Ball Hogs

[rick.oliver]

Originally Posted by Travis Hoffman

I could see the ball hog concept being more of a problem when you get 3 capable offensive robots together, especially when the teams are used to being the "point man", the center of attention. Who will swallow their pride and relegate themselves to more of a support/defensive role? Or how will they concoct a devious strategy where 2 robots manipulate one of the trackballs in glorious coordination and the other bot is free to use the 2nd trackball at will?

We've been discussing a strategy we've dubbed "Dancing with 1114" that was inspired by the Q&A they submitted concerning hurdling. The description contained in their question reveals a "devious strategy" or more accurately, an ingenious strategy whereby three capable machines share the two trackballs. Cooperating in the process of hurdling such that neither ball ever touches the floor. Oh, by the way, both balls are on the overpass at the end. I imagine that the GDC envisioned just such a game play. It should be fun to watch. Imagine if six robots were doing the dance!! That would rival the six ringer auton at IRI

[Swampdude]

Originally Posted by rick.oliver

We've been discussing a strategy we've dubbed "Dancing with 1114" that was inspired by the Q&A they submitted concerning hurdling.

huh? link?

Edit, found it, but I don't get it? how can beta touch the ball without crossing the plane?

[Jon Stratis]

just to throw in my two cents -

When it comes to picking the teams for the elimination matches, teams will be looking for two good hurdlers and one fast scorer/defensive bot (at least, this is the most likely desired arrangement). Last year, our robot had the capability to pick up rings and place them on the rack. During the beginning part of the qualification matches, we realized that our robot wasn't doing this as effectively as other robots, and we switched our strategy to become a very good defensive bot. The same situation should apply here - If we end up not being one of the faster hurdlers, then we'll switch our own personal strategy to appeal to the other portions of the game play.

The whole point of the competition isn't to be able to do just one thing - hurdling. It's being able to distinguish your robot in some way that makes it a unique and valuable part of an alliance. If thats through hurdling, then great. If it's through keep away, then great. It it's through driving twice as fast as everyone else and expertly avoiding traffic jams, then great. All of these strategies are highly useful for a team to have and to show during the qualification matches.

I would encourage everyone to avoid trying to show off one particular aspect of their robot and instead look to whats best for their alliance. I know at our competitions, i'll be encouraging our drivers and strategists to do whats best for our alliance, and i'll be doing the same for the teams we work with.

Oh, and we plan to come prepared with information like average laptime and average hurdle time to help cement our position

[NOV8R]

Our team thought of hurdling early on in the design process and nixed the idea because of this potential problem (three alliance partners competing for two balls). So we decided to design a robot to go for 24 points in hybird (one part of which is knocking down our alliances balls from the overpass) and then make it difficult for our oppossition to get their balls. The only robot design that is allowed to go after oppossition trackballs is a herder. So our design is a very robust herder with a great hybird program. We also decided that control was more valuable than speed in this role. We'll see during the first few regionals if we're right.

[Madison]

We collect scouting data for each machine that gives us a reasonable idea of how successful teams have been in their practice and qualifying matches at whatever they try to do. We'll look at the number of hurdles attempted versus those completed and number of laps completed, among other things, and generate a picture of their robot's capabilities based on that information.

This allows me to go into each strategy meeting with something better than a subject guess as to which machine is the best for the job. I typically formulate a strategy that best supports to most successful robot on the team, though rarely we'll try something a bit more tricky. If your machine isn't up to the task that you want to take it, I'll know it and I'll tell you. It may not always be the nicest thing to do, but it's usually what's best for the alliance.

There should be as little ambiguity as to a team's role during a match as possible. Before ever getting out onto the field, all of the teams on our alliance will have an expectation as to how things should play out and a handful of contingency plans on top of that. I don't like showboating and teams rarely make good decisions under pressure. If you didn't have a strategy to beat the opposing alliance before the match started, you aren't going to come up with it when there's thirty seconds left in the match. I'm all about sticking to the plan.

[Brandon Holley]

Originally Posted by rick.oliver

We've been discussing a strategy we've dubbed "Dancing with 1114" that was inspired by the Q&A they submitted concerning hurdling. The description contained in their question reveals a "devious strategy" or more accurately, an ingenious strategy whereby three capable machines share the two trackballs. Cooperating in the process of hurdling such that neither ball ever touches the floor. Oh, by the way, both balls are on the overpass at the end. I imagine that the GDC envisioned just such a game play. It should be fun to watch. Imagine if six robots were doing the dance!! That would rival the six ringer auton at IRI

can you post the link to the Q and A official answer?

[GaryVoshol]

How is negotiating with a potential ball hog any different than a "ramp hog" last year? It made sense for only one of the 3 robots to deploy the ramp; who should do it was based on each bot's relative merits.

Granted there will be multiple opportunities to ball hog in a match, vs. the end-game aspect of ramps. But the strategy negotiations between alliance teams will be similar.




Footnote: Post 1000!

[Brandon Holley]

Originally Posted by GaryVoshol

How is negotiating with a potential ball hog any different than a "ramp hog" last year? It made sense for only one of the 3 robots to deploy the ramp; who should do it was based on each bot's relative merits.

Granted there will be multiple opportunities to ball hog in a match, vs. the end-game aspect of ramps. But the strategy negotiations between alliance teams will be similar.




Footnote: Post 1000!

Because not everyone was lifting or ramping last year, i'd say maybe half at most were. So.... this year, do you think theres going to be only half the teams interacting with the ball(s)? Id say something like 95% will be interacting with them...

[GaryVoshol]

Originally Posted by Brandon Holley

Because not everyone was lifting or ramping last year, i'd say maybe half at most were. So.... this year, do you think theres going to be only half the teams interacting with the ball(s)? Id say something like 95% will be interacting with them...

We're talking a difference of scale here, not substance. There will likely be more opportunities to negotiate, I admit that. But the negotiations themselves won't be much different than they were last year.

[Swampdude]

Originally Posted by GaryVoshol

How is negotiating with a potential ball hog any different than a "ramp hog" last year? It made sense for only one of the 3 robots to deploy the ramp; who should do it was based on each bot's relative merits.

Granted there will be multiple opportunities to ball hog in a match, vs. the end-game aspect of ramps. But the strategy negotiations between alliance teams will be similar.

Negotiations are one thing, but I think this game plays differently in that there are constant decisions throughout the entire match whether or not to pass on multiple scoring opportunities in lieu of your partners.

[GaryVoshol]

Originally Posted by Swampdude

Negotiations are one thing, but I think this game plays differently in that there are constant decisions throughout the entire match whether or not to pass on multiple scoring opportunities in lieu of your partners.

Fair enough, I agree. I thought the point of your original message was about an over-enthusiastic team member or coach who dominated the negotiations. We've always had to deal with that.

[Bob Steele]

Originally Posted by rick.oliver

We've been discussing a strategy we've dubbed "Dancing with 1114" that was inspired by the Q&A they submitted concerning hurdling. The description contained in their question reveals a "devious strategy" or more accurately, an ingenious strategy whereby three capable machines share the two trackballs. Cooperating in the process of hurdling such that neither ball ever touches the floor. Oh, by the way, both balls are on the overpass at the end. I imagine that the GDC envisioned just such a game play. It should be fun to watch. Imagine if six robots were doing the dance!! That would rival the six ringer auton at IRI

Hey Go Thunderhawks!!! I know you guys from being coach of the RoboDevils a few years back and from nationals last year.

We are designing for just such an eventuality... At nationals we hope to be part of a triumvarate that acquires the trackballs at the beginning of the game and passes them to each other over the overpass so that the trackball never hits the ground. To that end, we should be able to "catch" the ball another team puts across the overpass without crossing the line. One ball would be in transit the entire time while 2 robots are passing..

This could be very interesting to watch...and a real teamwork effort for an alliance and it is EXACTLY what (in my humble opinion) the GDC was thinking about when Woodie talked about being on the other side of complexity..

Of course I may be completely wrong.
I hope we can play with you guys at Nationals...
We will be there!!!

Good luck to everyone!!!

Bob Steele

[JesseK]

We initially designed our bot to BE the ball hog slash galmour bot. It has a great way to capture an unpredictable ball. It has a great way to place the ball on the end, and a fast enough way to hurdle. It has a way to knock only our own alliance's ball off the overpass in the beginning. No guts no glory right?

As it turns out, there are bots that are able to hurdle much better than us. We intend to support our alliance using any means necessary, including giving up the ball to a launcher. We all benefit when we use our most efficient methods to score, and I think everyone echos that sentiment for eliminations.

However, the question of whether or not another hurdler is "better" than us at hurdling will inevitably come up in seeding matches, especially in the early matches since we've never seen the other bots play. This will be a judgement call, on the spot. Even to the detriment of our own team, we'll probably give up the ball in order to give another bot a chance while we simply play support. We do, after all, have ways to score that a launcher does not: reliable placement at the end. Knowing the competitiveness of the later matches though, this may not always be the case.