for your opponents

[pfreivald]

Originally Posted by Grim Tuesday

BTW, GoW, were you the guys we played with in the match when we had 2 bots with no field connection and we scored 5 goals on our own?

We played you in Q55, and won 8-2, so unless you managed to penalize yourself three points, then no, it wasn't us!

Patrick

[BJT]

The people promoting gracious professionalism wrote these rules. I would hope that nobody would think that following the rules as written would go against gracious professionalism.

[nrhinkle]

Today and yesterday at the Oregon Regional we were also pondering this year's bizarre seeding system. We noticed at least one match in which an entire alliance appeared to have coordinated to shoot into only the opposing alliance's goals. Initially we were baffled, until we looked more closely at the rules. Then we saw the brilliance and counterintuitiveness of this system.

Unfortunately, we didn't think to do this until after all of the qualification matches were over, but teams at later regionals may find this useful. We wrote a LabVIEW VI which displays the raw score and seeding score for each alliance, based on the number of points and penalties input for each color.

The source code, a compiled EXE (requires labview installed) and a full installer (does not require labview installed) are available on our team website. Below is a screenshot. Hopefully some teams can find this useful for the subsequent weekends. It would be really awesome if someone figured out a way to build this into a custom dashboard, so that during the game players could see where to shoot to get optimal points. If anyone does anything super awesome like that, please let us know; we'd love to hear about it!


(Isn't it weird? The red alliance scored 8 points and got 2 penalties. The blue alliance scored nothing. And yet the blue alliance got more seeding points.)

Download/info on our website

[Chris is me]

What is the particular benefit of everyone scoring on the same goal versus intentional tie? All I see is only one alliance can "defect" with a lopsided match, but both can with a tie.

That's a fantastic idea, by the way. Thanks for the handy app. Now I just need an iPhone version.

[Vikesrock]

Originally Posted by Chris is me

What is the particular benefit of everyone scoring on the same goal versus intentional tie? All I see is only one alliance can "defect" with a lopsided match, but both can with a tie.

That's a fantastic idea, by the way. Thanks for the handy app. Now I just need an iPhone version.

A "defection" in a one-sided match will result in a few more QPs for the winner (only 2 per goal). A "defection" in a high scoring intentional tie (say 7-7 going to 8-7) results in a massive seeding points loss for the loser, they go from 21 seeding points down to 8. If instead they had chosen to go 14-0 for one alliance a single defection would give them 14 points and their opponent 16.

[mcf747]

Originally Posted by nrhinkle

Today and yesterday at the Oregon Regional we were also pondering this year's bizarre seeding system. We noticed at least one match in which an entire alliance appeared to have coordinated to shoot into only the opposing alliance's goals. Initially we were baffled, until we looked more closely at the rules. Then we saw the brilliance and counterintuitiveness of this system.

This was our alliance team 1515 in our last Qualifying round. Is took quite a bit of convincing to get the other teams to play along because as you said and a few people on this thread have said.....alot of team don't understand the seeding rules.

There is an inherit issue here though at a large regional with 60 or so teams not only are you boosting your seeding score it is also boosting everyone elses on the field or about 10% of the teams at the regional.

Because of this it is really only an effective tacktick for your last match because if this was done the entire time the efectivness of it would diminish.

In a sense it almost turns the game into a 6 on 0 and sees who has the best set of 6 bots on the field instead of 3.

Although the match didn't turn out as well as we expected (the other alliace couldn't really score and were having some control issues) the best part was listening to the announcer (who was a great guy) but had no clue how the seeding worked and could not figure out what we were doing.

[rant/]
This raises another important issue of the scoring. I am not going to get into it (as there are already many other threads) but it is very complicated and can be taken advantage of. I look at it both in a positive and negative way however. I consider it part of the game just like a "field element" and strategy has to be developed to deal with it. In the same light though because the "match score" and not the "seeding score" is showen on the score board there is alot of confusion and I think misleads people who do not know the rules as much as us rule worms that study them extensively. A great example of this was at the organ regional there were 2gopc reps walking around and they came to watch a match. I had spoken to them earlier so they aproched me and asked me who won. They were very surprised to find out that the team that got 0 won. If FIRST really wants the public to enjoy the game I think they need to have more strait forward scoring or at least more clearly display the seeding score which is really the only thing that matters durring the qualifiers.

[/rant]

Sorry for the long post,
Matthew Forman
Team 1515

[Chris Fultz]

Originally Posted by pfreivald

We saw this strategy and we thought that it was very much against Gracious Professionalism, and poor gamesmanship besides. We played every game to win, no matter what...
Patrick

But in this game a "win" is irrelevant and not even tracked. All that matters is your seeding points. And sometimes you get the most points by scoring in one set of goals, and sometimes you get the most points by scoring in another set.

[MikeReilly]

We certainly did not ever "collude" for the fabeled 6v0 approach. And I do think this scoring system is bizarre, but does suggest high-scoring close matches. I agree with Jeff, read the rules. I think there's way too much "Gracious" and not enough "Professionalism" (and I'm not a sore loser, we won). It was very frustrating to see a team ranked in the Top 10 that had trouble moving most of the time.

On one positive note from this strange twist, we did get some respect for our scouting team, and shared our scouting with others. There were teams in the Semis we KNEW were high penalty teams and avoided them. THAT paid off, and that's professionalism, AND we shared our info with whomever wanted it.

Play the game, not the angelic/doormat/everyone wins. We also need to teach our kids how to lose well (that was our lesson last year, and the year before, and ...)

[pfreivald]

Originally Posted by Chris Fultz

But in this game a "win" is irrelevant and not even tracked. All that matters is your seeding points.

That is absolutely false. In this game, showing off your great robot to other teams matters a great deal more than your seeding points, and how well your robot performs overall is very, very highly tracked -- by scouting teams.

The scoring system this year is really, really wonky, but if you look at the top-seeded teams at the FLR, they really were the best robots -- and that was for the most part without these kinds of shenanigans.

As an *overall* strategy, this idea does nothing but maintain the status quo -- it doesn't help you stand out, and it doesn't help your rankings when compared to the other five teams on the field. Perhaps, if you know you are going to get absolutely crushed, you might consider doing this, but in any other circumstance it just doesn't seem viable to me. We've won too many close games that we "should have lost" in the past five years to not go for the win even when we're down.

[Chris Fultz]

Let me clarify -

Seeding points is FIRST's way of ranking teams.
Scouting is our way of ranking teams.

For ranking, win loss is irrelevant, it is purely seeding points, and that is what I was referring to.

For scouting, we don't even look at win/loss or even seeding points.
We look at our data on each robot's performance - scoring ability, hanging, defense, control on the field, consistency, penalties, etc. This data helps us craft a strategy for each match, and that is also why a low "ranked" team is often higher on our draft list than a "higher" ranked team.

Strong scouting is more important this year than most and you cannot rely on the seeding points ranking system to identify the strongest teams.

[Bjenks548]

Sorry if i did this wrong, its my first post, but we effectivly used this strategy in our last qualifying match. We had two defencive robots, and one that had only part of a working drive train. We played the match and lost around 9 to 2, however because of field errors we had to play it again. While they were fixing the field, our alliance decided to score for the other alliance. We lost 12 to 0, after one penalty for the winning alliance. We got 13 seeding points and the "winning" alliance only got 12!

[martin417]

We noticed that since the matches were low scoring and close, often 1-0 or 2-2, wins didn't help much, and we moved up in the rankings every time we lost. So, we briefly considered an unusual strategy. We would have two of our robots pull into our goals and park to block them, and the third bot would do everything they could to score in the opponents' goals. Since there were almost always penalties, this strategy would give us as many or more seeding points than the winning alliance. We decided at the last minute that we would just play straight up. In retrospect, we should have gone with the original plan.

[jwfoss]

FLR - Qualification Match 63

All teams involved 6v0

250, 191, 3181 (v.) 229, 3173, 1511

Final Score of the Match 12 - 0

Teams Received:
Red: 12
Blue: 11 (Due to Penalty)

The announcer couldn't figure out what we were doing till the last 20 seconds. We (all six teams) discussed the strategy the morning before the match being that many of the teams were well connected with each other. Thanks again to all teams involved, I fully expect this strategy to be used in the weeks to come.

[Bjenks548]

While we used this same strategy, we believe that it is against the values of first to do predetermine matches. When we did this, we didn’t tell the other alliance of our plans to maximize the seeding score. But I will say I have never seen the killer bees crowd (who were on our alliance) so quite during one of their matches.

[Integral]

This is essentially the 6v0 strategy.

Did you spend 6 weeks building this robot to play the game or game the system?

To anyone engaging in this strategy I say SHAME ON YOU.

If this is proposed to your team just say no, it is simply not in the spirit of FIRST and is a betrayal to team mates, friends and family.

Please play the game, don't game the system.