Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Rules/Strategy (http://www.chiefdelphi.com/forums/forumdisplay.php?f=6)
-   -   Strategy (http://www.chiefdelphi.com/forums/showthread.php?t=80089)

Chris is me 01-03-2010 14:10

Re: Strategy
 
Regardless of your opinion on the issue, everyone's robot's in the crate, so you'd better tune a strategy to whatever you just shipped.

What can you do better than average? Capitalize on that. (The teams that can do more than one thing better than average don't need a CD post to tell them how to make a strategy.) If you have higher traction than normal, run screens for more agile alliance partners. Got a sweet mecanum drive? Swerve around your defenders and score. Got possession, but no kick? Secure balls and get them to your partners to score. There's something for everyone.

Can't do better than average on anything? Design is an iterative process, you have at least 3 days, probably more to improve something. 65 pounds is a lot.

Lil' Lavery 01-03-2010 14:42

Re: Strategy
 
Quote:

Originally Posted by Chris is me (Post 930020)
Can 148 or 217 go under the tunnel or suspend robots?

Though the best antithesis to the everything argument is probably 148 in 2008.

I'm not saying I agree or disagree with them, but I've heard many a person suggest that 148 was "carried" in 2008. They aren't the end-all, be-all argument that you propse they are.

There are many excellent teams that focus on single aspects of the game, and there are many who "do everything." Look at 254/60 in 2004, for example. Then there are great "specialist" bots from great teams, such as 233 in 2007. It all depends on individual design and strategy, and there usually isn't a right answer to the question. Doing additional tasks does not automatically mean you sacrifice the quality of your "main function" (if you even set a specific task as your main).

Good teams can identify roles within a strategy and how to best fill them. They can create designs to fill those roles, and see how well they fill them and if a single design might be able to fill multiple roles (or if it can be combined with other designs).

Chris is me 01-03-2010 15:04

Re: Strategy
 
I'm a bit more strongly emphasizing "you do one thing well" than "do everything" when it's not that simple. Essentially I agree with Sean. The point I want to make is that you shouldn't spend the build season insisting you have to do everything in order to be competitive; more thought needs to be put into it than that.

XaulZan11 01-03-2010 17:16

Re: Strategy
 
Quote:

Originally Posted by Chris is me (Post 930020)
Though the best antithesis to the everything argument is probably 148 in 2008.

As Sean elluded to, 148 was the last team selected in Atlanta. If that #1 alliance picked someone else, I think you have to really question 148's decision to go with a lap bot.

I think for the prime example of 'Do everything well, not just one thing" arguement would have to be 148's alliance partner, 1114. They were the most dominate team in 2008. They were probably the best hurdler and I think could have been the best herder and lap bot if they wanted to. Just because they just did one thing, doesn't mean that was the only thing they could do.

thefro526 01-03-2010 17:28

Re: Strategy
 
Though I agreed with 148's 2008 machine being the antithesis for do-everything machines my belief on the subject is this:

Do what you can with the resources you're given. If you have a large number of resources available (Time, Talent, Manpower, Money etc...) then try to do as many things as you can without spreading your resources too thin. Inversely, if you only have few resources focus on doing one thing, and doing that thing to the best of your ability and then once you have that thing/function/mechanism worked out move your resources onto improving it or adding additional function to your machine.

We, (I), try to follow this second approach on 816. In 2006 and 2007 we learned that trying to do everything half-a**ed was worse that trying to do nothing at all. From now on we always try to focus on building a strong drive-train (most important part of any good machine, IMO), then a good/decent primary mechanism (Our elevator in 2008, our conveyor in 2009, or our kicker in 2010), and then a good secondary mechanism or function (Our multiple auton's in 2008 with robo-coach assist in 08, our autons and Driving Practice in 2009, and our hanging mech this year (which we removed)) This works pretty well for our team and our limited resources, I'd suggest that anyone else in a similar situation look into doing the same... Or Not.

Chris is me 01-03-2010 17:30

Re: Strategy
 
Quote:

Originally Posted by XaulZan11 (Post 930126)
As Sean elluded to, 148 was the last team selected in Atlanta. If that #1 alliance picked someone else, I think you have to really question 148's decision to go with a lap bot.

For the sake of the argument, I'm more referring to their regional performance, where their hybrid got them #1 seed at St. Louis and #2-3? seed at Bayou. At Championships their hybrid mode didn't seem to work right, which was key to their success. I doubt they would be the 24th selection if they got 5 or 6 lines every match. I remember reading a post that said going by scouting data they were the second highest scorer in St. Louis, below only 217.


All times are GMT -5. The time now is 02:16.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi