Oh, boy, scouting will be difficult this year!

[smurfgirl]

Originally Posted by Nibbles

You only need three items from that list. Separate design elements from success elements. Accuracy is important, but not for success.

I think design elements can play into how successful a robot and/or an alliance is. For example, if an alliance is comprised of three robots which can only collect balls from a payload specialist, won't this have an impact on the dynamics of that alliance? They will frequent certain areas of the field more often, perhaps stopping in front of them often for short periods of time. Similarly, wouldn't you want to know if this was true of an opposing team an/or alliance? Knowing things about design elements can help you to strategize for your alliance, as well as to try to predict what an opposing alliance will do.

Also, how do you figure that accuracy is not important for success? If a particular team only scores about 5% of the game pieces they attempt to score... isn't that a way of measuring how successful they are at their offensive maneuvers? Would you choose a team like this to play offense on your alliance? Would this data affect the way you strategize against an opposing alliance?

[Nibbles]

Originally Posted by smurfgirl

I think design elements can play into how successful a robot and/or an alliance is. For example, if an alliance is comprised of three robots which can only collect balls from a payload specialist, won't this have an impact on the dynamics of that alliance? They will frequent certain areas of the field more often, perhaps stopping in front of them often for short periods of time. Similarly, wouldn't you want to know if this was true of an opposing team an/or alliance? Knowing things about design elements can help you to strategize for your alliance, as well as to try to predict what an opposing alliance will do.

Also, how do you figure that accuracy is not important for success? If a particular team only scores about 5% of the game pieces they attempt to score... isn't that a way of measuring how successful they are at their offensive maneuvers? Would you choose a team like this to play offense on your alliance? Would this data affect the way you strategize against an opposing alliance?

It does happen to affect how well the robot performs, yes, but knowledge of it it isn't necessary to select the best robot, since accuracy isn't an economic good in this game. If a robot can only score 5% of the time, I wouldn't choose them not because of the 5% number, I wouldn't choose them because they simply fail to score enough points period. Likewise if they break down continuously, this will affect the total number of game pieces scored and affect the choice accordingly. Accuracy is not a limited good, neither is robustness, but time and game pieces are, and thus you need only be concerned with how efficiently the robot uses them if you are selecting alliance partners (selection of partners is all I was concerned with).

You are right, I did almost completely neglect the strategy element (I only really study the qualification numbers, I should put up a disclaimer, actual implementation will vary by needs). For instance, knowing a robot has Ackerman steering is an exploitable weakness that can reduce its overall effectiveness during a game. Knowing the different autonomous modes it has can help too. I have found it difficult to come up with objective data that would be useful, and have tended to ignore those points altogether and leave strategy to someone else in the pits. In any case I'll be sure to keep a better eye on strategy this season

[smurfgirl]

Originally Posted by Nibbles

You are right, I did almost completely neglect the strategy element (I only really study the qualification numbers, I should put up a disclaimer, actual implementation will vary by needs). For instance, knowing a robot has Ackerman steering is an exploitable weakness that can reduce its overall effectiveness during a game. Knowing the different autonomous modes it has can help too. I have found it difficult to come up with objective data that would be useful, and have tended to ignore those points altogether and leave strategy to someone else in the pits. In any case I'll be sure to keep a better eye on strategy this season

Yeah, I don't know if only going by qualification rankings is really the best way to determine the best robot... actually no, I know that it isn't. The fact that you haven't been running strategy this way and have left it to someone else is good, but you seeing the light on the importance of strategy is even better.

Originally Posted by Nibbles

It does happen to affect how well the robot performs, yes, but knowledge of it it isn't necessary to select the best robot, since accuracy isn't an economic good in this game. If a robot can only score 5% of the time, I wouldn't choose them not because of the 5% number, I wouldn't choose them because they simply fail to score enough points period....

Ok, I can definitely see your argument here. I would figure that accuracy and performance are generally related, but I think we're basically arguing the same thing; ultimately, robots that score lots of points are good.

[SammyKay]

Originally Posted by JDeCola

throwing an orbit ball is a lot harder than throwing a basketball or baseball, they would need a lot of practice, something we are all scrambling to get rightnow as far as the human player in the game

It actually isn't as hard as you think it is. Once we got our orbit ball, a few of us went out into the hallway at our workshop and threw it to each other while we kept on moving to kinda mock a moving robot.

[Chris Fultz]

I think many of the more successful teams elevate the importance of the scouting team and the scouting data.

Good knowledge of your partners and your opposing alliance can make a huge difference in a match. In many ways, the more experienced team members are better at getting this data than the new members (aka "the freshmen"), and can use the scouting time to learn and teach newer members about the different designs and strategies on the field.

Painful, yes.
Sometimes boring, yes.

But, Incredibly important, yes.

[kirtar]

I don't know about that this year. Depending on what your data fields are, there will probably be a good degree of inaccuracy. If a robot scores balls in autonomous, how many did it score? Were 7 loaded, or only 6 (are you feeling lucky?). Of course, scouting is not all about data but also about the team's opinion on how well another team competes. Most likely I'll have simple data collection done by newer members but analysis and observation (in terms of just taking notes) done by more experienced members.

Also, the data that we collected at Boilermaker last year pretty much wasn't used to determine our picks (although this could partially be because it wasn't compiled well).

[JVN]

Originally Posted by Chris Fultz

I think many of the more successful teams elevate the importance of the scouting team and the scouting data.

Good knowledge of your partners and your opposing alliance can make a huge difference in a match. In many ways, the more experienced team members are better at getting this data than the new members (aka "the freshmen"), and can use the scouting time to learn and teach newer members about the different designs and strategies on the field.

Painful, yes.
Sometimes boring, yes.

But, Incredibly important, yes.

Agreed. When I ask myself "Who has the best scouts in the world?" and "Who is the best team in the world?" it is the same answer.

-John

[gorillamonky]

Originally Posted by Vikesrock

Our results were not this conclusive. Our setup used trash cans with caster wheels underneath as the trailers and students as the robots. The rocks were represented by a combination of soccer balls, volleyballs, dodgeballs and basketballs. We did not have any Alliance Station or Outpost walls, but the Outpost players were required to stay in their chairs. With the number of balls we had, we allowed balls that rolled back to any of the human players to be thrown in again. This means we were probably throwing a bit more than 20 balls per match, but I would guess not more than about 30 at the most.

Our human players were getting about 2 to 4 scores per player per match with only 2 "robots" on the field per alliance. The most I scored in one match from the corner was 6 and the most one of our Outpost players scored was 8.

we did the same thing and got even better results using a fabricated trailer, however when we bought real orbit balls and tried again, we found that 1 human player will be lucky to get 3-5 balls in using static robots that where relatively close to te human players

ps, did you remember the hight of the wall, when we where doing ok until we added that, then out accuracy dropped drastically

[Xenosthebest]

One scout per team is fine... for really small teams, which isn't as abundant as one may think. The scout team should be anyone not working in the pits, on the field, or miscellaneous work that cannot be left for later (such as getting that one, specific wrench out of the truck in the parking lot). Let's face it, there are always those seniors that don't work in the mechanical field, such as chairmans or animators, etc, and all they do is sit in the audience and cheer. Give 'em a scout sheet and have them keep an eye on a team and it's human/robot players.

This way, when/if you get into the top 8 and need to pick alliance partners, you have more teams to look at than that one, single freshman recorded in the 2-minute rounds. Also, video footage is helpful, so if you have a camera (with a tripod?), make good use of it.