The math of the cans...

[asid61]

So after discussing with some of my team members about whether we should make our step can grabbers faster (Yes!) I decided to run some numbers on whether it would be worth it in elims to grab, say, 2 cans, faster than anybody else. Or 4 cans but remove something major from our bot.
Or literally have a bot made of can grabbers that grabs all 4 cans in a fraction of a second, using superior firepower (all the cims + as many motors and pneumatics as you can fit in that 120lb-limit), plus our drivetrain for shoving noodles or totes.

I am not counting the value of noodles due the unknown factor in how long it takes to put them in cans or how many can fly to the other end of the field.

Most of this is calculated with a 3-stack assumption for really good teams. This is lowballing IMO, but you can see at the end that more stacks/ team means a more effective can grabber.

With 2 cans to each alliance, then it's evenly matched points-wise per alliance. If, in one match, an alliance can score 5 full stacks of totes + can (not counting noodles here) and one non-can stack that's
36 points/stack * 5 stacks + 12 points for non-can stack = 192 points.
A high number, to be sure, but doable depending on where you are (thankfully, my team is not attending Dallas lol). With no cans, only the 3 you begin with, then you run a point total of
36*3 + 12*3 (tote-only 6-stack) = 152 points.
Okay, that's not too bad, your average still remains relatively good, especially if against other teams you can match taking 2 cans off the step. Plus, you can still score totes instead of picking up cans once you run out to lessen the impact. The other team still gets only 192, which could be beatable overall with other elims scores.

TL;DR: If a really good robot this year can score 2 canned stacks (as in, that's the effective maximum for even the really good teams) then the can grabbing bot is somewhat ineffective in elims. Very situation-specific as to its other capabilities.

But...

If two alliance members can stack 3 fully loaded stacks, and the can grabber can only grab 4 cans, then that's a total capacity of
6*36 = 216 points.
with canned stacks. By contrast, if you can only get 2 cans off the step and a decent stacker (2 loaded stacks) with the two good stackers that's
5*36 + 3 * 12 = 216 points.
TL;DR: At any level of competition, two demon-stackers and a decent stacker w/ two-can grabber would do as well as an alliance containing a bot that grabs 4 cans and flops over with two demon-stackers. If the can-grabbing-flopper can cap 6-tote stacks with cans or load cans with noodles or something, then it has the upper hand.

If an alliance with the capability to get 4 cans the fastest makes it to semis, then it gets very interesting. Because each team plays 2 matches, and there's only 4 teams playing, a lowered score due to loss of the cans could impact the score enough to outweigh the fewer points that a specialty can-grabbing bot would make compared to a fast stacker.
Using the 3-stack/bot model,
4-Can grabber alliance:
6*36 = 216 points.
Other alliance:
3*36 + 6*12 = 180 points.
In matches without facing the 4-can grabber, but still obtaining two cans:
5*36 + 4*12 = 228 points.

Cutting it pretty close there at 228 vs. 216. But, at 6 matches worth 228, and 2 matches worth 180, the average is only 216 points! What an interesting coincidence.
So if you have that 4-can flopper, pushing some noodles around could possibly win you semis.
TL;DR: A 4-can bot that does nothing else can tie with an alliance made of really good stackers if it has stackers of similar level oiance in semis. It can beat the opposing alliance if it can push stuff around or do anything other than flop over.

On Einstein, the 4-can-and-flop is an almost guaranteed win. If every robot there can make, say, 3 loaded stacks over the course of the match, this is the flopper's alliance:
6*36 = 216 points.
This is the other alliance:
3*36 + 6*12 = 180 points.
In order to beat the flopper's alliance, the opposing alliance has to score an additional 18 totes. That's 3 extra stacks, or essentially one entire extra stack per robot on an alliance. And everyone's on Einstein here. 3 stacks is a lot of upper hand.

I'll have to wait until Dallas to see what the metagame is- how many stacks can the best of the best get? 3? 4? 5? The more stacks they can make, the more effective the can-grabber-flopper bot becomes (assuming the alliance is made up of good robots), as it becomes increasingly likely that they can score all 7 canned stacks every single match, and the opponent's average score drops even lower.

Take Einstein again. The opposing alliance has to make do with 3 cans. With only two functioning robots on the flopper's alliance after auton, at a capacity of 4 stacks per non-can bot, the flopper alliance makes
7*36 + 1*12 = 264 points.
Compared to
3*36 + 9*12 = 216 points for the no-canners.
AKA 24 extra totes to score to catch up. I'm not doing the math for semis, but I think it's still going to be pretty close for a can grabber vs. everybody else with higher points totals, although the can grabber would have a better upper hand.

So to summarize:
-On Einstein, cans are important. It could all be decided in the literally the first second.
-In elims, cans are important, but not as much. A can grabber has to still carry its weight, at least a bit.
-If a really good robot can stack 3 canned stacks per match, then a 4-can grabber is about equal to a 2-can grabber with 2 canned stackes in a match.
-If the metagame is 3 stacks/bot, then the 4-can-grabber is pretty good. As the metagame gets better, to 4 or even 5 stacks per robot, the value of the can grabber becomes much higher to the point of being overpowered.
-If a very high-seed team with 2-can grabber picks the next fastest 2-can grabber, it stands poised to win if it can pass through quarterfinals.



Personally I wouldn't want to build just a 4-can grabber as the robot unless we spent all season getting the code and mechanism perfect to the point where it was a guaranteed grab of all 4 in 1/10th of a second, and I knew it was impossible to be faster than us.
Which is totally possible.

Thoughts and errors in my concepts? Assuming all tote stackers are created equal is a fallacy, but it doesn't look like there will be a massive fissure between the fantastic teams and the good teams this year. Still, the 4-can flopper has to do something else other than grab the cans...

[faust1706]

I hope on einstein both teams can grab all 4 bins from the middle, and they all grab them at the same time and it comes down to whose grabber is stronger.

[asid61]

Originally Posted by faust1706

I hope on einstein both teams can grab all 4 bins from the middle, and they all grab them at the same time and it comes down to whose grabber is stronger.

Zip ties on the can or just can shear is my guess. Those things are only so strong.
They might end up simply getting rid of the center cans or something. Nobody could let go for the entire match.

[AllenGregoryIV]

Gamesense had a good discussion on this. Your alliance still has to execute even if you get all 4 cans off the step. A knocked over stacked could end an alliances hope at making it out of any round.

Also noodles in the RCs mean that the opposite alliance needs another 3 totes per stack to catch up. There is something to be said for efficiently noodling the cans and ensuring the noodles stay in the cans. One of the reasons why so many people are amazed by 148 is their ability to play two different roles. They can both acquire cans, put noodles in them and put them on top of stacks while they stack at the same time. If that gives them an extra stack or possibly two out of the feeder station (say 4 or 5 by one robot with littered RCs on top) that is a huge advantage.

The average stacks per robot gets harder as more good teams get on an alliance. 30 totes behind the glass (5 stacks) and 18 right side up in the landfill (3 stacks). The 9th stack is likely way slower for even the best alliances since it's either upside down or pulled from the step.

The arms race (as talked about on GameSense) will be very interesting this year. It's going to be like 2011 but worse since teams can keep throwing power, weight, and design time at it. 3310 has set the bar at their reported 1/4 sec pull. I'm waiting for the 1st team to start wind tunnel testing their mechanism to reduce drag.

[JohnSchneider]

Originally Posted by AllenGregoryIV

The arms race (as talked about on GameSense) will be very interesting this year. It's going to be like 2011 but worse since teams can keep throwing power, weight, and design time at it. 3310 has set the bar at their reported 1/4 sec pull. I'm waiting for the 1st team to start wind tunnel testing their mechanism to reduce drag.

start wind tunnel testing?

[Michael Corsetto]

Common, who doesn't like seeing an entire season decided in less than a quarter of a second?

[faust1706]

that's how breakaway was on einstein. If you didn't stop....469(?) in autonomous you lost the match.

Originally Posted by faust1706

469(?)

Yes, 469.

[asid61]

Originally Posted by Michael Corsetto

Common, who doesn't like seeing an entire season decided in less than a quarter of a second?

Just the last match of it.
Consider:
http://www.youtube.com/watch?v=pmyph...ature=youtu.be
Even if it's sped up, it's definitely possible to reach those speeds and beyond with the infinite motors this year.

[AllenGregoryIV]

Originally Posted by JohnSchneider

start wind tunnel testing?

Have I underestimated you guys?

Originally Posted by Michael Corsetto

Common, who doesn't like seeing an entire season decided in less than a quarter of a second?

We all signed up for this drag race, right?

[Whippet]

Forget wind tunnel testing, we need to start looking at the relativistic effects that could be caused by time dilation at high can grabbing velocities...

Seriously, don't you guys have a regional to be getting ready for or something? Get some sleep.

[Daniel_LaFleur]

Originally Posted by faust1706

I hope on einstein both teams can grab all 4 bins from the middle, and they all grab them at the same time and it comes down to whose grabber is stronger.

I'm waiting to see a robot towed over the step in a tug-o-war.

[Boltman]

I asked that tug-o-war question on Gamesense after seeing all the auto can grabbers.

Cans will be HUGE in Einstein...in Elims not so much

I still think there may be fouls called over alliances "fighting" over RC's from the neutral zone but those decisions should be rare.

[Cash4587]

I really think at this point having the ability to grab cans in auto will just make you 1) a more appealing robot to scouting teams 2) prevent the other alliance from potentially scoring those cans which will be good for you even if you don't score them yourself 3) Make you more valuable in auton to robots who have a 3 tote auton, or 148 who can get all 3 cans and all 3 totes.

Ultimately, I think grabbing all 4 cans in auton with one robot will prove to be too slow unless someone can get all 4 crazy fast like 3310. I will be willing to bet that on Einstein, if divisions get lucky, the winners will be an alliance that can grab all four cans with 2 robots, while the third does a 3 tote auton, or even just drives forward to the auto zone. I am not too sure yet depending on alliance selections and the depth of the divisions.

[lynca]

The CAN race is the only thing that keeps this game from a glorified skills challenge.

The GDC will improve the overall game by allowing ONLY 1 team from each alliance to touch the cans.
(restriction only applies in the autonomous period)