unique strategies that worked....... or didn't

[AGPapa]

A unique strategy during F3 at FLR that I thought worked well was executed by teams 1507 and 378. 1507's robot usually shot from right behind the low goal (like many robots this year). When 1507 was under heavy defense, team 378 (who's robot is about the same length as the low goal) would go up against the wall and 1507 would use them to line up instead.

Here's a video of it in action.

[BrendanB]

Originally Posted by Andrew Schreiber

Defense... See attached. That's the result of that defense on a cable that is buried deep inside our robot. Imagine that, breaking your opponents via burying a claw deep inside their bot is a winning strategy

I heard that isolating your opponents ball works great too.

[Anupam Goli]

We knew we were screwed for our last couple of quals matches at Palmetto this year, so on Friday night, I was tossing around strategies with some of the other mentors and students. We came up with this strategy that we call "the magic school bus", which was a formation where the three robots surrounded the ball, and just kept in formation while running down the field, gathering the 3 assists, and finally scoring it. We thought it would be stupid and crazy enough to work.

Crappy visualization:

R1 0 R2
... R3

Basically, inbound it so the ball lands in the middle of this robot cluster, R3 pushes the ball against the wall for the first assisst, then the formation moves the length of the field, with R2 pushing the ball, and at the final zone, R1 would pick the ball up and score.

The strategy never came to fruition, and I don't remember why, but I think everyone's minds would've been blown at the event if we unveiled it on the field.

[CTbiker105]

Originally Posted by who716

They were brutal matches to say the least but it surprised me to see this strategy worked.

Brutal is one way to describe those matches...

One of my favorite strategies I saw being used this year was at CMP in Archimedes. 33 would inbound, truss it to their first HP who would inbound it to 1671. 1671 would shoot it horizontally across the field to their second HP. Finally, the second HP would inbound to 1625 to finish the cycle in the high goal.

Very cool to watch.

http://www.thebluealliance.com/match/2014arc_qm109

[Cam877]

Originally Posted by CTbiker105

Brutal is one way to describe those matches...

One of my favorite strategies I saw being used this year was at CMP in Archimedes. 33 would inbound, truss it to their first HP who would inbound it to 1671. 1671 would shoot it horizontally across the field to their second HP. Finally, the second HP would inbound to 1625 to finish the cycle in the high goal.

Very cool to watch.

http://www.thebluealliance.com/match/2014arc_qm109

This was actually done in eliminations in the Curie division, with 573 inbounding and trussing to human player, and then either 1718 or 2451 taking it and shooting across the field to the robot left over. Great strategy, they ended up finalists only losing to the poofs' alliance. I also like the give back to human player strategy utilized by 2590, 1477, and 1625 in Archimedes elims and Einstein.

[Pault]

I guess now that the season is over I may as well reveal the strategy that we wanted to use, but didn't. FYI, we knew it would have taken a really good shooter to execute, and would only work when partnered with a special few robots (probably 2-4 at any given competition). We were going to make sure that we could play the game the "normal" way before we could play it this way.

The prerequisite is a partner robot that can catch and can shoot into either the high or low goal from directly in front of the low goal (obviously high goal is better). They would stay in that position for the entire match. Your robot, which is a tall shooter, would park in the low goal corner on the opposite end of the field, get fed a ball from the HP, pass it full court to your partner robot, and they score it. The 3rd robot would play defense.

10pts truss
10pts catch
10pts 2 assists
1/10pts score
= 31/40pts per cycle

Now, the full court pass may sound crazy, but we were getting well more than 54ft of distance with our flywheel shooter during our robot's early stages (later changes unintentionally decreased that distance dramatically). The real question is whether you can get the accuracy, but I think that a good team could have tuned it enough to be reasonable (I would say you need 66% accuracy for this strategy to be really effective). Game breaking strategy, we just didn't have the capacity to execute it.

[Caleb Sykes]

Low goal extension

[Boe]

Originally Posted by inkling16

Low goal extension

Definitely my favorite match from state

[brennonbrimhall]

Take a look at how alliance #8 beat #1 in Archimedes Quarters. They almost put up 400 points without fouls.

[who716]

Originally Posted by Andrew Schreiber

Defense... See attached. That's the result of that defense on a cable that is buried deep inside our robot. Imagine that, breaking your opponents via burying a claw deep inside their bot is a winning strategy

we would never deploy a strategy to intentionally damage another robot. playing defense always increases the chance of damage, and that was just a victim of the defense level. it's unfortunate that we won the match with you goes out, but we have all had issues like this.(glad that you made it back out in time for match 3-3) In our semi finals 2-3 we got hit which reset the CRIo and we take preventative measures against this and since I have been on the team it has never happened until that one match. making the loss deficient a lot larger then what it might have been.

Originally Posted by Chinmay

Back on subject, I think that one of the most effective things I saw was the "pass back" to a human player but bouncing the ball off a robot, or throwing it into a robot and having the robot spit it right back out.

we deployed this in the 16th finals and although it only actually happened once it worked great for a real quick three assist cycle. the strategy was to only do it if both robots were available and the opponent was inbounding.

Originally Posted by Cam877

This was actually done in eliminations in the Curie division, with 573 inbounding and trussing to human player, and then either 1718 or 2451 taking it and shooting across the field to the robot left over. Great strategy, they ended up finalists only losing to the poofs' alliance. I also like the give back to human player strategy utilized by 2590, 1477, and 1625 in Archimedes elims and Einstein.

I find this strategy interesting and I really want to try it, unfortunately our robot does not have the distance on its shot to do it but I can see it being very effective.

[CTbiker105]

Originally Posted by brennonbrimhall

Take a look at how alliance #8 beat #1 in Archimedes Quarters. They almost put up 400 points without fouls.

This was the only elimination match I missed on Archimedes prior to us getting eliminated. I came back to the stands, saw the score briefly and asked a mentor sitting beside me, "Did they really score 390 points against 2056?" His response was, "Yeah, the ball never really touched the ground. It was very impressive."

I'm surprised there weren't more bots that could preform the role 1918 did.

[Wayne TenBrink]

Originally Posted by CTbiker105

This was the only elimination match I missed on Archimedes prior to us getting eliminated. I came back to the stands, saw the score briefly and asked a mentor sitting beside me, "Did they really score 390 points against 2056?" His response was, "Yeah, the ball never really touched the ground. It was very impressive."

I'm surprised there weren't more bots that could preform the role 1918 did.

There were a few that could park in the corner, catch the ball from the HP, and score without moving (254, 469, 2771 for example - 2771 was the only one that I saw really using it). Perhaps we were best suited for it since we were tall and didn't have any parts hanging out that could be exploited to interfere with the shot. This wasn't something we specifically designed our robot to do, but we realized it was a capability early on. We never had many opportunities to try it during the season, and it never worked as well as it did with 51 and 2485 - they were the key that made it work.

It was a very interesting strategy, but it had its problems. The fact that we lost in the semi's is proof that it isn't a game lock strategy. As Jared Russell pointed out, it takes one robot out of the defense role and it requires clean execution. Our HP was the one feeding our bot, and it really took a toll on his nerves. A few missed shots into the bot cost us cycle time we coudn't afford to lose. The fact that we were parked in the corner instead of playing defense made it harder for 51 & 2485 to get the ball to the forward HP & easier for our opponents to run their cycle.

In hindsight, I think it would have been possible for us to take a more active role in the match and still be in the corner when we needed to be. We still have the MARC and the WMRI coming up. Perhaps we will get some chances to try it again!

[Chris Hibner]

Originally Posted by Wayne TenBrink

In hindsight, I think it would have been possible for us to take a more active role in the match and still be in the corner when we needed to be. We still have the MARC and the WMRI coming up. Perhaps we will get some chances to try it again!

You nailed it here. Later after being eliminated, I told Ken Patton that I blew it. Every time you guys left the corner to track down a loose ball, you always made it back to the corner - even when defended. I told Ken that I should've made that strategy adjustment in the semis. I thought about it, but I was too afraid to make the call to do it. All I saw was the 390 and I thought to myself, "if we make that change and lose, then I'll have to live with the 'why did you go away from what worked?' questions." I beat myself up that night about that.

What really kills me is that in both of the semi-final matches that we lost, the opposing alliance missed an auton shot and we had the opportunity to to try to shut them down at the start of the match, but we were too focused on getting the cycle started because "hey, we scored 390 in the first match". In retrospect, putting up a score that high in the first match was probably the worst thing to happen to us. After our first loss in the semis, I should've had us go 2 bot cycles for as long as we could defend them from clearing their missed auton ball to build a little lead before starting the 3-bot cycle. Then I thought, "but we scored 390" and chickened out. Oh well. Then again, the alliance we lost to was really great so there's a good chance we would've lost anyway. Ifs and buts, candy and nuts, and all that stuff. I guess I can't complain.

[DanBrowne]

Originally Posted by brennonbrimhall

Take a look at how alliance #8 beat #1 in Archimedes Quarters. They almost put up 400 points without fouls.

I remember watching this from the sidelines. At first I was all "2056 will crush, hands down." like everyone else. As the match progressed I was just in pure shock! when the match ended and the scores went up, my jaw dropped to the floor . I looked at one of my teammates and we couldn't believe that someone had just put up that many points right off the bat in elims.

[Carrington]

Originally Posted by Cam877

This was actually done in eliminations in the Curie division, with 573 inbounding and trussing to human player, and then either 1718 or 2451 taking it and shooting across the field to the robot left over. Great strategy, they ended up finalists only losing to the poofs' alliance. I also like the give back to human player strategy utilized by 2590, 1477, and 1625 in Archimedes elims and Einstein.

It's funny you should mention that because both strategies were actually things that 1625 and 2451 had a chance to practice in the weeks leading up to worlds. We were lucky enough to have a fun bunch Illinois teams down to our shop to practice several different strategies including the ones mentioned above. Based off our practice we had a pretty good feeling that these strategies would come into play but to what degree we weren't sure. In Archimedes the giveback strategy was utilized by many teams in the qualification rounds. 27 was rocking that strategy almost exclusively in qualification matches and were extremely successful with it(1010 assist points, 8-2 record, 3rd seed). A similar form of that strategy was also used by the Galileo alliance in the semis on Einstein.