Soooo just some thoughts

[benhulett]

I've only been involved with FIRST for 2 years, my junior and senior year in high school, 2 of the 3 years that team # 1895 has existed. I'll probably come back a little bit as alumni to give some pointers to next years team and such, but to be honest, I have seen some things I don't like in the actual games....

But before I go on, I have many more positive points (in fact WAY WAY WAY more) than I have negative points

First thing on my mind is "mentor built" bots....

In my opinion, this sort of ruins the game... the mentor is supposed to be there to mentor, not build the robot. You have mentors that have way more experience in engineering than your everyday FIRST robotics high school student. The student is supposed to gain the experience, not let the mentor do the work, give a quick run down of how it works, and then go do the next part of the robot. How is this helping the student. Our team is focused on the students building the bot, while the mentor gives tips. You go to these competitions, and we look at some of these rookie teams, and we sit there and think to ourselves, there is no possible way just the students built that work of art. I don't know if any of you all have seen this, but it just bothers me that they get that advantage.

The only other thing that has bothered me the past two years has been the ranking system. A team could have the best engineered bot in the entire game, but be matched up with other robots that impede their ability to score points and be last in rank. A team could have a not so up to par bot but rely on their partners to give them an awesome rank. At first I thought well everybody must be thinking the same thing I am, the ranking system really doesn't determine how well your bot really does. Your performance as a whole is basically decided upon by your team mates. This is a random selection, so it's hard to counter it. But it turns out from my view (@ annapolis) that every team doesn't see this. The top 8 teams were given the choice of alliances, and only 2 bots outside of the top 24 were selected. I saw much better bots than some of those inside the top 24 that were impeded by team performance but were outstanding performers. So I thought that maybe the bots should be graded on their individual points, assists, and team wins. That way you wouldn't have such a ridiculous drop in rank if your team loses.

These are just my thoughts.... I personally think that FIRST could come up with a better ranking system.

I just want to hear YOUR thoughts, am I being completely unreasonable here? Have I missed a point or two? I appreciate comments from anybody regardless of whether or not they agree.

[EricH]

I'm going to give you a tip--NEVER bring up "mentor-built vs. student-built". Do a quick search on that, and you'll see why. The short of the matter is that not many care one way or the other, provided inspiration happens. You'll see the long when you do the search.

Now, on to the ranking system...
I might be willing to bet that some of the teams in the top 8 didn't have scouting crews. If they did, better robots might have been selected. Or, the better robots you saw might not have complemented the teams doing the picking. Both happen. It might also have to do with marketing, or lack thereof. (Note: I hope you aren't referring to your own team as one of the better robots unless you actually were.)

Your proposed solution would actually be more work for the scorers and lead to a decrease in teams working together. With limited scoring objects; all one team has to do is deny another the _______ and they get an increase in their ranking. But that isn't what FIRST is about, is it? Oh, and the every-man-for himself game format was already tried between 1992 and 1998. Seems that higher-seeded robots at the end of Friday kept getting defended. Collusions was suspected, but not proved. So it became required, and the alliance system was born.

[benhulett]

Indeed you are correct about my proposed idea, I didn't think about it that way. But even though teamwork is involved, potentially awesome robots could be totally knocked out of the way due to bad match ups. I'm not saying our bot was fantastic in anyway, I saw absolutely brilliant robots out there that our robot couldn't even compete with. In no way am I saying our robot is amazing, that wouldn't be fair. But my point still does stand, your team's match up can totally kill your performance in the competition. I'm not trying to be a pompous $@#$@#$@# here, I'm saying there is a lot of hours put into these robots and it's not exactly fair that your bot doesn't have the same chance as other random match ups. I would vote for more matches to let the ranks even out a bit more to be honest.

If your alliance members during qualifications have affected your performance as an individual team, then you've already failed. The reason I say this is because in order to have their performance totally screw you over, you must have decided to rely on other teams. By that nature alone, the robot wouldn't be elim worthy. A lot of people are going to disagree with me on this, but here's how I see it:

In order to be a high power competitive team, you need to be able to do everything on your own, with no outside help, and be able to do it FAST with very little driver influence. Until a team can find this niche, you won't see them on Einstein.




(note: This response is without having looked at your team's robot. So please, if anything is insulting directly towards your team, it was made as a blanket statement, and not towards you/your team individually.)

[benhulett]

I have also thought about that. But when your matched against 3 extremely high scoring robots, and you have two extremely low scoring robots, how can your rely on your robot to win that match. You'd have to have an extremely powerful robot with an extremely experienced team. This is hard to come by. I do understand that the entire idea behind this competition is teamwork, but in a way you have to rely on your team mates. My point isn't to have a single robot based game, I'm saying that the ranking system in itself relies too heavily on the outcome of the match rather than a mix of performance of your bot, and the outcome of the match. IE, let's say you do 60% of the scoring in the match. Your other team mates do the other 40%. Yes you've done a fantastic job!! Awesome! But when you look at the ranks, you've all gone down nearly the same amount. What you've just done has no impact other than the fact that you lost overall. This really made our team feel like crap.

Originally Posted by benhulett

I have also thought about that. But when your matched against 3 extremely high scoring robots, and you have two extremely low scoring robots, how can your rely on your robot to win that match. You'd have to have an extremely powerful robot with an extremely experienced team. This is hard to come by. I do understand that the entire idea behind this competition is teamwork, but in a way you have to rely on your team mates. My point isn't to have a single robot based game, I'm saying that the ranking system in itself relies too heavily on the outcome of the match rather than a mix of performance of your bot, and the outcome of the match. IE, let's say you do 60% of the scoring in the match. Your other team mates do the other 40%. Yes you've done a fantastic job!! Awesome! But when you look at the ranks, you've all gone down nearly the same amount. What you've just done has no impact other than the fact that you lost overall. This really made our team feel like crap.

Winning isn't what scouting teams see. A good scouting system will analyze how YOU play the game, how many times YOU score, how YOU drive, and nothing else. Honestly, most of the high power teams have such a complex scouting algorithm set up, it's pretty impressive to see how it all works out. If your robot can perform, then make it do so; you'll be chosen if you can actually score, and score well.

About the rankings: They don't mean much. If you're good, you will be chosen. The high ranked teams usually have complex scouting systems, and so if you perform, you'll get picked for elims. If not, you won't. It's that simple.


As for my previous post, yes, it is fairly insulting. However, by "do everything," I'm trying to get across that the bots that you will see winning have more capabilities then the ones that you will see losing. Taking last year as an example, teams like Jesters, 254/968, and other very strong teams could do everything, and well, won.

I'm not calling every robot that fails to do all game aspects a total failure. If you'll read again what I said, you need to do whatever you do, or everything you do, on your own, without help.

[benhulett]

Indeed you are correct sir, which is why halfway through that competition when I realized our rank would be stuck very low because of bad match ups, that I could depend on good scouting to pick the best bots. So while watching the alliance selections, I decided to count. I counted two robots outside of the top 24 that were selected. I believe that there were far more suited bots for the top 8 than those in the top 24. This is why I have such little faith in the majority of teams scouting abilities. I did see some awesome scouting by some teams, but not too many.

[fredliu168]

Originally Posted by benhulett

Indeed you are correct sir, which is why halfway through that competition when I realized our rank would be stuck very low because of bad match ups, that I could depend on good scouting to pick the best bots. So while watching the alliance selections, I decided to count. I counted two robots outside of the top 24 that were selected. I believe that there were far more suited bots for the top 8 than those in the top 24. This is why I have such little faith in the majority of teams scouting abilities. I did see some awesome scouting by some teams, but not too many.

You don't have to rely on other team's scouting abilities. So you did poorly on qualifications, go out and meet other teams and market your robot. Make posters of your robot's abilities and hand them out personally to every team. Also, record your own statistics and show it to other teams. Get your team's name out.

In my opinion there are some teams with poor scouting systems that may not look at every robot's abilities. These teams are the teams that just view FIRST ranking method. Most would be grateful if you made their job easier for them by proving your robot deserves to be selected. However, make sure not to be overly aggressive.

Drawing from personal experience (in FVC) last year, there were 99 competing teams and we were ranked 94th due to some controller/pairing problems. However we decided to market ourselves to the top 8 and ended up in an alliance and performed very well. If we didn't market ourselves, there would have been no chance of us being selected (as far as I know, the next worse seeded team was ranked in the 60s).

[Tom Line]

One of the biggest improvements our team made was consistent scouting. It took some butt-kicking (by the student leaders) and some consistent pressure (by the student leaders) and a vastly improved scouting system (again by the students). That resulted in the students 100% handling our alliance selections.

This year has been a "break through" year for the team. We lost a lot of seniors last year and a lot of underclassman have had to step up and learn to be leaders. It's never a pretty process, but they've been up to the challenge and done a good job.

Probably the best feeling I got was when they actually un-invited the mentors to the alliance picking. They felt they could do a good job - and they did!

The point of all this? Not one bit of it was robot related. While the kids DID build probably 90% of our robot (even learned to weld aluminum, etc), that is actually the smallest and least important part of what they've done this year. Anyone can learn to drill and tap. Learning to be a leader is something that few people can do, and it's tremendously exciting to see it happen.

Go Team 1718!!!!!!

[GeorgeTheEng]

I think in some ways there is a very important point about what I see as an intentional design of the ranking system and the alliance system.

We often talk about first as building life skills. Some folks will crucify me and others will love me, but FIRST is not about the robot. The robot is a means to an end, not the end.

Is the ranking system flawed from the perspective of "the best robots should be in the elimination matches"? Yes, definitely. Does the ranking system provide a situation that is similar to real life? Yes, definitely.

When you're in elementary and middle school, performance is all about you as an individual student. Even in High school to some respect. In college, at least with engineering, and the "real world" individial performance means much less. Every day you will thrust into situations where you must rely on working with a team of folks you may or may not know. Who may or may not be able (or will) to do what they claim. In the end though, the team performance will be all that matters. This is something the alliance and ranking system simulates beautifully. I work for a large defense contractor where we are teamed on some programs with companies that we compete against on others. It's the nature of our business. Teams are just the nature of engineering in general.

In addition, the best design is not always the successful one. A group that can market a design that is good enough can often be more successful then an extermely well designed but poorly markets product. This is why scouting and marketing your team are very important. There are other factors that go into alliance selection as well, such as the ability to work as a team player. I know teams that will choose a slightly less capable team that is easier to work with and is able to be a team-player over an excellent team that does whatever it feels like.

I think the ranking system puts a large emphasis on team work, strategy, and being able to think on your feet, that an individual perforance based ranking system would not.

Personally, it is things like this that set FIRST apart from other similar competitions and enables students to be better prepared for college and beyond.

[Brandon Holley]

I have to say here that, and please dont take this the wrong way, in the heat of competition some things can happen and cause some frustration afterwards that lingers for a while.

This sounds a bit like one of these kinds of posts. Yes it is unfortunate to not be picked for the eliminations, but honestly you have to trust the people you've never met, the scouters of other teams.

In arizona this year we didnt receive our robot until 3pm thursday afternoon, and we had a lot of work to do on it. We decided to forego our morning matches on friday so that our robot was fully able to function. Well our plan didnt work out because we werent fully working until the end of the day friday. Saturday morning we hit the ground running and were able to show that our catapult was more than capable.

Teams saw this, and even though we had a horrible rank, we were picked, and we made it to the semifinals.

There are hundreds of stories just like this one all over chiefdelphi and FIRST. For each time scouts miss a quality robot for elims, theres a 100 times they don't. Unfortunately everyone cannot be in the finals, you gotta sell yourself.

[EricH]

Originally Posted by CraigHickman

About the rankings: They don't mean much. If you're good, you will be chosen. The high ranked teams usually have complex scouting systems, and so if you perform, you'll get picked for elims. If not, you won't. It's that simple.

Hey, Craig, you coming to L.A.? If not, come soon. I still remember the year when one team let TWO other teams use their pick list because those teams didn't have one.

By the way, on the mentors/students debate (to stop it before it starts):
Some good discussion on an extreme case
More...
Still more...
If you want more reading...
And finally, one of the best threads on this topic I've seen.

Originally Posted by EricH

Hey, Craig, you coming to L.A.? If not, come soon. I still remember the year when one team let TWO other teams use their pick list because those teams didn't have one.

Hm. Were those potent teams? I have yet to see that happen around me... Then again, I may have missed it.

Also, as far as the whole not seeing much scouting going on, most of the time you won't see it. For the team I used to be with, it was 6 people with clipboards up in the stands, just watching and writing. Then, that night, all the data was compiled and scored. You wouldn't see it happen, but the results would be a very smexy list with scores, ranks, and all other data about a team, with a pick list on the side.

[EricH]

Originally Posted by CraigHickman

Hm. Were those potent teams? I have yet to see that happen around me... Then again, I may have missed it.

No. Other than the one that allowed the others to use the list. (Last year, there were some obviously unprepared teams and one that prepared hastily.)

Also, as far as the whole not seeing much scouting going on, most of the time you won't see it. For the team I used to be with, it was 6 people with clipboards up in the stands, just watching and writing. Then, that night, all the data was compiled and scored. You wouldn't see it happen, but the results would be a very smexy list with scores, ranks, and all other data about a team, with a pick list on the side.

You got that right. My old team did the same thing. I'm trying to develop that for my current team.

[Alex Golec]

Originally Posted by CraigHickman

If your alliance members during qualifications have affected your performance as an individual team, then you've already failed. The reason I say this is because in order to have their performance totally screw you over, you must have decided to rely on other teams. By that nature alone, the robot wouldn't be elim worthy. A lot of people are going to disagree with me on this, but here's how I see it:

In order to be a high power competitive team, you need to be able to do everything on your own, with no outside help, and be able to do it FAST with very little driver influence. Until a team can find this niche, you won't see them on Einstein.

I disagree with just about every word of this. While you should build a self-reliant robot, but calling a robot a failure if it fails to do everything in the game is asinine. Try telling every ramp robot from 2007 a failure because it relied on its alliance partners for points!

Is a high-power, competitive team the only answer?