Chief Delphi - Week 1

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)

- Rules/Strategy (http://www.chiefdelphi.com/forums/forumdisplay.php?f=6)

- - Week 1 (http://www.chiefdelphi.com/forums/showthread.php?t=83901)

Quote:

Originally Posted by DeepWater (Post 932142)

As an inspector at Bayou I will sum up to problem with the ranking system. It is Saturday morning and we have 1 robot left that hasn't passed inpection. They are not even close and they only have 2 more qualification rounds left to play this morning. I seriously doubt they will ever get their robot on the field.

Oddly enough the same team is currently ranked 2nd at the end of Friday. Something is seriously wrong when a team can show up at a ROBOTICS competion and never field a ROBOT but be ranked 2nd. Now think about the implications when that team is in the top 8 and gets to pick...

This is exactly why this ranking system needs to go out the window. Go back to Win-Loss-Tie. At least the common person could understand that.

Quote:

Originally Posted by Chuck Glick (Post 932150)

This is exactly why this ranking system needs to go out the window. Go back to Win-Loss-Tie. At least the common person could understand that.

Even with Win-Loss-Tie that robot may still be ranked #2. If the alliance partners are that good, or the opposing alliance has problems, the same thing would happen. I've seen boxes on bricks (not even on wheels) that rise in the ranks every year because of lucky random alliance selection. It may be slightly more pronounced this year, but I'd chalk that up to the number of teams that simply don't know how ranking works this year.

And go ahead, score on yourself the entire match. Try to make it 10-0. I'll see what you're doing, and as a coach, instruct my drivers to score on ourselves too, so we end up winning the match 10-4. Enjoy your 10 seeding points, I'll enjoy my 18.

Quote:

Originally Posted by Doc Wu (Post 931988)

We're seeing a number of teams that haven't even got a fully functioning robot yet. They're struggling to make that work, not understand some esoteric rules that make no sense.

Even teams that are fully functioning and understand the "coopertition" and excessive penalties, want to win. It's human nature to want to win. Try telling your teenage drivers to stop scoring and let the other alliance catch up.

Plus, it's confusing to spectators who are struggling to understand the game that we've had seven weeks to figure out. How do we sell a game to the public where winning is a secondary goal?

I liked the mechanics of this years game, but the rules have spoiled it for me. It may get corrected somewhat after the first week, but how fair is that to those who play in the first week of regionals?

To be frank, the rules should have been read and understood before the robot was designed, let alone built, never mind struggling at competition to make it work. They make perfect sense when you aren't reading them for the first time after playing 5 matches, blowing out your opponents, and wondering why you're still in the bottom half of the rankings.

Teenage drivers should understand before they're out on the field why you'd tell them to let the opposing alliance catch up, or even to score on yourselves. Winning is still a primary goal if you'd like to seed high, but it's in your best interest to not shut out your opponents.

I don't know if it's just me, but it's really bothersome how many people didn't bother reading the whole manual this season. Students on my team were required to read the manual, and score 100% on a 180 question written test to even attend competitions, let alone drive the robot.

In the engineering world, if a customer hands you specs for a widget to design, you'd darn well better read every line. Otherwise, you may miss the part that says "Widget must be capable of X, and must not do Y," and build it without X but with Y. Show that to your customer and see what happens.

It will be interesting to look at a comparison of win/loss record (2 point win, 1 point tie, 0 point loss) vs. ranking points for all teams playing this weekend. I'm thinking it will be pretty close to a straight line.

I agree with the fact that its like watching paint dry, but I can say that my team did not build a robot to sit by or give up, no matter if it drops us from the elimination round or selection we are going out to win. Lets have fun and play the game if your strategy is defense so be it, ours is to score and hang.
and do it often.

Quote:

Originally Posted by SteveGPage (Post 932137)

Thanks for doing all these stats, it will be interesting to see if the averages go up today - and if averages start to go up in week 2.

They actually start to go up as the regional goes on. Combined match scores (red plus blue) also have a general trend of going up as the regional goes on and teams get better acquainted with their robots and driving on the field.

Code:

EVENT        AVG. SEED PT.

SDC        5.74

DC        4.98

GA        4.31

KC        6.56

NH        7.41

ROC        7.08

OR        6.16

NJ        5.05

GT        5.69

GG        6.82

Note: SDC, OR, and NJ are still in progress. GG hasn't been updated in a while.

I'm willing to wait until our team tries the new ranking system before I form a final opinion on it, but after seeing it in action I do have a few concerns.

1) It is counter-intuitive. Perhaps FIRST might suggest a better term is counter-cultural, as one of the stated goals of FIRST is to "change the culture", that might be a positive term. The problem is, however, that the idea of losing 10-0 being a "better" result than winning 4-1 requires more than a thirty second explanation before it "makes sense". If we want to engage people in our events, they shouldn't walk away going "what the heck was all THAT about..?"

2) The goals of the qualifying rounds are not related to the goals of the elimination rounds. In one it is a good idea to lose massively, and assist your opponent in running up the score, in the other it is a good idea to win. This not only means that top-8 teams may have done little to establish their credentials as teams worthy of leading alliances into the elimination rounds, but will also make it more difficult for top-8 teams to scout potential alliance partners. How do you establish a reputation as a good defender or two-way player in a ranking system that emphatically discourages defense? Likewise, how do teams that have built offensive machines that can work around defenders distinguish themselves from their more dainty cousins?

3) The ranking system does not encourage quality robots. If you have two "box-bots" and one non-functioning robot facing three top quality machines, all the weaker alliance needs to to is post their own robots in front of their own goals to ensure that their opponents can't score on their behalf, and they achieve the exact same result as the teams that built much better machines. Likewise the ranking system encourages "giving up" and following this strategy any time you are facing a likely superior alliance. Their is no need to strategize or collaborate with "opponents"... just fold right from the start and let your "opponents" carry you to higher ranking points.

There is no doubt that the rules are the rules. I don't expect them to change. I think it is smart for teams to follow the rules and therefore smart to collaborate to arrange blowouts in qualifying or to immediately give up when outmatched. I might not LIKE it... but I think it is the smart thing to do.

Jason

Quote:

Originally Posted by LinuxArchitect (Post 932116)

Perhaps I missed it, but I'm surprised that there isn't a general good/bad sportsmanship rule. If it were my decision, I would yellow card any team that chose to use a strategy to "game the system."

...
George

Ok, I'm a rookie mentor (at the age of 50, so I do have some life experience to share). I checked with those with more experience and they universally say that gaming the system is part of the game itself. Go ahead and help the other team score if appropriate.

I'm not sure if i can ever agree with that, but I'm a team player...

We were going to do a match where two of our robots would plug up *our* goals, and then our extra robot would proceed to help the other team from scoring. We eventually decided against barring the goals, but we took a grand loss.

my thoughts on the ranking system are that the only time a point is bad is when you fall behind and that in a close match winning is almost 3 times better than losing:ahh: so i think a good strategy with this ranking system is just to score as many points as you can and only play defense when it's tight