At FIT DCMPs, my team advertised ourselves as a defense bot that could climb and score in sandstorm, however we did not advance to eliminations after playing defense on teams like 2468, 2417, and 118 at champs and 148 in a previous competition. I think I noticed a trend in alliance selection: the third pick was a scoring robot that was sent to play defense typically. Do teams have a preference for teams that can score as a backup Incase some thing goes wrong? Is there not a preference on an alliance partner that can climb to level three or level two? I don’t mean anything negative to the teams that didn’t pick us, I just want to know what we did wrong because we will almost defiantly be a defense bot at worlds as well.
Usually our metric for a “defense” pick is one of two things. They’re good at defense or they’re insane enough scoring-wise that playing defense isn’t necessary. If an alliance captain can climb to Level 3, then they’re likely looking for a Level 2 (unless a double climb exists). One other thing we prioritized was Level 2 start, since then you’d be ahead a cycle (3 pts = 1 cargo) to start the match.
So in essence, what I’d look for in a good defense robot:
Level 2 Start
Front hatch or side hatch in Sandstorm
Resistant to pushing/able to push
Strategic blocking & not just being aggressive
Pushing pinning calls to max time without drawing penalties
Reliably parking on Level 1 or climbing to Level 2 EVERY match.
For a scoring robot, it’s essentially the same with the exception of scoring fast and effective, and NOT being pushed when doing it, especially with placing hatches (which require intense precision).
One last thing: be reliable. Radio reboots, mechanism failures, and battery/brownout issues in even more than one or two matches can lead to teams doubting your abilities. If you can climb to Level 2, do it EVERY match. Show off what you can do well and people will notice. At some point in quals, you’re going to need to be more “how can I score the most points and show off my abilities” rather than “how can I rank higher.”
I typically like to have the 3rd robot have offensive capability if possible for back up if something breaks on one of the 2 main scores.
This weekend at 7 rivers we were able to get the 6th ranked robot on our pick list as the 23rd pick. Hard to pass that up even if there was another robot that was better at defense left (there wasn’t but doesn’t mean there couldn’t have been).
Sometimes we’ll pick for just defense, but it needs to be supernaturally good (see 747’s game at their last 2 events), especially if there is scoring depth at the event. Worlds will be all the more likely because you get that 4th robot, so there is some flexibility for running triple offense or 2+ D.
From what I’ve noticed, teams like to have a defense bot that can score at least a little bit as opposed to not at all. Our defense bot at the Blacksburg event was a ramp bot that I believe could do hatch panels. At the first finals match of the same event, team 384, a good scoring bot, played defense (not sure of the reason) while their defense bot scored for their alliance, so I’d assume teams want a scoring defense bot in their alliance for a situation like this
I cannot speak for everyone else, but we ended up being the #7 seed alliance captain at this event. After reviewing our scouting data we thought that it was likely that we could get two offensive bots that averaged more game pieces than us and hopefully could climb to L2 and we would play defense and climb L3. We felt this strategy would be our best chance at winning so we were not even looking at defensive bots. I’m sure everyone else picking had their own criteria that they used.
I was shocked at some of the teams and the quality of robots that did not make it into eliminations at this event. It was just a really deep field.
One of our main criteria for looking at teams is driver ability. It doesn’t necessarily depend on focusing towards defense or offense.
Teams in the 2nd half of the alliance captains are most often looking for more offensive power rather than defense–they may be defense specialists themselves like 1817 was at FiT DCMP. So if you are a defensive bot, you have now narrowed your opportunities to just 4 or 5 draft slots. There may be a number of other ones available. Given the difficulty in judging the relative defensive strength of each team because defense is played sporadically in qualifying, it is difficult to show that your strengths are exceptional compared to another robot that shows great driving skills but spends all of their time scoring.
Sometimes teams just scout poorly. They’ll rank the 25 best scorers, pick team number 19 or whatever in the second round, and then ask them to switch to defense. If that is actually why, then there’s nothing you can do - so don’t really even consider this option as it won’t drive any improvement on what you can control.
- Did you get penalties? More than once? This would plummet you down team lists.
- Did you successfully demonstrate your L2 start and L2 finish most matches?
- Did you demonstrate your sandstorm scoring most matches?
- Were you ever broken?
Teams will often pick those that consistently work and avoid penalties over the high ceiling low floor teams. Top seed alliances often want consistency over almost all else.
Here is what I want from a defensive bot, in no particular order (except for #1):
-Does not draw penalties while defending (this is the most important)
-Level 2 Climb.
-Ability to score hatch on the front of the Cargo Ship in Sandstorm.
-A reliable robot that can withstand hits on/from other teams.
-Enough weight where you aren’t getting pushed around.
-A drivetrain that allows you to play defense (no mecanums, preferably a tank with all/mostly traction, the less omnis the better).
-Can play some mediocre offense if one of our other robots breaks.
-Will listen to us and our other pick.
So far, we haven’t gotten everything that we wanted out of a defensive second pick. Our first defensive pick (6091) was probably our best defender, our second defensive pick (6117) was a really good counter-defender and a good scorer when we had robots down (which happened a couple of times at Troy), and our third defensive pick (7809) was a good defender that could score their front hatch (not the most consistent, but you can’t ask for too much at a district event) and could theoretically send it to level 2, although they just barely missed it. At states and worlds, there should be more good robots and it may be more likely to get a robot that does most of or all of these things.
Remember, you don’t need all of these things, and being really good at some of them can negate not being able to do others, but the more you can do of these, the better of a pick you will be.
Agreed, one of our top scouting metrics for alliance partner #2 which is unfortunately not easily quantitiable is driving ability.
Do people think that a defense bot will have more acceptance at worlds due to the nature of a 4th pick? My team will be having a strategy talk on Monday concerning worlds. Basically all of the kids want to focus on sandstorm, defense, climb and only score in certain situations where it is advantageous to our alliance. All of the mentors want us to go all out scoring and don’t really see the value of defense. I’m the team captain, and I can pretty confidently say we are an average scoring bot (best case averaging 8-9 cycles not including sandstorm), but a good borderline great defense bot (shutting down teams like 118, 148, 2714, 2468). Thoughts?
As far as consistency problems go, at district champs, there only two instances I can think of concerning reliability for my team. One match we lost connection to the field (this has been addressed and will not happen again) and one match we got 9 penalty points for being outside our perimeter (first time getting penalties in defense). However, there were two weird situations in quals. Somehow in one of our matches 2417 got stuck on top of us temporarily resulting in a yellow card (I still don’t know how this is physically possible). And one time we were beached on a ball and used our HAB mechanism to get us in stuck. We drew 6 penalty points, but continued to play defense for 40 secs and got back to level 2 climb, so I think it’s defiantly worth it. (We are addressing this problem by closing our bumpers and sacrificing our floor panel intake.)
I would look at these and say that it’s actually 4 instances of consistency problems. Most scouts will see these issues and make a note of it. I would see the scout’s note and see that you had “weird situations” in 1/3 of your matches.
If you decide to focus on defense, I would suggest that you do it in every match. I would also focus on making sure that you can’t get beached, can’t get pushed around, and don’t leave your frame perimeter while on the opponent’s side of the field.
On Einstein I imagine that matches will be won on defense but lost on defense fouls.
You definitly need to show you can play defense in lots of matches and commit to being a defence bot. We picked 5468 because they played a lot of defense… for 2 whole events. they showed they could play defense against every team they were put against including my team. They could score well and there were better offensive picks than them, but they were denying 15-20 points a match placing a panel in auto, and getting on the hab with seconds left every match. that is more than another offensive bot could do and its one less bot on our side of the field
2522 averaged 22pts in teleop offensively and had an L3 climb (albeit I can only find 1 attempt of it, though successful).
5468 definitely showed more defense though.
There’s a lot of good advice in this thread, but one other aspect is that scouting for objective defense data is nearly impossible in this game, and therefore it’s all subjective. It’s all based on the “feel” of the person scouting them, making it hard to rank options. Just being decent at defense might not have gotten you as much attention as you think.
Got us a regional win and being first picked by an alliance at our next event.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.