Value of defense

[Ekcrbe]

Originally Posted by Ginger Power

Will it be so easy to shut down an elite shooter? In other words, can an elite defender easily shut down an elite shooter? I think an elite shooter should have a decent advantage.

Logic: The shooter can shoot from near max height at an upward angle from the outer works. The defender will have to give a little buffer between them and the offensive robot so as to avoid penalties. If the defender isn't tall enough to block the ball, and isn't allowed to hit the shooter, what can they do? An elite shooter should be able to hit >80% of their shots.

I don't see blocking shots as being what good defense will really be about. The main focus of a defensive bot should be, just as it was in 2012, '13, and '14, to slow down the process of collecting the ball and moving into shooting position. If the shooter isn't in perfect rhythm with its feeders, the balls will spend some time on the ground before the shooter can pick them up. That's where the defender comes into play—its job is to steal the balls and send them back the other way or at least ferry them to the secret passage and play sentinel over them until a teammate can come collect them. Then, once the shooter does get a ball, the defender should hound it all the way until it retreats to the outer works or squeezes onto the batter. A good driver (and that's what this strategy crucially relies on) could conceivably cut the output of all but the best shooters by 50% or more.

[Racer26]

As someone who comes from the region that 1114, 2056, 1241, 610, and many more call home:

I'm really used to planning for being up against the strongest possible teams. The number one thing those teams do is dissect the game and do what is necessary to seed number one independent of the skills of their alliance partners. In FIRST Stronghold, that means ensuring that they get the two extra points available in each qualification match.

To do that, they must DAMAGE 4 of 5 DEFENSEs by CROSSING them each twice, AND score 8 BOULDERS into the enemy TOWER. Conveniently, those two objectives mesh nicely. So long as you CROSS a DEFENSE each time you bring a BOULDER into the enemy COURTYARD and score them, you'll achieve both (assuming your alliance partners are at least functional enough to drive up onto the base of the TOWER to CAPTURE it).

For this game, I don't see any reason a single elite bot shouldn't be able to single-handedly achieve that objective, and I expect the top teams will probably do just that.

As in every previous year I can remember, I expect the top teams will CROSS two DEFENSEs in auto and score 2 BOULDERS. That means they only need to score 6 more in teleop. Totally doable.

[Ekcrbe]

Originally Posted by Racer26

As someone who comes from the region that 1114, 2056, 1241, 610, and many more call home:

I'm really used to planning for being up against the strongest possible teams. The number one thing those teams do is dissect the game and do what is necessary to seed number one independent of the skills of their alliance partners. In FIRST Stronghold, that means ensuring that they get the two extra points available in each qualification match.

To do that, they must DAMAGE 4 of 5 DEFENSEs by CROSSING them each twice, AND score 8 BOULDERS into the enemy TOWER. Conveniently, those two objectives mesh nicely. So long as you CROSS a DEFENSE each time you bring a BOULDER into the enemy COURTYARD and score them, you'll achieve both (assuming your alliance partners are at least functional enough to drive up onto the base of the TOWER to CAPTURE it).

For this game, I don't see any reason a single elite bot shouldn't be able to single-handedly achieve that objective, and I expect the top teams will probably do just that.

As in every previous year I can remember, I expect the top teams will CROSS two DEFENSEs in auto and score 2 BOULDERS. That means they only need to score 6 more in teleop. Totally doable.

I agree that it's doable for an 1114 or 2056, but it might not be the most efficient way after a few weeks or in eliminations when they can count on their partners for something. At that point, specialization could make everyone's job easier. That's the time when an opportunistic defender can do some damage.

[Racer26]

Originally Posted by Ekcrbe

I agree that it's doable for an 1114 or 2056, but it might not be the most efficient way after a few weeks or in eliminations when they can count on their partners for something. At that point, specialization could make everyone's job easier. That's the time when an opportunistic defender can do some damage.

Don't get me wrong, I agree it is rarely the optimum usage of an alliance to do everything yourself and carry your alliance partners, and those top teams recognize that too.

I'm merely stating that the perennial powerhouses will be ABLE to carry their alliance to those 2 ranking points without help, in the event of barely functional alliance partners. As always in FRC, the best defense is an untouchable offence. The 2013 Championship Alliance mentioned earlier in this thread and other powerhouse offensive teams routinely embarrass all but the best defense by simply being faster, more agile, and better practiced. The best drivers practice executing their offensive strategy under defensive pressure, and since most defense is an afterthought, the drivers are usually less practiced, and the offence simply drives circles around them.

I'm expecting that at CMP the GDC will be taking their option to adjust the number of BOULDERs required to weaken the TOWER, as many teams present will be able to score 8 BOULDERs alone.

[evanperryg]

Originally Posted by Ekcrbe

I agree that it's doable for an 1114 or 2056, but it might not be the most efficient way after a few weeks or in eliminations when they can count on their partners for something. At that point, specialization could make everyone's job easier. That's the time when an opportunistic defender can do some damage.

I think it's interesting that you bring this up, because I had a feeling the game would progress exactly the opposite of how you described. At low-to-medium levels of play, I expect to see no robot that can solo a breach and a capture in the same match. At higher levels, even, I predict that no more than 3 teams will do this in the first 4 weeks of competition. As a result, I think it will be much more likely to see alliances that have one robot breach and one robot capture at early events, and at lower levels of play. However, as the game gets more advanced, and fields get deeper, I expect to see alliances with three robots all capable of contributing high goal shots and defense crosses. Utilizing all three of these robots most efficiently could entail many things, based on their relative strengths and weaknesses. However, in the case of an alliance with three hybrid scorers, I expect them to all contribute to both objectives.

[Citrus Dad]

Originally Posted by ahartnet

I think a key thing everyone here is missing is that the defender may not be focusing so much on blocking shots or messing up a robot trying to shot but preventing boulder pickup from the secret passage.

The team update made it very clear the rule is designed such that trying to get a boulder from the secret passage is going to be dangerous if there's an opposing robot over there. If there's not then if you have two boulder scoring robots you can pretty well guarantee a selection of boulders to choose from without leaving the enemy courtyard.

Yes, this will be a key defensive point--it could well draw off an offensive shooter from the other side to defend the feeding station.

There's also defensive solutions to key outworks.

Defense will work much like it did in 2012 and 2013 (and basketball). You will not be preventing scores but rather slowing down the rate of scoring, and trying to disrupt certain strategies e.g., 2013 FCS.

[Citrus Dad]

Originally Posted by joelg236

Easily hey? I wouldn't guess this happens until regional semi-finals in most regionals.

I can see a single top robot completing this task alone (or with minimal help) in qualifications.

[IronicDeadBird]

Generalizing the value of defense is risky, when in all honesty its case by case. If you run a match where none of the robots can be defended against then defense has no value.

[Citrus Dad]

Originally Posted by slickvic2252

What won in 2013 I believe it was 3 shooting robots....

However, in the 2013 Curie final, 4814 was a defensive specialist leading two shooters in 1918 and 67, and nearly won the division.

[Sunshine]

Originally Posted by ahartnet

I think a key thing everyone here is missing is that the defender may not be focusing so much on blocking shots or messing up a robot trying to shot but preventing boulder pickup from the secret passage.

The team update made it very clear the rule is designed such that trying to get a boulder from the secret passage is going to be dangerous if there's an opposing robot over there. If there's not then if you have two boulder scoring robots you can pretty well guarantee a selection of boulders to choose from without leaving the enemy courtyard.

You are correct if alliances do not plan for this. Offensive bots need to rotate through the field so they are not getting in each other's way in the court yard. Have one in court yard while other is in secret passage area, rotate. You'll be in greater risk of penalty.

[mjc49]

While defense is risky, I would expect that when elims come around, teams will have enough scouting information to have a strategy where they may want to slow down the opposition if they are able to get a decent lead during the Autonomous period.

Since the mysterious force known as Murphy's Law happens, we want our drivers to know how to play defense in case something goes wrong. Scoring is what we are building the robot to do and our strategy to plan for but stuff happens!


I'm really looking forward to seeing the game strategies evolve this season.

[Ekcrbe]

Originally Posted by Racer26

Don't get me wrong, I agree it is rarely the optimum usage of an alliance to do everything yourself and carry your alliance partners, and those top teams recognize that too.

I'm merely stating that the perennial powerhouses will be ABLE to carry their alliance to those 2 ranking points without help, in the event of barely functional alliance partners. As always in FRC, the best defense is an untouchable offence. The 2013 Championship Alliance mentioned earlier in this thread and other powerhouse offensive teams routinely embarrass all but the best defense by simply being faster, more agile, and better practiced. The best drivers practice executing their offensive strategy under defensive pressure, and since most defense is an afterthought, the drivers are usually less practiced, and the offence simply drives circles around them.

I'm expecting that at CMP the GDC will be taking their option to adjust the number of BOULDERs required to weaken the TOWER, as many teams present will be able to score 8 BOULDERs alone.

To be fair, they do need to get their partners to at least cross a defense and help them capture, but I agree with most of this. I just think that the majority of "strong" offensive teams early in the season will really be good-ish drivers with functional robots, and they won't be good enough to outplay some smart defending. There will always be teams that really are that good, though.

I also predict that the tower will get strengthened for CMP, although that provision has been in the rules for a number of years now and has never been used.

Originally Posted by evanperryg

I think it's interesting that you bring this up, because I had a feeling the game would progress exactly the opposite of how you described. At low-to-medium levels of play, I expect to see no robot that can solo a breach and a capture in the same match. At higher levels, even, I predict that no more than 3 teams will do this in the first 4 weeks of competition. As a result, I think it will be much more likely to see alliances that have one robot breach and one robot capture at early events, and at lower levels of play. However, as the game gets more advanced, and fields get deeper, I expect to see alliances with three robots all capable of contributing high goal shots and defense crosses. Utilizing all three of these robots most efficiently could entail many things, based on their relative strengths and weaknesses. However, in the case of an alliance with three hybrid scorers, I expect them to all contribute to both objectives.

At low/medium levels of play, there will be all sorts of bumbling around and that may involve a division of breaching and scoring robots, but those matches won't be very high-scoring or interesting to watch. The unlikelihood of getting three or even two of the robots you described together on an alliance will inhibit the hybrid alliance strategy. I think that the qualification matches of a team that ends up seeding very high in the first couple weeks will feature that team scoring points wherever they can and their partners struggling to get through the class A and C defenses or defending the whole match. Very rarely will an alliance breach and weaken in one match, but many of the times that it does happen, I think one robot will be responsible for a majority of the breaching and a majority of the scoring. The other partners still will probably be focused on those one or two defenses that the shooter doesn't want to worry about.

What I consider specialization, which I don't expect until the second half of the season, it one robot staying in the opponents' courtyard and collecting balls that were deposited by another robot feeding and breaching at the same time. I know it's a goal for a lot of teams to be this feeder robot, and I think that's a great strategy to go for because it can allow you to be quite productive, but I think the population of these robots that work effectively will be quite small in the first few weeks and so they won't take over the game dynamic from the start. Once those robots become common and effective, some will get efficient enough to convince a shooter that they can handle breaching and feeding, and that's when scores will begin to take off.

When you mention having three capable robots together, though, I think that that setup will lend itself to the strongest shooter taking over and getting part-time help from one partner who also plays intermittent defense, while the third is the dedicated breacher/feeder. We saw something similar a lot in 2012, where one robot would stay on offense shooting, one would shoot some but collect balls from the defensive side of the field, and the other would stay on defense, potentially lob stolen balls over the bump, and initiate the balance sequence. I don't expect many alliances of three robots playing relatively equivalent roles.

Of course, I could be underestimating the level of play overall, in which case the "lone wolf" alliances won't come to exist as often.

[shadow747]

Originally Posted by JesseK

After walking around AutoDesk's CAD'ed field last night I came to a striking conclusion about the outer works. I bet the zone for effective offense is at least 6-12" away from the outer works. The defending drivers cannot see the rear bumper of an offensive robot , and therefore are really unable to estimate whether the bumper is over the outer works or not. This is huge. I bet it will cause a LOT of penalties in early weeks.

this is true in many ways but in some cases it may not matter. the rules are there to keep us safe and the game fun. defense is impotent but if you focus on defense what happens when you get a team of only defense. or all you know how to do is defense. but your right it is hard to know where your bumper is and it will be interesting to see how many technical arise in this game alone. can we get a counter from first on how many happen?

[Nemo]

A breacher bot could find itself with a period of time toward the end of a match when it's already breached all of the defenses, and it's trading 2 point cycles with the opposing alliance's 5 or 10 point cycles. If that situation is anticipated, the breacher might be better off spending some amount of time during the middle of the match playing defense before the end game rules kick in. As long as the breacher leaves enough offense time to breach and capture for their own alliance, playing a bit of defense could be the right move in some cases.

[GeeTwo]

Originally Posted by Racer26

As in every previous year I can remember, I expect the top teams will CROSS two DEFENSEs in auto and score 2 BOULDERS. That means they only need to score 6 more in teleop. Totally doable.

That will still leave seven defenses to cross. Each robot can only get credit for crossing one defense during auto.

The rules already stated this, but it was clarified in Team Update 01:

Originally Posted by Team Update 01, Section 3.1.3

Once a ROBOT CROSSES a DEFENSE in AUTO, any additional DEFENSES that ROBOT REACHES or CROSSES by that ROBOT during the AUTO period will not decrease DEFENSE STRENGTH nor contribute points to the ALLIANCE.