Any Wicked Awesome Goalies out there ?

[TheKeeg]

Okay, this is something that we came up with at the beginning of build season but never used it (blocking is not in our strategy). The perfect blocker would be a large pole mounted on the edge of a lazy-susan turn table. When you wanted to block you have a motor turning the table and the previously small rod now takes up close to three ball widths (or more) Couple this with rapid robot movement and it could become a very effective blocker.

https://encrypted-tbn0.gstatic.com/i...IXJv_Fw3vy9vRv

The table could look like this and you mount the pole on one of the corners. (the table would obviously be bigger).
Figured I would share the idea for anyone who wants a blocker. What do ya' think?

[JamesCH95]

Originally Posted by ohrly?

My team put a LOT of consideration into the idea, but ultimately the mentors vetoed the dedicated goalie.

You can find an animation here: http://www.youtube.com/watch?v=408YXf6kGJ0

The idea is up for grabs if you can build it in a week...

PM me if you want a detailed .blend file.

Really creative design, I like it. If made it probably would have been one of the best goalies out there.

However, it would have suffered from the same Achilles Heel as every other goalie robot out there: counter-defense. All it would take is one robot parking sideways in the goalie zone to block the goalie from moving through the goalie zone quickly. the goalie would either have to push the counter-d robot sideways as fast as the shooting robot could re-position and shoot; or retract its blocking extension, drive around the counter-d robot, get back into the goalie zone, and re-deploy its blocking extension before the shooting robot could re-position and shoot.

This is why we did not pursue a goalie design. Defense is important, but a goalie robot is simply too easy to effectively counter.

[Nemo]

Originally Posted by ohrly?

My team put a LOT of consideration into the idea, but ultimately the mentors vetoed the dedicated goalie.

You can find an animation here: http://www.youtube.com/watch?v=408YXf6kGJ0

The idea is up for grabs if you can build it in a week...

PM me if you want a detailed .blend file.

Even though I don't think the goalie strategy is good this year, I like this design. It's quite clever and well thought out.

[gpetilli]

Originally Posted by Nemo

Even though I don't think the goalie strategy is good this year, I like this design. It's quite clever and well thought out.

Unfortunately, it is against the rules.

G12

An ALLIANCE may not POSSESS their opponent’s BALLS. The following criteria define POSSESSION:
1. “carrying” (moving while supporting BALLS in or on the ROBOT),
2. “herding” (repeated pushing or bumping),
3. “launching” (impelling BALLS to a desired location or direction via a MECHANISM in motion relative to the ROBOT),
4. “trapping” (overt isolation or holding one or more BALLS against a FIELD element or ROBOT in an attempt to shield them).

[gpetilli]

Originally Posted by TheKeeg

Okay, this is something that we came up with at the beginning of build season but never used it (blocking is not in our strategy). The perfect blocker would be a large pole mounted on the edge of a lazy-susan turn table. When you wanted to block you have a motor turning the table and the previously small rod now takes up close to three ball widths (or more) Couple this with rapid robot movement and it could become a very effective blocker.

https://encrypted-tbn0.gstatic.com/i...IXJv_Fw3vy9vRv

The table could look like this and you mount the pole on one of the corners. (the table would obviously be bigger).
Figured I would share the idea for anyone who wants a blocker. What do ya' think?

A lazy-susan does not help much since you can not have it free spinning.

G12

An ALLIANCE may not POSSESS their opponent’s BALLS. The following criteria define POSSESSION:
1. “carrying” (moving while supporting BALLS in or on the ROBOT),
2. “herding” (repeated pushing or bumping),
3. “launching” (impelling BALLS to a desired location or direction via a MECHANISM in motion relative to the ROBOT),
4. “trapping” (overt isolation or holding one or more BALLS against a FIELD element or ROBOT in an attempt to shield them).

[TheKeeg]

Originally Posted by gpetilli

A lazy-susan does not help much since you can not have it free spinning.

G12

An ALLIANCE may not POSSESS their opponent’s BALLS. The following criteria define POSSESSION:
1. “carrying” (moving while supporting BALLS in or on the ROBOT),
2. “herding” (repeated pushing or bumping),
3. “launching” (impelling BALLS to a desired location or direction via a MECHANISM in motion relative to the ROBOT),
4. “trapping” (overt isolation or holding one or more BALLS against a FIELD element or ROBOT in an attempt to shield them).

I guess it is a good thing we didn't go with that design.

[ohrly?]

Originally Posted by JamesCH95

Really creative design, I like it. If made it probably would have been one of the best goalies out there.

However, it would have suffered from the same Achilles Heel as every other goalie robot out there: counter-defense. All it would take is one robot parking sideways in the goalie zone to block the goalie from moving through the goalie zone quickly. the goalie would either have to push the counter-d robot sideways as fast as the shooting robot could re-position and shoot; or retract its blocking extension, drive around the counter-d robot, get back into the goalie zone, and re-deploy its blocking extension before the shooting robot could re-position and shoot.

This is why we did not pursue a goalie design. Defense is important, but a goalie robot is simply too easy to effectively counter.

We did a lot of analysis, and we found that mathematically, we still benefit our alliance as long as it takes more than two seconds per cycle for this counter-defense robot to get into place (because we slow them down). Actually, we found that if we block a single ball in autonomous, we are pretty much guaranteed a win. Of course, our mentors correctly pointed out that while it is mathematically advantageous, we still have to get picked, and it doesn't look good if we get blocked.

Originally Posted by gpetilli

Unfortunately, it is against the rules.

G12

An ALLIANCE may not POSSESS their opponent’s BALLS. The following criteria define POSSESSION:
1. “carrying” (moving while supporting BALLS in or on the ROBOT),
2. “herding” (repeated pushing or bumping),
3. “launching” (impelling BALLS to a desired location or direction via a MECHANISM in motion relative to the ROBOT),
4. “trapping” (overt isolation or holding one or more BALLS against a FIELD element or ROBOT in an attempt to shield them).

Well noted. We were more than a little irked with this significant rule change after two weeks. Our prototype was fast enough to lock position before impact, but the struggle was indicating clearly to the referee that the post was stationary.

[Oblarg]

Originally Posted by gpetilli

3. “launching” (impelling BALLS to a desired location or direction via a MECHANISM in motion relative to the ROBOT)

Emphasis mine.

I think this rule is ambiguous; it can be argued effectively that such a mechanism impels the ball in an arbitrary direction not at all determined or "desired" by the drivers.

It could also be argued that there is a "desired" direction, namely "not towards the goal."

Without a clarification, the legality of such a mechanism is unclear.

[RaxusPrime]

Originally Posted by ohrly?

We did a lot of analysis, and we found that mathematically, we still benefit our alliance as long as it takes more than two seconds per cycle for this counter-defense robot to get into place (because we slow them down). Actually, we found that if we block a single ball in autonomous, we are pretty much guaranteed a win. Of course, our mentors correctly pointed out that while it is mathematically advantageous, we still have to get picked, and it doesn't look good if we get blocked.



Well noted. We were more than a little irked with this significant rule change after two weeks. Our prototype was fast enough to lock position before impact, but the struggle was indicating clearly to the referee that the post was stationary.

If someone did do this, the launcher wouldn't need to launch an opponents ball. Instead why wouldn't you use that catapult as a possession of your own alliance's ball. Then you would be able to get an assist without having to move. Though I guess that this would only work if an alliance partner that could catch... But still it would be pretty awesome

[BBray_T1296]

Originally Posted by gpetilli

Unfortunately, it is against the rules.

G12

An ALLIANCE may not POSSESS their opponent’s BALLS. The following criteria define POSSESSION:
1. “carrying” (moving while supporting BALLS in or on the ROBOT),
2. “herding” (repeated pushing or bumping),
3. “launching” (impelling BALLS to a desired location or direction via a MECHANISM in motion relative to the ROBOT),
4. “trapping” (overt isolation or holding one or more BALLS against a FIELD element or ROBOT in an attempt to shield them).

If the post is sliding left/right, this would only be "impelling the ball to a desired location" if the ball were to be deflected significantly left or right. Yes, while an off-center bounce would send the ball off to the side, The major (>>95%) "impulsion" would be caused by a deflection rather than a launch.

I would invoke the "grandmother clause" in this situation. A move like this is pretty clearly a "block" more than anything else (blocking being a deflection).

I think people are way over-thinking the "possession vs defense" side of this game, especially when it comes to goalie. IMHO a robot like this is quite obviously legal, especially because the intent of the device is clear (to move the defense device in front of, not "impel" the ball)

[Whippet]

Originally Posted by BigJ

One could argue "away from the goal" is a desired direction.

This seems useful here.

[Kevin Leonard]

Originally Posted by ohrly?

My team put a LOT of consideration into the idea, but ultimately the mentors vetoed the dedicated goalie.

You can find an animation here: http://www.youtube.com/watch?v=408YXf6kGJ0

The idea is up for grabs if you can build it in a week...

PM me if you want a detailed .blend file.

I love this robot. It's completely legal, and I really wish you had built it. It's really well thought out and would be an awesome addition to nay alliance, especially if you could truss it from the goalie position.

[Andrew Schreiber]

Originally Posted by bEdhEd

From my experience, being offensive wins matches more often than being defensive. I've found that acquiring more points for your own alliance is much more valuable than preventing the other alliance from scoring.

This is incorrect. Logic would show you that preventing an opponent from scoring is exactly the same as scoring the same number of points yourself when it comes to determining who won the match. 0-1 match scores, while boring, count the same as 0-120 scores from a WLT perspective.

However, this changes when you start looking at it from a ranking perspective rather than a winning perspective. Take, for example, years where your second sort was based off of your opponent's score. During those years it was beneficial to allow your opponents to score so long as you could outscore them.

So, what I'm saying is that offense or defense wins matches (it doesn't matter which one) but offense tends to rank higher. Now, the really fun part of the discussion involves the difficulty and risk of effective defense as well as the metagame of the area you are playing. You also need to take into account the difference between effective defense and crappy defense (I'll give you a hint 90% of defense is of the crappy variety, maybe more).

[bEdhEd]

Originally Posted by Akash Shah

Defense is totally underrated. I can see that defense will be a major part of the game after the ball is passed to your alliance member. You will have nothing else to do while you wait for the robots on your alliance to score for the next ball to be put into play. So I can see defense becoming a big part of the game after these week 1 competitions coming up.

I totally agree that defense is underrated. I'm not saying that defense is not useful. As a drive team mentor, I have to think of defensive strategies, too, and defense can be extremely useful in certain situations, but defense is something that can be done well by many teams at a competition, although some robots are better designed for defense than others.

My point is that as far as I know, high performance teams tend to err on the side of being offensive, because the acquisition of points is solely on one's own alliance alone, and does not depend as much on what the opposing alliance does, compared to being defensive. It's all about who dominates the field strategically, and how well a strategy is planned. This is why I mentioned a good possibility of a very strong offensive team choosing a good goalie. This might give a team an advantage, and it might just make the difference between a win or loss, but defense alone usually will not win a match. There's gotta be some offense along with it. You can have an all offense strategy and win a match, but I don't think you have an all defensive strategy and win, unless the opposing alliance scores nothing, and your alliance earns foul points.

[BBray_T1296]

Originally Posted by Andrew Schreiber

This is incorrect. Logic would show you that preventing an opponent from scoring is exactly the same as scoring the same number of points yourself when it comes to determining who won the match. 0-1 match scores, while boring, count the same as 0-120 scores from a WLT perspective.

However, this changes when you start looking at it from a ranking perspective rather than a winning perspective. Take, for example, years where your second sort was based off of your opponent's score. During those years it was beneficial to allow your opponents to score so long as you could outscore them.

So, what I'm saying is that offense or defense wins matches (it doesn't matter which one) but offense tends to rank higher. Now, the really fun part of the discussion involves the difficulty and risk of effective defense as well as the metagame of the area you are playing. You also need to take into account the difference between effective defense and crappy defense (I'll give you a hint 90% of defense is of the crappy variety, maybe more).

I completely agree with this. If your scouting system fails to appreciate the value of defense in a single piece game like this, you are setting yourself up for an outright spanking when you pass up that total shutdown defensive bot in eliminations because they have no OPR.
If a robot is able to effectively seal off one side of the field (or simply deflect even as low as 25% of shots) it doesn't matter how far you can shoot, or how well you can catch, if you struggle to end a cycle, even mediocre robots will take the banner away from you.

Points not being awarded to your opponent are subtly points being added to your own score. If you can slow down the other alliance such that they can only complete one less cycle per match, that effectively gives you an entire cycle's points worth of a head-start on them.

Matches are only ~2 minutes long, and when your 20fps shooter fires a ball right into the awaiting goalie post of the other robot, it is going to go flying in some direction, wasting precious seconds. If a defensive bot can deflect even as low as 2 shots per match, I wholeheartedly believe that they will significantly hinder your score.

Also take it psychologically: from a driver perspective (as a former driver), getting your shots blocked is very frustrating, in the heat of a match nothing makes you madder than an effective defensive robot hindering your progress. As you get angry you stop thinking clearly, only digging yourself deeper. Even the most trained and level-headed drivers will face the pressure of constantly being shut down.

This makes me wish we had made a defensive bot this year.