Triple Cyclers or Defense?

[XaulZan11]

Originally Posted by CalTran

So that's where the question came in. 2013 is somewhat comparable to this game, where cheesecake blockers were a prime pick for alliances, yet the 610-1241-1477 machine swooped in and took Championships with triple cyclers. Do you think that this strategy could win again?

But 2 offense and 1 defense won in 2011, which is probably more similar to Stronghold than 2013 (no full court scoring, protected lane by opponents goal, feeding was a valid strategy...).

[Rangel]

Originally Posted by XaulZan11

But 2 offense and 1 defense won in 2011, which is probably more similar to Stronghold than 2013 (no full court scoring, protected lane by opponents goal, feeding was a valid strategy...).

Well that year 12 tubes was the real max scoring you could do that year since the bottom rows were worth almost nothing. If more than 12 tubes was needed, I could see triple offense being viable. The goal that year was to score 4 logos and prevent your opponents from scoring 4 to give some wiggle room for minibot races.

[BrennanB]

Originally Posted by Rangel(kf7fdb)

Well that year 12 tubes was the real max scoring you could do that year since the bottom rows were worth almost nothing. If more than 12 tubes was needed, I could see triple offense being viable. The goal that year was to score 4 logos and prevent your opponents from scoring 4 to give some wiggle room for minibot races.

And this years max is 8 boulders, the rest is worth significantly less

Originally Posted by Lil' Lavery

Not having a presence in your own courtyard allows the opposition to poach balls out of your secret passage.

There aren't any balls there to begin with.

Originally Posted by Chris is me

I don't know what you guys are talking about - triple cycling is defense. You have to break this paradigm of "are we gonna play defense (push people) or are we gonna play offense" because the best third robots don't fit such a clean paradigm.

Exactly. This is a key point to this game, and any game with non-alliance specific game pieces. However just because offense is defense it does not mean it's the best defense.

[Chris is me]

Originally Posted by BrennanB

There aren't any balls there to begin with.

What are you talking about? There are balls there nearly every match. Teams across the country have been taking advantage of these balls to get "easy" captures. Even against good human players, all it takes is one offensive robot to drive in and block a fast roll to create opportunities for balls to be re-scored.

Exactly. This is a key point to this game, and any game with non-alliance specific game pieces. However just because offense is defense it does not mean it's the best defense.

Ultimately, especially when you are outgunned, you are trying to prevent an opposing capture while ensuring your own. Slowing down scorers with physical defense is one way to do it, and it is effective, but at my events I've found ball control to be more effective. Your mileage will vary.

[Bryce2471]

Originally Posted by Lil' Lavery

Not having a presence in your own courtyard allows the opposition to poach balls out of your secret passage.

Agreed. There are a few reasons why defense will be important for getting out of a division, and this is one of the most overlooked.

[Rangel]

Originally Posted by BrennanB

And this years max is 8 boulders, the rest is worth significantly less

I diagree on them being significantly less. Assuming both alliances get a capture, the winner is who can get the most points on top of that. That includes autonomous, extra boulders, and scaling. One or two extra balls could very well be the difference on Einstein.

[Kevin Ray]

Chris is Me you are, IMO, right on the money. Although we lost in the tie breaker in the finals with a defensive robot on us, I feel that we could have won had our alliance robots been shooting as we had in the quarters and the semi's. We put up 197 points with three high goal shooters. We were missing auto shots and undefended teleop shots which was the reason for the loss. We lost by 15 (three high goal shots). Had we made the two auto shots which we had been making, defense would have had not made a difference.
Now, none of this takes away from the amazing play of 1156 and 287 and their top notch shooting ability, nor the great defense played by 263. I just feel (especially seeing all of the unforced errors by our alliance) that three hi goal shooters could definitely beat a two shooter/one defender alliance, even without having them shoot from the outer works.
And yes, I know this is going to open a whole lot of comments but remember, this is just my view.

[Edxu]

Originally Posted by Kevin Ray

Chris is MeI just feel (especially seeing all of the unforced errors by our alliance) that three hi goal shooters could definitely beat a two shooter/one defender alliance, even without having them shoot from the outer works.
And yes, I know this is going to open a whole lot of comments but remember, this is just my view.

Let's build two second-pick robots and compare.

Robot 1: 50% high goal shot, can consistently shoot 3 boulders per match and cross 3 defenses
Robot 2: defense specialist, no boulder manipulation at all

Robot 1's contribution to the alliance will mainly be adding to their own alliances score by crossing 3 defenses for 15 and scoring 5*3*50% for 7.5 per match. I'm going to assume that endgame and auto are constants and irrelevant for our calculations to make them easier.

This means that Robot 1 contributes ~23 points per match on average to the alliance. However, this doesn't take into account a number of other issues, the least of which being congestion at the defenses and in the opposing courtyard, and the most pressing having no boulders to score after a while.

In Robot 2's case, while they don't contribute directly, they contribute by adding to the opponent's cycling speed through a combination of increasing intake distance by knocking balls away, protecting the secret passage from being locked down and directly blocking opponent's shots.

I would argue that if Alliance Captain and Pick #1 are equally strong on both sides, the side with the defensive robot would win, because it's a lot easier to slow down 23 point's worth of cycles than it is to jostle for balls in the neutral zone, decide on the spot which defense to cross, organize with your alliance and then cross, nevermind shooting positions.

Of course, if the third robot is also an effective cycler, then I would agree that the three-cycler alliance would win. It's just a matter of whether or not that third, effective cycler is there for you to pick.

[Rangel]

From watching many different high level regionals, I think the game will become more like 2014 if anything. Basically what Chris is talking about where robots that currently don't have a ball or where a ball isn't available to be contested, will be productive by playing neutral zone defense until a ball becomes reasonably available to be contested. I think just having a pure defensive robot wastes a lot of time whereas a 3 cycler alliance has the flexibility to take turns playing defense and offense when it's convenient for them. Even if an alliance can't get a third good high goal scorer, a robot that feeds balls is almost as good and may even be a better approach because they will spend less time in the courtyard and more time ready to steal balls or play neutral zone defense. If an alliance does decide to play 2 scorers and one pure defender, I think they better be sure they are the fastest at collecting and scoring balls that they will be able to keep up with 3 shooters if defense isn't as effective as they hoped it would be. Even so, I think alliances that are in the situation where they are the two best scorers would be playing it a lot safer to just pick another scorer to extend their lead. Of course with 4 robots per alliance, captains can pick both a third scorer and a pure defensive robot depending on the situation.

[SoccerTaco]

Originally Posted by Edxu

Let's build two second-pick robots and compare.

Robot 1: 50% high goal shot, can consistently shoot 3 boulders per match and cross 3 defenses
....
Robot 1's contribution to the alliance will mainly be adding to their own alliances score by crossing 3 defenses for 15 and scoring 5*3*50% for 7.5 per match.
....
This means that Robot 1 contributes ~23 points per match on average ...

If the captain and first pick are capable of breaching - and they generally are - then I would suggest that the "3 defenses for 15" is not a contribution that is needed from the second pick robot. The other two robots can cross on their cycles.

In the scenario above, I think the mid-game contribution would be closer to 7.5 rather than 23 - the other points are redundant.

[Theseusgoats]

It is definitely possible but I don't think that there is enough depth in teams to have 3 high goalers, especially at champs

[Kevin Ray]

Originally Posted by Edxu

Let's build two second-pick robots and compare.

Robot 1: 50% high goal shot, can consistently shoot 3 boulders per match and cross 3 defenses
Robot 2: defense specialist, no boulder manipulation at all

Robot 1's contribution to the alliance will mainly be adding to their own alliances score by crossing 3 defenses for 15 and scoring 5*3*50% for 7.5 per match. I'm going to assume that endgame and auto are constants and irrelevant for our calculations to make them easier.

Realistically, a 50% high goal shooter who shoots only 3 boulders is more like a third pick--especially at the Worlds. A 50% high goal shooter making only three shots per match may not even be a viable second choice for an alliance with any expectations of winning. Although this may have occurred in matches at regionals/districts, it is not what realistic, consistent expectations are for the caliber of teams attending the Worlds and as such doesn't make for a fair comparison for a three shooter alliance.
Looking at the number of good high goal shooters already qualified for the worlds it seems that it is quite possible to have at least 3 alliances with good (or better) high goal shooters. Realizing that a high goal shooter can switch to the role of a defense bot but the reverse most likely cannot happen, I would venture to guess that all good (or better) high goal shooters will be picked before defense bots (unless a defense bot is outstanding).

*An interesting observation- At SBPLI the serpentine selection process brought to light an interesting situation. The possibility exists that if there is not a sufficient # of good high goal cyclers the high seeds loose out and the argument becomes moot. The One and Two seeds had to pick defense bots for their third pick.

[pandamonium]

Originally Posted by Theseusgoats

It is definitely possible but I don't think that there is enough depth in teams to have 3 high goalers, especially at champs

Depth!?! With over 50 regionals and tons of districts there will be at least 13 event winners in each division. It is not ridiculous to think that divisions may have 15 - 20 robots capable of high goal autonomous modes. and roughly the same number capable of 4 - 7 high goals a match. 6th 7th and 8th alliance may have 3 shooters capable of cycling, the 24th robot though may not be leaving the 1st seed potentially at a disadvantage.

[Kevin Leonard]

A couple things I'd like to point out-

1) The defensive setup that worked at your specific event isn't necessarily the end-all be-all to be emulated across the world. When your opponents are prolific outer works shooters your strategy should probably be different than if they're batter shooters and should be different than if they low goal exclusively.

2) 3 cycling robots can theoretically occupy your own secret passage enough to deter poaching if an alliance is well coordinated, but a dedicated defensive robot can arguably do so better.

I think that we'll see both types of alliances in the coming weeks.

Originally Posted by Theseusgoats

It is definitely possible but I don't think that there is enough depth in teams to have 3 high goalers, especially at champs

At the Finger Lakes Regional, our alliance of 5254 and 2791 picked up 2383 as our third robot. 2383 was having a bad weekend compared to their usual standards for sure, but they were an alliance captain at both Orlando and South Florida when they were working better. We helped them a bit with autonomous and their shots, and in our first quarterfinal match they scored 3 high goals individually during teleop, as well as one during auto, contributing 25 boulder points of the 82 scored that match. That was despite 2383 missing a few shots as well. Unfortunately they had issues in semifinals that caused them to start missing, but they rebounded from FLR to be the 5th alliance captain at South Florida.

[g_sawchuk]

Let's look at this from an eliminations point of view, as it is at this point that you would be debating whether to formulate a triple cycler alliance, or a double cycler and a defense robot alliance. FIRST has a picking order as follows.

Captain  1st    2nd
1          09     24
2          10     23
3          11     22
4          12     21
5          13     20
6          14     19
7          15     18
8          16     17

If we look at the 3rd collumn (2nd picks), it's evident that, under the assumption that alliances make the correct picks (which is a variable we can't quite account for), that the 8th alliance should have a better 2nd pick robot than the 1st alliance. One could take this and draw the conclusion that, because of this, the 2nd pick robot on the 8th alliance should be more capable of completing tasks that satisfy point-scoring options of FRC Stronghold, as opposed to the 2nd pick robot on the 1st alliance, which may only have a drive train.

For the next sector, we will assume that the alliance captain and 1st pick are both on default cycling for any alliance, and the debate only exists in regards to what the 2nd pick should do.

Since we've seen that the 8th alliance's 2nd pick robot should be strongest, we will decide whether or not they should have this 2nd pick robot playing defense when playing the 1st alliance (a typical QF match). There are several factors that could make having a 2nd pick cycling as well an intelligent move, or a complete mistake.

Clutter
FRC Stronghold as just a field is already cluttered. The goals are all very close together, as they are all on a fairly small tower, opposed to a game like Aerial Assist where there was a very wide goal, or Ultimate Ascent, where the goals were less compressed. On top of this, the surface area of the places to score from is very limited. The game has enforced a rule to only be allowed to shoot from the Courtyard, meaning that you have to get reasonably close as opposed to a game like...

Aerial Assist: Half court shots to score, or full court to get the ball to the other side of the field to then be scored.

Ultimate Ascent: Shoot from anywhere on the field, even full court.

The most recent game that comes to mind in which the area which it was only really easy to score from a close range was Breakaway. However, in this game, the goals were also spread out on opposite ends, unlike the compressed nature of FRC Stronghold.

Now, assuming that we go against this logic of clutter, and do a triple cycler anyways, it's quite plausible that someone who has some decent sense of strategy would play a defender robot on top of all three robots. Now you have four robots in one courtyard. That's going to be fun. And not only that, you now only have, at maximum, two robots in the other courtyard, allowing them to have plenty of room and free space to carry out fast and efficient cycles without delay.

In such a compressed and compact game with limitations on where you can do what, clutter is a problem that will appear with triple cyclers, especially if the other alliance plays a defense robot.

Stopping The Other Alliance From High Goaling
Keep in mind that we are still considering the value of a triple cycle alliance from the view of an 8th alliance, as it is the most plausible situation where this strategy could be valuable.

High goaling has become extremely frequent this season, clearly showing itself as a major factor that determines whether or not an alliance emerges a match as a victor. However, the high goaling, on average, isn't great yet. Or even good, for that matter. Very few robots have shown the ability to shoot from the outerworks (a safe shot) without being blocked. 2056 is capable of shooting from there, but their shot is at a low enough angle that it can be blocked with a tall robot. 118 demonstrated their shot in their reveal that cleared a defender. However, in all honesty, in match, 118 hasn't really been able to carry out this shot very well, and have been favouring the batter shot.

From this, we can continue on the analysis based off of the notion that all robots can be defended against, and that no robots have a safe shot that can't be blocked.

Many robots have displayed very long shot line-up/shooter speed-up rates. I feel like it is safe to say that examples of these teams aren't necessary, as these long line-up times have become an average in FRC Stronghold. Because of this, a robot on defense can effectively push a robot out of their line-up position before they release the ball. Some more recent robots that cannot be pushed as easily, but can still be blocked from getting to what could be considered their "lock down spot" are 1241 (beside the batter), 118 (batter), and 610 (on the wall beside the batter). However, it's fair to say that defense can still be effective on them, in regards to either stopping them from getting to their "safe spot", or slowing them down.

The real kicker in regards to why it's important is simply that, if they aren't defended, they're almost unstoppable monsters, and it's just a free walk in the park for them for shooting. Coming from the view of an 8th alliance once again, you'll be up against some of these elites, and you need to stop them. Slowing down these high goals is essential to success as a lower alliance.

Secret Passage
The secret passage is a gold mine in regards to scoring objects, should it be left undefended. If no one is there, you can go in and out quickly, scoring, and reducing your cycle time phenomenally. If you are able to get more than 6 balls into the goal, you can keep on going, and they'll be forced to feed the balls onto the field for you, so long as you can keep up.

On a side note, I'm shocked that I haven't really seen people putting the boulders through the battrice (I think that's what it's called, it's so forgotten about), when high level teams start to play this strategy and there is no defender. It would certainly mess up their cycle, considering that they're expecting the balls to come out of the lower human player slots.

To be brief, you simply need a defender to stop the opposing alliance from getting these boulders. The importance isn't as much because it's always extremely valuable to have a robot defending to stop them from getting these balls, the importance is that it will hurt a lot if you don't have a robot defending to stop them from getting these balls.

Overall
It seems that it is safe to conclude that, in an 8th alliance, a 2nd pick is needed to defend to stop the 1st alliance from having a scoring palooza, while a 1st alliance (on the other end of the spectrum) is typically the weakest robot selected for an alliance, and thus has a very low chance of being valuable as a cycler, and may just get in the way more than it can help.