Your Team Sucks At Defense: Here's Why

This guy defends!

2 Likes

From your use of the present tense, I presume you mean #7, the mecanum beast. It’s more about teaching than learning. I am no longer with a team, so I am trying to lead some workshops/seminars about getting the most from the KoP drivetrains with extreme versions, while using even fewer tools than I had with 3946. I also provide a number of examples which are closer to mainstream, but still well outside the AndyMark manuals. I am trying to (and currently expect to) build this with no power tools apart from a drill press and a portable band saw, though if I have to use a handheld drill, I will. (Handheld drills in unskilled hands are a recipe for wrongness.) So far I don’t see any need for a hand drill. I plan to include a full BoM and details of construction more for instruction than as a blueprint. I also plan to provide details which will provide the clues as to what the problems were and how I solved them and why the numbers are exactly what I ended up with.

I also present a number of construction hacks at each seminar. One of the hacks this year will be cutting 6" cleanly and consistently off each side of the KoP endplates with a handheld bandsaw, a handful of 10-32 bolts and nuts from the KoP, and a workbench with a vise. OK, add a 5/32 allen key and a 3/8" wrench. Another will be how to use a TB cluster gear as the best tool for installing the retaining clip on a CIM being used in a Toughbox series gearbox.

I mean you can’t suck at defense if you have never done it :joy:

1 Like

A lack of something is a vacuum, which…sucks.

2 Likes

Brilliant :joy:

Grade the defense and tell me how to do it better! Team 4926’s first event playing D. I think we did ok!

Indiana District Champs F-1

One plus i see for strafers- May get out of a T-bone easier than WCD. When i got T’d by 195, i couldn’t move at all, until they let me go. If you could strafe fast enough, to overcome the T-bone speed, you might be able to get out of it (of course, depending on where you are).

This is a good reason to ask other nearby teams to scrimmage for even a few hours. It also serves as a shakedown for your robot so you have some idea what may break.

With respect to defense, I think that 2014 and 2017 were better games from a design perspective because it allowed for robots to play both offense AND defense and you could play defense on the entire length of the field.

2016 and 2018 limited defensive opportunities to zones. 2016 was far defense and 2018 was near defense. The fear of getting pushed into the null zone was a deterrent for far defense in 2018, but it still could be played.
2019 limited defensive opportunities in three ways. Only one robot on defense. The defensive robot had to stay within the frame and the defensive bot typically had to be the worst offensive bot of an alliance. It was extremely rare for a robot to switch from offense to defense, usually it involved some kind of malfunction of an offensive mechanism.

iirc 2017 had a penalty for touching opponent robot in their feeder station.

I’ve shown videos of 5012’s OCR semis performance to at least a dozen drive teams throughout the season, using it to coach quals partners, our wonderful defense picks at our regionals, and even 1257’s drive team when they played defense at FMA champs. Thanks @Jeremy_Germita for giving us a crash course from the experts :slightly_smiling_face:

I was quite surprised they slipped to us (#1 alliance) too, considering they were the clear best defender in Darwin quals. They knew exactly where to go to be in the way and got there quickly. We were excited to play with them and relieved to not have to face them again in playoffs.

Thank you for all of 1114’s work with them, it made my life easier on Saturday morning of champs. They really were ready to play championship-level defense.

3 Likes

In Rebound Rumble, we could make 1-2 shots in autonomous and then couldn’t make a freaking basket (not through lack of trying during the build) – but we had an absolutely beastly octocanum drive train which made us particularly good at two things:

  1. Endgame balancing, and
  2. Utterly shutting down robots that scored from the fender.

We spent many games basically just parked at our opponent’s fender, because if we left they could score points. The only times we had to move from that position during teleop was (a) to get there or (b) if our opponents were good key shooters.

We played back-to-back games against 191 at Buckeye that year, and in the first we simple mauled them–they couldn’t score a bit, and we had to back off defending them so as to not tip them over on more than one occasion. The next game, the moment we got up against the fender their driver took his hands off the controls and literally grabbed his hair because there was nothing he could do to score any points until the endgame.

(This was also the year we intentionally shot one of 1240’s balls out of the air when they shot from the key. Fun, on purpose, and probably impossible to ever repeat.]

tl;dr version: Yup. Positional defense can sometimes involve not moving, and yet still be excellent defense.

Gotta say, i personally think that honor belongs to 250…
One thing i have to ask: what made you swap them out after SF1? Clearly it ended up being the right call but it was really unexpected

Imo, “excellent defense” should require some sort of skill and effort. Simply parking in a key, unprotected location should not merit the status of “excellent.”

For 2019, simply parking was excellent defense.

1 Like

Theres no doubt that it was effective…but is that something that should be encouraged in game design?

1 Like

I think we can all agree, that it’s a bad game dynamic, but you have to play the game, and gain every edge possible.

2 Likes

Well, if that’s what you got out of the tl;dr version, then maybe you should have read the whole thing…

The reason why parking at the fender was excellent defense was because we were preposterously good at defending the fender from anyone trying to score from that position–and to get to the point where simply being there was a deterrent, we had to demonstrate that we were so good at defense from that position that there was no point in trying to get past us. There were few robots who were as good at it as us, so defending the fender was much more than just “I’m here, so you might as well go do something else” for most defensive teams–but we were that good, so we were playing excellent defense by simply parking there and letting them not score because of our presence.

(The key was protected, and we had to be much more active when defending key shooters).

I don’t think he’s arguing that your defense wasn’t executed well, but more so that the game should not be designed in a way that parking at the fender should be very beneficial for the defender.

Yeah, I know–but that’s not what I said.

I said that we could be effective just parking at the fender because we demonstrated that we could not be powered through on defense but could power through you when you’re on offense. Lots of other teams could be powered through while trying to defend the fender, so parking there wasn’t by itself a game design flaw that was super-effective for “the” defender.