Am I the only one who LOVED the seeding system this year?
 

(after week 1)

So I've been thinking since week 5 about this new seeding system that no one seems to like all that much. I think I break from popular opinion when I say that this system does an awesome job of what qualification systems are supposed to do: rank teams. The system made the highest scoring robots seed 1st, regardless of schedule. At least, on paper it did. Strong schedules where a team won every match, weak schedules against multiple powerhouses, and mixes between the two were all good things, and teams that earned it could become top 8 robots with a little ingenuity.

If people throw out their preconceived notions of "winning" matches, and are more willing to consider the tradeoffs of playing matches 6v0 or 4v2, then the system is really great. Unfortunately, far too many teams decided that these strategies were "unfair", "cheating", etc. which really ruined the experience for those who understood the ranking system and wanted to take the challenge. The way the system forced you to honestly evaluate your chances of winning a match while calculating the gain in seeding 6v0 versus 3v3 was in my mind very smart, and I really didn't like how this year there was so much negativity against outside the box thinking.

I especially liked how losing several matches on Friday did not count your team out of the top 8 like last year. A lot more teams had the potential to put on a really great show for a match or two and rocket back up to the Top 8.

I think if the GDC spells out at the beginning of the year in big letters "PLAYING TO WIN MAY NOT BE THE BEST MOVE" to generally quell the downers and teams that decide because you're playing to win the regional and not the match you should go on their DNP list. Because other than that, I loved how this seeding system worked to put the top at the top.

Does anyone else agree? Am I missing something?

Re: Am I the only one who LOVED the seeding system this year?
 

Personally I loved the seeding system from the beginning. The only situation I was afraid of was the 6v0 but the GDC quickly took care of that. This system puts the best scorers at the top of the rankings almost always. In the past, so many times I found that the best teams were rarely ranked first and lots of teams that weren't great were ranked first based on luck of schedule.

Re: Am I the only one who LOVED the seeding system this year?
 

Chris,

I am with you 100% when I say I love the seeding system also. The systems makes teams think creatively when playing the games, and makes the, have strategy instead of just win, win, wind no matter what it takes (as long as it is in the spirit of FIRST).

As you know one of the big factors of my team being a QuarterFinalist at CT is because of the seeding system. We came out of that even 5-6-1, which wouldn't get us into the Elims without being picked.

I hope that as we enter the Off-Season and we start competing at some of very favorite Invitationals, that they decided to stick with the seeding system, instead of W-L-T.

Re: Am I the only one who LOVED the seeding system this year?
 

I loved the seeding system too once we added the 5 point bonus. Hopefully the bonus for winning will stay proportional to the number of points expected in a match. In Lunacy a 25 point bonus for winning might have been necessary.

Re: Am I the only one who LOVED the seeding system this year?
 

After the 5 point fix, I agree that this system generally did a great job of ranking teams according to my own subjective take on their abilities. Sure, there were still some flukes, but in general far fewer than in years past. Under this system, a great robot that has bad luck in one match (a dud battery, a one in a million flip) can still seed high even with a loss or DQ.

Re: Am I the only one who LOVED the seeding system this year?
 

I dislike the outliers this system generated.
During match 87, the field was comprised almost solely of elimination caliber teams. The score was 20-16.

57 of team 67's 276 qualifying points came out of that match. I think before that match they were ranked around 30th, and after that match they were ranked 1st or 2nd.

I also didn't like the way the seeding updated throughout the day, an average would've been better than a cumulative.

As far as separating out the teams, I agree this system did a better job. Its imperfect, but better.

Re: Am I the only one who LOVED the seeding system this year?
 

I echo this (and as a minor addendum, DQs need to be made more obvious when looking at the rankings).

Re: Am I the only one who LOVED the seeding system this year?
 

I interpret that as proof that it works. If only one team on each alliance were elims caliber and that match happened, I would be very dissapointed, but as you said, every team in that match was very good.

Re: Am I the only one who LOVED the seeding system this year?
 

I liked it, with the 5-point addition.

Now, add in a DQ display, and we're all set.

Re: Am I the only one who LOVED the seeding system this year?
 

There is still a mathematical and motivational disconnect with the seeding system.

Your seeding score must be based on your alliances score in order to have you motivated to do well.

Case in point: Archimedes. Going into the match Team 33 had the #1 seeding score (the eventual event #2) going against team #254 the eventual event #1. The only team that could contest this position lost a match before us. We could have done a 6v0 or even sandbagged the match in order to ensure we kept the lead (anything less than us scoring 13 pts. and loosing). This match had a great set of teams and had the potential of being spectacular. We knew this, and our opponents knew this. We made a conscious decision that rather than throw the match, we would go for it. then end result was a spectacular 20 to 18 defeat that catapulted 254 into the lead. This set the new seeding record of 61 points. For us, a 22 to 0 defeat would have been much better with us blocking shots on our goal likely playing against the entire other alliance and one of our partners that had a vested interest in our opponents doing well. That's what I don't like about this seeding system.


In order to get the same benefits, they could have done:

Winners Seeding= Winners points + 2* loosers points + C
Losers Seeding= 2*loosers points.

This would have had all the benefits and not given the incentive to do a 6v0.

Do we regret Qualifier Match 119? Heck no. We had spent two days getting one of our alliance partners ready for the Battle Royal, and it was arguably the greatest match of the year. 20 to 18 with 4 bots off the floor and the last 2 points scored in the final 10 seconds! It was spectacular. Great job 254, 330, and 45 and thanks to our partners 233, and 1111 for helping put on a great show!

And that is one of the reasons I am not in love with this seeding system.

Re: Am I the only one who LOVED the seeding system this year?
 

That match was awesome! I don't see why winning a high scoring, tight, tough, exciting match like that shouldn't count more than a match like the debut of HOT's Breakaway robot in Qualifications 1 at Kettering, where the final score was 3-2.

The scoring system does what it was designed to do: provide an incentive for teams to score big, and make sure only teams that can consistently put up high scores, and beat other good teams, make it to the top 8 and become alliance captains.

I wrote a long post very early in the season about my high opinion of the seeding system, when everyone else was vehemently opposed to it. And I haven't changed my mind since then. I am in full support of the new seeding system and even think it should be used at offseason events. :eek:

Re: Am I the only one who LOVED the seeding system this year?
 

You can add me to that list. When I first read about the seeding system on paper I was very skeptical and thought it was going to suck. But having seen a season of regionals play out with the system in place (and the GDC's quick-thinking modifications), I think it worked amazingly well.

The big thing for me is how the new system allows you to play the game and not worry so much about winning. In previous years, at Virginia at least, losing two matches was reason to discontinue your scouting efforts. All of the teams in the top eight were usually 7-0-0 or 6-1-0. After the almost-inevitable loss due to unlucky alliance pairing on Friday, the rest of the regional was very tense. Waiting for a final score at the end of each match was very stressful for me, as I listened for that single penalty that would make or break my team's chances at the top eight.

Now, with robots ranked on performance rather than W-L-T record, I found the regional to be much more enjoyable for myself and my team. You can just relax, play the game, and let your robot do what it's designed to do, knowing that you will be ranked accordingly. Losing one match by one point isn't such a big deal anymore, and you know that so long as it was a tough loss and your alliance performed well, you will be still be ranked accordingly.

By the same token, I believe this system has done a much better job than that of previous years in correctly ranking the top eight. I didn't see nearly as many complaints about unfair ranking this year. In previous years, on my high school team we always had something to complain about with regards to our schedule, and we never really felt like our ranking was fair. I didn't see any of that this year.

Re: Am I the only one who LOVED the seeding system this year?
 

More or less this new system shows which teams are the best. But like the old WLT system an team can still get lucky and get really good alliance partners and even up in the top 8, when really they are not a top 8 team. It's not perfect, but I think it's better then the old one.

Re: Am I the only one who LOVED the seeding system this year?
 

I see your arguments in where the seeding system was good thing to rank teams, and I agree that it did a great job of ranking good teams (post 5 point addition). However, for one aspect in the game it was not good. The GDC worked hard this year to make sure that spectators could show up and watch, and automatically "get it". But when I had family that came out and watched, and didn't know the game until they showed up and watched a few matches, and when all 6 robots began to score in one side, they didn't understand why. So obviously, 6v0 and 3v3 was something teams had to decide strategy wise, and teams that did it properly succeeded. But is there a way to keep a system where teams are ranked well, and make sure that spectators are not confused (short of them learning all the benefits of how to maximize seeding? :D)

Re: Am I the only one who LOVED the seeding system this year?
 

I agree. Winning big matches puts too much of a skew on the rankings because the LOSING alliance doesn't really get enough credit for the fight they put up. Losing a match 20-2 is much easier to do than losing a match 20-19, and yet, they're worth the same to the LOSING alliance. The WINNING alliance gets far more benefit from the close match. Why shouldn't the LOSING alliance get a proportional benefit to the closeness of the match?

Re: Am I the only one who LOVED the seeding system this year?
 

Agreed. The only thing that I do not like about the seeding system is that the loosing alliance is not accredited for their own scores. I'm going to add a little bit to the proposed scoring system:

Winners Seeding= Winners points + losers points + 2*difference in score
Losers Seeding= 2*losers points.

My major gripe with the system was when a match had many semi-heavyweights, and the scores were close because everyone thought they could win the match, the seeding scores were high. when there was A heavyweight or two going on an alliance, they would sandbag the match and the match where the two heavyweights should have gotten much more qualifying points then the semi-heavyweights.

Re: Am I the only one who LOVED the seeding system this year?
 

Hence Winners = W+2L
and losers = 2L

For the example I gave, this switches it from a 61 vs. 20 seeding score to a 56: 36. It still pays dividends to win, but isn't quite as bad to loose a close high scoring one.

Also if you win 20:2 then winners get 24, losers get 4. instead of 24 & 20 respectively.

As I tell the kids, DO THE MATH!

Re: Am I the only one who LOVED the seeding system this year?
 

Yeah, to be honest, I also prefer this new seeding system over the w-l-t system. The only thing that I would like to see change is that the winner should only get the same amount for a goal in the loser as in the winner... a ball in the opponent's goal is worth twice as much as in your own goal! So, I'd recommend changing this by either:

Winner: 2W + 2L
Tie: 1W + 1L
Loser: 1W

Winner: 1W + 1L + 10
Tie: 1W + 1L
Loser: 1W

I'd actually prefer the first option, as it would make the winner's advantage over the loser and tie proportional to the level of play at the tournament... Just my thoughts!

Re: Am I the only one who LOVED the seeding system this year?
 

The seeding system does seem to help the best teams rise to the top, however, there still seems to be a quite a bit of luck involved. In the 16 - 20 match, team 67 won and moved into 2nd place (which they really deserved.) If our alliance had won (with the luck of a few balls bouncing in our favor,) we might have had a much higher seed than maybe we deserved, and team 67 would have been left much lower.

I like the concept of the seeding system but it seems a bit too sensitive to the match outcomes.

Re: Am I the only one who LOVED the seeding system this year?
 

The issue with this is system would be that you are rewarded for driving down the other guys score. Defense guys will love this system, but Offense guys will hate it. Anytime difference is added into the system, the net results will be more defense unless scoring 1 point is significantly easier than stopping someone from scoring a point.

Re: Am I the only one who LOVED the seeding system this year?
 

I liked the seeding system this year but I was not in love with it. Losing teams should get more credit for losing. I really don't like 6v0. At MSC one of the other teams wanted to go 6v0 and I (along with most of my team) refused. If the losing team is rewarded more for their fight. The match our alliance wanted to go 6v0 was against 469 201 and 2612. Had we not gotten all the dogma penalties it would have been a fairly close match and if we had been rewarded for fighting we may have been ranked higher and that might have affected what we did with our robot (we made a lot of changes at MSC)

Sorry that is very poorly put together but I am tired, If you need me to clarify please ask.

Re: Am I the only one who LOVED the seeding system this year?
 

I didn't really like the seeding points this year. It helped us a MSC, but it screwed us over at Atlanta. Luck has a lot more to do with where you seed. It depends on who you go against.

Re: Am I the only one who LOVED the seeding system this year?
 

What if you did something like:

Win: W+2L+C
Lose: (1/(W-L) * L)+ L + W
Ties: 3T

This way:

A 20-2 match gives 29 SS to the winner, 22.1 SS to the loser, and a 20-18 match gives 61 SS to the winner, and 47 SS to the loser.

Yes, a match won or lost by 1 point results in the C element being the only difference. I think this is a good thing.

Re: Am I the only one who LOVED the seeding system this year?
 

To echo what has been said, I like the seeing point system. It is a good way to keep the best teams somewhere near the top.

THE PROBLEM with the seeding system is that no other sport does it that way. To the outside public, it is seen as very confusing when compared to a WLT record. I had a tougher time describing the POINT SYSTEM to outsiders this year then the GAME ITSELF!

Re: Am I the only one who LOVED the seeding system this year?
 

A 18 point difference from the winner and the loser gets a 7 point SS difference? I agree with most of the earlier ideas that winner's points should not get included in the loser's SS.

If there's a 20-2 match, I say the alliance that only scored 2 deserves only 4 seeding points even if it means a 25 SS difference, since it isn't fair for a horrible alliance to get a big boost in seeding just because they got caught against 3 powerhouses.

With a 20-18 game, it was a narrow defeat, so I'm fine with the loser getting a nice big boost in seeding (36 loser points vs. 61 winner points. Makes much more sense than 18 vs. 61

Alternatively, what if the score difference subtracted from seeding points, such as this formula (winner remains same as current)
loser: (2*L)-(Difference/2) (nothing below zero)

Re: Am I the only one who LOVED the seeding system this year?
 

Rather than editing my post, I'd like to clarify:

20-16 means more points than 33-0.
Also, whats more degrading: A shut-out of 15-0, or your opponents scoring for you for a score of 10-5? (Not sure if that happened this year, but it was a common occurrence in 2006).


I also dislike the fact that this scoring system makes it better to win by penalties than to out-right win.

Re: Am I the only one who LOVED the seeding system this year?
 

I think the losing alliance should have got more QP's.

Re: Am I the only one who LOVED the seeding system this year?
 

I liked the system. I don't like the teams who did not get the system and played defense. They made it a clear goal to point out that defense was not the way to go, until finals.

Re: Am I the only one who LOVED the seeding system this year?
 

I agree that the system worked fairly well, though I would have preferred a bigger bonus be given to the losing teams in the closer games.

One other thing I didn't really like was that it is basically pointless to play defense, especially if you know you're going to lose the game. This makes it nearly impossible for defensive bots to gain a high seed unless they switch to playing offense. Take 294 - they were #1 seed in Newton division, but I believe they told us they prefered to play more defensively from the far zone. They had to use a completely different strategy to gain that seeding spot than one they wanted to use in the elimination rounds. This problem was worse for weaker bots: they could play defensive and hope they were noticed by the scouters on a high seeded team or switch to offense and hope they could get some seeding points.

Re: Am I the only one who LOVED the seeding system this year?
 

The seeding system hurt us sometimes, and helped us sometimes. But, I will say, that without a doubt, it did it's job. I've never seen a system where at the end of qualifications, the top 8 were as proper as they were this year. And I think this is proven by the large number of inner top-8 picks I saw at the events I attended.

I think the problem with the seeding system was the way it was worded...because people still don't get it (as proven by many posts above). There are no winners/losers in qualifications. Only points. So...it should have read something like this:

"There are four goals...two red, two blue. Robots score balls into these goals and points are added to the red score, or blue score based on the goal color. Robots are encouraged to score in either goal to try and get as many seeding points as possible. Various strategies and coopertition efforts are expected (and encouraged) to get the best seeding scores based on the strengths and weaknesses of all robots playing in that round. Seeding points are given as follows:

- Which ever color goals receives more balls, all robots on the field get a seeding score equal to the number of balls scored.

- The similar color alliance will receive two bonuses: 5 points for getting more balls + 2 times the number of balls in the other color goals.

- The other color alliance will receive no bonus.

- If there is a tie for the number of balls scored in the red goal and blue goal (with alliance penalties deducted), each robot receives 3 x the score.

- Point penalties are then assessed to each alliance, if appropriate."

Or something like that...

Re: Am I the only one who LOVED the seeding system this year?
 

I think the seeding system works well at certain events.

For example, at worlds, the seeding system was awesome. The right teams ranked in the right places, and each team had a chance to be #1.

At Peachtree on the other hand, the system didn't work. Blowout match scores were overshadowed my teams that barely scraped off a win.

So I'm not sure, but I like the system...I just don't think it's perfect.

Re: Am I the only one who LOVED the seeding system this year?
 

I hated this years system, they need to go back to win / lose record.

Re: Am I the only one who LOVED the seeding system this year?
 

I guess I kind of made this thread partly to see where everyone lies, and partly because I don't want to lose something that did such a fantastic job of ranking teams because they want to hold on to the concepts of winning and losing being the most important things in a match. I think it makes students and adults think a lot harder when there's more than one qualification strategy on the table. Deciding to take the high risk high reward close match win, or the no risk minimum reward 6v0? Throw a match to maintain your seed, or go for more points with an epic qual you might lose? I guess I like that these choices exist, and I only dislike that people thought it was immoral or un-GP to choose one path over the other.

Re: Am I the only one who LOVED the seeding system this year?
 

Team RUSH had the pleasure of Dr. Woodie Flowers stopping by our pit to chat for a few minutes and, among other things, a discussion of the seeding system came up. I don't remember the exact wording of things, which I'll attribute to the rigorous Champs Schedule, but I was left with the impression that the whole 6v0 idea was not something that the GDC necessarily intended or wanted. It has been a few days, and I'm still exhausted, but I don't think that those strategies aimed at manipulating the seeding system were a desired, or expected, outcome based on the conversation.

That being said, we did find it strategically advantageous to play 6v0 once this season, so we did.

And all things aside, I thought that this year's system, while it could be confusing to newcomers, did a very good job of getting the "good" teams to the top while cutting back on the "fluke" top seeds.

I would love to see a similar, albeit slightly modified with more reward for the losing alliance in a close match as mentioned above, system return next year.

Re: Am I the only one who LOVED the seeding system this year?
 

The bottom line is seeding is based on the degree of how you win in any given match vs just winning. Winning "better" has it's merits in this year's system than the W-L-T. I can live with it but it depends too much on who you play with and against based on a random match schedule. 1 match can have a huge impact on your seeding as noted earlier.

Re: Am I the only one who LOVED the seeding system this year?
 

I guess i may just be repeating what other people have already said, but ill throw another opinion out there. now from a somewhat outside view (i havent had much time for first this year) i cant say i agree with a system where you could make #1 seed by losing every match. it may not be the most GP way to go, but that kind of style would have worked (or so it appears to me). id much rather see teams go undefeated to the top than these seeding points. maybe its just that i didnt have much experience with this system, after 5 years of the old one, but i really hope the GDC goes back to its old ways.

Re: Am I the only one who LOVED the seeding system this year?
 

I think this seeding system is significantly better than WLT, /BUT/. It needs some function of the "closeness" of the match going to the loser. I rather like the formula I described above, or possibly this one:

W: W+2L+C
L: W+(2L/(W-L))
T: 3T

Re: Am I the only one who LOVED the seeding system this year?
 

I loved it pretty much from the beginning (6v0 and all), and loved it more when they added the 5 point winning bonus. I really hope they use it again next year.

The loser getting 2L seems like a good compromise to get rid of 6v0. Alternatively, you could make the loser get W+L so that their seeding performance is tied somewhat to the quality of their opponents. If you got blown out every single match by strong opponents, it doesn't necessarily mean your robot is bad, it might just mean that you've had a really difficult schedule. Having the winner's score affect what the loser gets is a good idea, but having it as the only determinant of the loser's seeding points probably isn't a good idea.

Re: Am I the only one who LOVED the seeding system this year?
 

Only one other person has disagreed with the system and I will be the second. In what world do you get/deserve awards for losing. In a game you either win or don't. The way it was last year and years before the was still ranking points for ties with win/loss records. Yet I see on this thread that maybe we should give the losers more points to show that they tried.

Be honest everyone. Did you not find that the elimination rounds were a lot more interesting and competitive than the qualifying matches? Not just because of the teams playing but because we all knew that you had to win.

I guess that I am tired of the "we need to make everyone feel good" approach to life. If we don't work hard, compete hard and strive to be the best then why should we expect to get rewarded? If I am getting rewarded for getting something I didn't work for then I don't feel right about it.

In the working world you will not be rewarded because you showed up. If you don't produce you will be pounding the streets looking for another job.

Re: Am I the only one who LOVED the seeding system this year?
 

Something this system offers that previous scoring systems didn't is the ability to achieve high seeding points regardless of the difficulty of your opponents. (Given that you are a fairly competitive robot with the ability to think strategically...). The seeding points system seemed balanced, especially after the conclusion of my teams final two qualification matches.

The Situation: Qualification Match 125 (Galileo)

Blue Alliance: 3164, 2467, 1466
Red Alliance: 78, 1058, 2834

Due to the strength of the teams on our alliance, we predicted a high scoring match. The outcome of the match was 21-7, which resulted in 40 seeding points. As stated by Johh Fox:

The innate balance of this system makes me believe that the GDC never intended on teams scoring for one another in the first place, to make the scores as close as possible. This came to me immediately after the conclusion of our final qualification match, which consisted of the following alliances.

The Situation: Qualification Match 144 (Galileo)

Blue Alliance: 1717, 3138, 2283
Red Alliance: 78, 2036, 85

It was known that this match was going to be difficult compared to Match 125. The final score of the match was 12-11, which resulted in 39 seeding points for the winning alliance.

So now to look at the matches side by side, in regards of their overall difficulty and the resulting seeding points.

Five less balls were scored in total in match 144 than in match 125, yet there was only one less seeding point generated in that match than in match 125. This was because the difficulty of opponents was somewhat proportional to the amount of balls we scored (obviously). But the genius in this system lies in that the stakes are high even when the match scores aren't blowouts. (Like in qualification match 144.) This is why I believe that the GDC did not mean for alliances to score for their opponents at all, because in a standard "score for yourself" match, regardless of the difficulty, the resulting seeding points awarded to the winning alliance is balanced depending on the difficulty of the opponents faced.**



I am describing the balance of seeding points of the winning alliance based on the difficulty of their opponents, the seeding points awarded to the losing alliance is a whole different story. (Which I think needs a bit of revision)

Re: Am I the only one who LOVED the seeding system this year?
 

[quote=Steve W;955850]Only one other person has disagreed with the system and I will be the second. In what world do you get/deserve awards for losing. In a game you either win or don't. /[quote]

In auto racing, the 2nd place qualifier doesn't "lose" qualifying. In the Collegiate BCS series, the rankings are based on schedule. A close loss to a highly ranked opponent is more favorable to rankings than blowouts against nobodies. Same is true of NCAA Basketball Tournament. In FSAE, the Auto-cross times are used a qualifiers for the schedule for the main endurance portion. Good Auto-cross times gives you a timeslot with other fast vehicles and at a more favorable time of day.
Qualifications are exactly that. Qualifiers. They are not the tournament, but a schedule designed to have fun, compete, and most importantly sort quality and ability for a seeding structure. A good seeding structure and algorithm matches most teams opinions of rankings. This year's seeding structure was better than most years. With a 12 matches at the FiM tournaments, it was scarily accurate.

There are two good systems proposed in this thread. Mine rewards offense and penalizes DEFENSE in the qualifying rounds. There is another using a difference metric that also rewards DEFENSE in the qualifying rounds. This really depends on what you want the qualifying to be. Both are better than the current metric.

Re: Am I the only one who LOVED the seeding system this year?
 

But in what world is every situation a Win or Lose situation? Doesn't everything vary to a degree with the opportunity costs that the choices and actions entail? It makes each match have a little more real worldly in my opinion.

I loved the seeding system because it left room for recovery, every team will go out and lose communication with there robot at least once this year, but if they have a chance to recover from it with each match, as to work up to overtaking the weaker teams ranked above them, then that is the way it should be. I personally believed that at the three events I attended that the Original top 8 was correct with only a single robot that didnt belong there. Although not in the right order necessarily. But the system is design to put the top 8 best offensive robots out selecting. Which it accomplished almost flawlessly.

Side comment: At The West Michigan District with the top 16 selections in the Alliance Selections, the top 16 ranked robots were picked or were captains. AND the top 5 alliances chose the robot ranked directly below them.

I believe if the Seeding System can rank like a picks list (Through the 1st round at least) then it was doing its job of finding the best.

Re: Am I the only one who LOVED the seeding system this year?
 

Yes! I also don't like that the game has to be played differently for qualification and elimination rounds.

I've also said this before. If we don't learn that the world is a big bad place and you need to try your hardest, how can we expect to compete in life? We really need to get away from this entitlement society we live in. Why should you be entitled to any success if you don't work hard for it. Isn't that the message Dean is always pushing?

Another problem with the seeding system was the huge swings that could be generated through collusion or how a match was played. We had a match at Championships where we played strong defense against a top eight team. They still won the match, but we cut their normal high scoring in half. That team plummeted down almost 40 positions in the ranking system. They never fully recovered to the top eight. We were not trying to damage their ranking position. We were playing the game to our strength which this year happened to be defense. A team should not be affected in that way by winning a match. I'd hate to think how damaging it would have been if we'd won that match.

The system needs work so that it isn't so volatile.

Re: Am I the only one who LOVED the seeding system this year?
 

The problem is that you can lose to a slightly stronger alliance, while another team wins against a much weaker alliance. If you just count win/loss records, that other team looks better than you, even though you might be confident of defeating them directly. The Breakaway seeding system deals with that quite well, in my opinion.

You don't get awards for losing. You get more seeding points if you lose against strong opponents than if you lose against weak ones. It's not a perfect scheme, but it appears to have been very effective this year at ranking robots in the "right" order.

I didn't find that to be the case at all. Maybe it's because the teams I focused on are traditionally high performers and did play to win in qualification rounds.

Re: Am I the only one who LOVED the seeding system this year?
 

I don't see the current seeding system as anything close to the "make everyone feel good" approach. On the contrary- the current system rewards actual performance over simple win/loss statistics.

Under the old W/L/T system, the first ranking metric was obviously wins/losses/ties, followed by ranking points. The problem with using a simple "win" as the primary rank decision is there's no differentiation between a 1 point to 0 win, or a 15 point to 14 point win. The differentiation is deferred to the second metric of ranking points. So lets say Team W, Team X, Team Y, and Team Z are competing at some regional. Teams W and X have awesome robots, while teams Y and Z barely have a moving drive train. The random match schedule pits Teams W and X against each other, and Teams Y and Z against each other. Team W edges out Team X, in a thrilling 15 to 14 point victory. Team Y and Z struggle to move, but Team Y manages to score one ball, winning 1-0. Under the W/L/T system, Team W would be #1 seed, Team Y would be #2 seed, Team X would be #3 seed, and Team Z #4. Team X clearly has a better robot than Team Y, but the rankings don't reflect that, because performance isn't included in the primary ranking metric.

Situations like that happened more often than not under the old ranking system, where an all but dead robot ends up in the top 8. This season, I saw much less of that (post week 1). Statistically, it will still happen if lower end robots end up paired with higher caliber robots in the random match schedule, but by making performance the main ranking metric, it's far less likely. Alliances with strong robots playing against strong robots will rise in the ranks much farther, rewarding overall quality of any given match over luck of the FMS draw.

It's not a perfect system by any means, and I'm not a fan of the potential 6v0 matches. But looking at it logically, on Curie 1114's 6v0 match that ended 29-0 netted 34 and 32 seeding points for each alliance. Team 78's last 2 matches on Galileo (matches 125 and 144), ended 21-7, and 12-11, netting them 40 and 38 seeding points, for tough and fun to watch matches. Those matches catapulted them from 15th place to 2nd, a move that would have been impossible under the old system. Galileo even had a 19-11 match, generating 46 seeding points. I'd take those matches over 6v0 any day.

Is there room for improvement in this new system? Definitely. Is it a step up from the old system? I honestly think it is, and would be sad to see FIRST revert back.

Re: Am I the only one who LOVED the seeding system this year?
 

First always tries to limit the value of defense in the game. With this scoring system they didn't find a balance and as a result "negative defense" (scoring for opponent) was worth more then scoring for yourself. The system works but just needs a little tweaking (Losing team gets points from their own score as suggested)

As I see it, in the working world when you start a project you will be rewarded for accurately assessing the situation and set your scope, and project goals accordingly.

That said from what I saw of championship qualifications vs. regional qualifications. A lot more teams at championships decided that they didn't have a shot at being a top seed and played full throttle to show off their skills.

Re: Am I the only one who LOVED the seeding system this year?
 

Steve, are you saying that a team winning a match 1-0 should be considered equally good as a team that wins a match 20-19, and similarly, that they should be considered BETTER than the team that LOST the 20-19 match?

Yes, winning should have value, but it should not be the primary metric of determining seed position, as it is mostly unrelated to robot performance, however, the alliance SCORES are much more indicative of good robots.

Re: Am I the only one who LOVED the seeding system this year?
 

Are you trying to say that a soccer game with a score of 21 - 16 has better teams than a game that ends up 1 - 0? If you watch basketball you can watch the last 2 minutes to see who is going to win in most cases. If the score was 2 - 0 in a basketball game does your response to the first question remain the same? What I see is an attempt by FIRST to make lots of scoring (like basketball) with no defense, then, say defense is important because we are changing how the game is to be played.

Maybe, if the playoffs were every alliance plays against every other alliance and their ranking scores decided who won the field I might go along with your logic.

Re: Am I the only one who LOVED the seeding system this year?
 

Winning seeding points = W + 2L
Losing seeding points = 2L
Tie = 3*score

I saw this suggested several times and I'd like to echo that it would be the best system for this game. Scoring is simple (not too much on-the-fly math involved when explaining to spectators), the coopertition bonus is still in place, and it removes the incentive to score for the other team, even in tough situations. You get what you score, and a little extra for winning.

Also, did anyone else feel that seeding points for each game should have been posted on the screen along with game points? It just seems strange to not show the actual points that each team gets out of a match...

Overall though, I did thoroughly enjoy the seeding system this year. It did a very good job of placing excellent teams on top.

Re: Am I the only one who LOVED the seeding system this year?
 

I liked it for the most part, after the tweaking, I like being rewarded more for a better performance, tho I would still like a win/loss system, I will take this too


but here is my crazy conspiracy theory (that I don't really believe since the evidence says this was not true):


The GDC came up with this scoring system to encourage the 6v0 games in order to teach us a lesson in life. While competition does drive people to do their best, in order to achieve the greatest possible outcome, we must all work together.



A bit preachy, don't like being preached to on the field, but I dont think its true, and sorry if this has been brought up before

Re: Am I the only one who LOVED the seeding system this year?
 

In Breakaway, it's obvious that a game with a score of 21 to 16 DOES generally have better teams than a game with a score of 1 to 0. I don't believe 1075guy was making any reference to soccer...

Re: Am I the only one who LOVED the seeding system this year?
 

I like the three alliance format with randomly drawn qualifications. Sure it means that we are sometimes paired up against a powerhouse alliance or are playing 3v2, but then again sometimes we are on that powerhouse alliance or we are the ones with the dead robot. It is a great lesson in the importance of paying attention to details. On the other hand... I do not like 6v0 matches. I do not like other teams scoring "for" my team. I do not like having to switch in the middle of the match to scoring for my opponent when it is obvious that I will either win/lose. I like being encouraged to try our best throughout the entire match. As such, I fall in the camp that think the seeding system needs some minor tweaking. My recommendation is similar to others in this thread:

Win: W+L+C
Tie: 2*T
Loss: 2*L

Generally this scheme promotes scoring; however, this scheme eliminates the 6v0 motivation, partially reduces the motivation to score for opponents since goals are weighted equally, and rewards both alliances for a high-scoring close loss. Perhaps obviously, the key component is the bonus points awarded for winning (C). In games where the GDC wants qualifications to emphasize offense, choose C to be small (or zero). In games where the GDC wants to promote winning through defense, let C be large relative to the expected value of the total score.

Re: Am I the only one who LOVED the seeding system this year?
 

Lots of good discussion.

My general comment is that the FIRST community is robust to many things. If we can survive the game Coopertition FIRST which had a flawed game, a flawed tournament structure AND put the drive teams in psychological pressure cooker that was beyond belief*, we can survive anything. But that does not mean that it was a good idea.

By which I mean, that yes, I agree, the sorting system in the end turned out to be not as bad as we all feared and in fact, you could argue it was even better at making sure certain kinds of good teams made it to the top of the ranking than FIRST's historic methods.... ...BUT... that does not mean this was a good sorting method. IMHO.

Here are specific cases to consider.

Curie Match 100.

Teams:

469 888 111 Vs. 1114 231 288

Team 1114 was ranked #1 at the time. Team 111 was ranked #2.

If 1114 plays it straight, it may have won, but it probably would have lost. Rather than give 111 a great seeding score by going down with a large losing score, 1114 and company decide to play it 6 vs. 0.

Final score: 29-0

Both 1114 and 111 get a good score by most team's standards but 111 is denied the opportunity to gain a lot of ground on 1114.

Is this right? I don't think so. Let me rephase that. Is this smart and legal play by 1114? Yes. Is this the type of activity that FIRST should be encouraging? 100% No (imho).

Curie Match 137.

Teams:

1379 440 115 vs. 1114 2667 3234

1114 is in 2nd place in the rankings (but 111 has played 10 and 1114 has only played 9 so no surprise there). 111 is in a pretty good position but 1114 can over take them if they get a monster seeding score.

1114 and company should win but their opponents probably won't likely put up enough points.

1114 decides to score for their opponents. The problem is they get a little too clever by half and score too many points for their opponents. The score should have ended in a 13 to 13 tie but for 2 DOGMA penalty points by their opponents.

Was is smart for 1114 to score for their opponent? Yes. Should FIRST be setting up a system where this is what teams need to do to seed high? Again I think not.

Your thoughts are welcome.

Joe J.


*one of the members of our drive team checked himself into the hospital after one regional, thinking he was having heart issues -- it was just the stress of the tournament -- stress that FIRST baked into the cake

Re: Am I the only one who LOVED the seeding system this year?
 

In general I liked the direction the seeding system moved this year toward rewarding robots that accomplish the game objectives well.

I do agree with the opinion expressed by many here that the loser should not receive any part of the winner's score. There are two main issues cause by that aspect of the current system. The first is the 6v0 problem where it may be in a teams best interest to forgo scoring for their own alliance. While the strategic implications brought about by this strategy are interesting, I believe most would agree that, given the chance to design from a clean slate, it would be best to make the 6v0 outcome less desirable. The other issue occurs at regionals with a small number of teams that are clearly a cut above the rest. At Northstar playing a match with 71 or 1986 in it (either with or against) was better than a win in most other matches. A few teams in the top 8 may have seeded a bit better than they should have by having 2 or even 3 matches with 71 or 1986, the fact that they didn't need to be on your alliance makes this scheduling much more likely.

With regards to defense being discouraged in qualifiers, I am all for that. I am not saying that defense is not an important and interesting part of FRC games, but to me it makes sense that a robot that primarily specializes in defense should not seed in the Top 8. NBA players like Bruce Bowen (a few years ago), Ron Artest and Trevor Ariza are known as great defenders, but if there were a giant redraft of all the NBA players they would not be anywhere near the top. The top would be players like Lebron James, Kevin Durant, Kobe Bryant and Dwight Howard, players that shine on the offensive end, but are also flexible and can contribute defensively in order to win games.

Re: Am I the only one who LOVED the seeding system this year?
 

I agree. I think any scoring system which can force teams to make unpopular or confusing decisions needs improvement.

I have no qualms about teams scoring for the other alliance. 188's strategy to make 1114 lose their super cells was one of the neatest strategies I've ever seen. If the scoring system forces top tier teams to score for their opponents to ensure they get a high ranking, I'm cool with that too. Do what you need to, I'd love to join your alliance! ;)

That said, it's confusing for the audience. And because people interpret it as borderline legal/cheating (when it is totally legal) it doesn't make many friends. And that's unfortunate.

FWIW, I think this system did do a better job than straight WLT. I like the idea of giving the loser their own score (or some multiple). Much beyond that I think it gets too complicated.

Re: Am I the only one who LOVED the seeding system this year?
 

I strongly applaud the efforts of the GDC to downplay defense and reward offense.

Unlike most of our favorite sports, defense is simply easier in your typical FIRST game. Scoring requires a robot to perform complex actions. To "play defense" is for the most part to simply get in the way. Two great teams might play a 1-0 match, but two poor ones will seldom play a 20-19 one.

Re: Am I the only one who LOVED the seeding system this year?
 

First, I just want you to all know that I dislike this seeding system immensely. I can understand the need to rank robots based on performance, but there were too many ways to exploit the system this year that it backfired in my opinion.

What I Liked:

1. Ranking robots based off of performance. The higher your alliance scored the better you did. I like this part! But why should you have to score for the other alliance if they suck? What's wrong with a 15 - 0 match? If the good alliance scored for the other wouldn't it just mess up the ranking system and defeat the purpose?

2. Two loses wouldn't kill your chances of being in the elimination rounds. My rookie team was in the top 8 in TC District event despite losing a few matches Friday due to technical problems. However on Saturday we performed well, and our rank showed it.

What I Hated:

1. Socialism. Yes, it is socialism. Your alliance could throw the match and get more seeding points than if they actually tried. Yes I understand that you could get the most points by scoring huge numbers for you AND your opponent, but that won't result in accurate rankings unless the alliance is composed of several top-notch teams.

2. Teams throwing matches. I understand why they did this, but honestly there is no honour in winning this way.

3. Irony. There were several times where I saw a team throw a match and then compete their hardest against teams in elimination rounds. Why not keep the seeding system from quals to elims if it is a good one? Basically they're saying that in qualification rounds teams shouldn't try and win, but instead get the most seeding points so they can be number one - even if that means throwing a match.

In order for this type of seeding system to work, the GDC needs to come up with a game that is designed around it. The trade off is that the game would be extremely hard to understand for the audience.

Personally, I'd like to see a game where robots work together but are ranked on their individual score. They could do this by RFID tags or something like that. But given the issues this year with the very simple automated scoring, I think refs could actually track the points better.

Re: Am I the only one who LOVED the seeding system this year?
 

I certainly didn't love the seeding system this year.

But I didn't dislike it as much as I thought I would.

But really, like or dislike is irrelevant in the face of the logical inconsistency inherent in the system.

Specifically, as has been mentioned previously... the losing alliance has basically no influence on the score they receive.

Consider an alliance where all three robots break down, flip, or are otherwise incapacitated in the first 30 second of the match. They lose, 8-0 and each robot... even though it doesn't work... gets 8 points.

The next match the losing alliance plays, and while all three robots work... they don't work so great... but manage to pull off a close 5-4 loss. They get 5 points.

The robots that don't work end up ranking higher than those that do.

So the system works great, if you're winning. But if you lose... particularly if you play hard, and play to win, but lose close matches... you end up ranked behind robots that don't work and were continually blown out.

I don't have a problem dropping the WLT ranking for a points based system... but I think FIRST could go a long way to improving the formula for next year. I wouldn't want to use this formula again.

Jason

Re: Am I the only one who LOVED the seeding system this year?
 

It also seems there is an appearance versus reality issue:

If you score for your opponent, it looks like you are helping them by keeping the score close. The reality is that scoring for your opponent only helps you.

If you really wanted to help your opponent, you would beat them 20 - 0.

The "coopertition" aspect seems backwards in this respect.

Re: Am I the only one who LOVED the seeding system this year?
 

i agree with Jason....and with Joe ... a good start towards a better system but full of imperfections.
What we saw at CMP was proof of that.... no teams should ever be put in a situation where if they win they may be at a disadvantage..or if theyplay hard and lose aclose match they could be at a great disadvantage...and losing a match big would help them..... the teams at CMP that did this... were playing by the rules and more power to them....

A team that puts in 10 points should get more seeding points than a team that puts in 5 points...
It is really as simple as that...

How about something really simple.... the alliances get their respective score ?

Teams playing other teams that are good will have to score more points to win...

This would, however, lead to more "blowouts" of teams and that would not be so good.

How about a scoring system that would be your own score divided by the difference between the two scores.... so a 10-5 win would give 2 points to the winner and 1 to the loser...???
hmmmm perhaps a bit extreme....probably lead to good teams scoring low against not so good team or scoring teams FOR the other team ...

3-2 would yield 3 points for the winner and 2 for the loser...
hmmm this would not lead to big scores...unless both teams were scoring...


Sliding scale? Multiply a winning team's score by 5 if they win by 2 or less
Multiply a team's score by 3 if they win by 2-5
Multiply a team's score by 2 if they win by 5-10
Multiply a team's score by 1 if they win by >10

Nothing seems to work very well here....

I think it should come down to points scored...by both teams... with some sort of bonus multiplier

Actually the old way we did it.... prior to this year... isn't bad...
Perhaps a modification of that...

Winner's team score plus loser's team score...

Loser gets 2x their own score...

match is 10-5 Winner's qual points 15 Loser's qual points 10
In a match like the one at CMP Winner 29 Loser 0
Winner gets 29 points Loser gets none... (hmmmm not much of a temptation to throw a match..

This actually seems like a pretty good scheme... and very simple...
A loser can NEVER get the same score as a winning team....and will always score at least 2 lower.
good points... promotes scoring... gives good losing teams a decent score based on their OWN abilities
inhibits throwing matches... no advantage for scoring for an opponent...

bad points... does not prevent blowouts...

any suggestions? perhaps a penalty for too big a difference? (I don't like that....I think that GP should prevail.... that teams should realize that beating up on a team that can't do much is the wrong thing to do....but....)


????
No perfect system here.. I think

Re: Am I the only one who LOVED the seeding system this year?
 

Win: C + 2*L + W.
Loss: L + W.
Tie: 3*score

Note: For the winner, L is unpenalized, W is penalized. For the loser, reverse that.

Strategy/Analysis: Getting your own score as a bonus if you win gives an incentive to score, score, score. Getting double the loser's unpenalized score is an incentive to not have a blowout. If it is a blowout, say 10-1, the loser gets 11. The winner gets 12 + the constant. If the constant is fairly small, it keeps the seedings close enough that one bad match can't throw you way down the rankings, but a lot of them will. Ditto for one good match and a lot of good matches.

Doing a difference factor would need to be carefully thought about, and it should probably be applied to both teams. I'd rather not get into that right now.

Re: Am I the only one who LOVED the seeding system this year?
 

I dont know, maybe its just my own preferences as a driver for 4 years, but defense is fun. I mean, shutting out a top team is just as rewarding, AND should be just as beneficial to your score as playing for the win. stopping the best team was the way to win in previous years, or so i observed. Also, maybe it was just because my team was only at NJ week 1, but every time we won a match we went down in standings. yeah, that felt awesome. now i know the rules were changed, and that would have made things different, but i still dont like to see the losers get more ranking points than the winners. thats just not how any other game works.

Re: Am I the only one who LOVED the seeding system this year?
 

This doesn't quite fix the problem. If I'm playing against a team ahead of me in the standings and have decided the match is a loss I will still score in their goal because every goal I score for myself moves them further ahead of me, while scoring for them moves us both ahead equally. Any new system can't give the winner more for the losers score then the loser gets.

Re: Am I the only one who LOVED the seeding system this year?
 

You also can't give the loser any more for the winner's score than the winner gets, which is what would have happened this year without the Coopertition Bonus and the addition of the constant. At that point, it's winner's and loser's scores (penalized and not) and whatever constants you throw in. Why even bother then? Go with total unpenalized points for the winner, loser's points for the loser, and watch teams who lose a lot of high-scoring matches by a point or two end up at the top of the standings. Then somebody complains that robots who aren't the top robots are up there, which is probably one reason we had this system in the first place.

Honestly, any system you do is going to disappoint somebody. This year's made a lot of people happy or almost happy (due to encouraging 6v0 in certain situations). The ones that aren't don't like 6v0, and I don't blame them.

Re: Am I the only one who LOVED the seeding system this year?
 

I liked that the seeding system ranked teams based on their performance. I would also say that there was a more accurate rating than normal. But I would not necessarily put this all on the new seeding system. This year at events around the country, people got to play more matches than they had in previous years. The effect of more matches is huge. It is harder to get lucky and carried to a top 8 position when you are playing 10-12 matches than it is playing 8. I would say at district events in 09 the top teams seeded in the top spots more often than not, not just because they were good, but because they played a lot of matches and were able to consistently perform.

Under the W/L/T system you had a straight forward and simple way to rank teams. Anyone could understand that. However if my roommate came to an event under this seeding system and just started watching he would be completely confused on why people were blocking their own goals and scoring for their opponents.

My team played in a 6v0 match at the Kettering District event, and debated doing it on multiple other occasions. The fact that three teams came to us before a match and told us they were not going to try and win makes me think this system is flawed.

This system would work fine as a second tiebreaker. Instead of Ranking Points I think that the Co-Opertition Seeding would be a great way to show how good teams were. But your first tiebreaker always needs to be wins.

If FIRST's overall goal is to make a game that is ready for TV, then they need to get rid of the seeding system. Games should be played to be won and anything that takes away from that undermines the enjoyment of what is on the field.

Re: Am I the only one who LOVED the seeding system this year?
 

I know this has been stated in other forms by other people, but I'd like to reiterate it:

My problem with the 2010 seeding system isn't so much the system itself, but rather it's implications. I don't like that the goals and thus the strategy of match play is different in qualifications than in eliminations. We should rank alliances using the same criteria we use to eliminate them. It makes no sense to do it any other way.

Re: Am I the only one who LOVED the seeding system this year?
 

That would be great. When FIRST adopted the W/L/T seeding, teams widely hailed it as a wonderful thing. Suddenly qualification match play aligned with the way elimination matches were going to be played.

Unfortunately, you don't get a reliably good ranking result with a dozen or fewer matches. Unless each team plays both with and against a large fraction of the other teams, a simple win/lose tally won't work well.

If the goal is to have the "best" robots play the elimination rounds, the Breakaway seeding algorithm is pretty good. Unless you've designed your robot to play defense at the expense of scoring ability, I don't think it should make a big difference in match play.

Re: Am I the only one who LOVED the seeding system this year?
 

1. Many years, the offense related task is significantly more difficult to do than the defense task of stopping said person by hitting them. Yes, there are very skilled defensive drivers that have mastered the art of pummeling bots, however it is typically easier to get good at defending than it is to get good at scoring.
Case in point this year, a bot could stop a scorer by driving between the scorer and the goal. In order to score the ball, the scorer had to then move the ball to the other goal without getting a penalty, without loosing control of the ball, and then score it. The other bot had to drive sideways and get into the way. For this reason I like offense related ranking systems.

2. Changing the seeding strategy from Quals to Elims also requires teams to modify their stratagies. This adds a new role and importance to the scouting/strategy team. We customized our strategy for each qualifying match in order to get the maximum amount of Seeding points. This was a very inspiring challenge for me and my team.

Re: Am I the only one who LOVED the seeding system this year?
 

It promotes a certain KIND of best robot. As you point out, not defense or even infinite loop machines like #469 (they do OK but they had to fight to be even a ranked team).

By the way, I am okay with FIRST deciding to promote a certain kind of robot. While all robots inspire (or can) some are more fun to watch than others. Watchability is something that helps advance the goals of FIRST so I am perfectly okay with that.

Perhaps my biggest bone of contention (which the winning bonus solved for the most part) was that if teams tried to maximize their seeding points, it was going to be very confusing to watch (my mom would have not been happy to see Match 100 on Curie -- rules and strategy be damned, it was confusing to watch).

Joe J.

Re: Am I the only one who LOVED the seeding system this year?
 

Not only would it be confusing for spectators, but potentially degrading for the other robots of the "losing" alliance of a 6v0. I'm sure 231 and 288 had agreed to 1114's plan in Curie match 100, but ultimately it cost each of them a chance to showboat their robots. By that point Friday, the rankings were starting to become established, and scouts of potential top 8 teams may have been in the stands watching. By sitting in the goals for the entire match, they weren't showing off their possible offense/defense/drive skill/ball handling abilities.

Re: Am I the only one who LOVED the seeding system this year?
 

To summarize some ideas:

Winners: W+2L+C
Losers: W
Tie: 3T
Promotes winners scoring for low scoring loosers, and known losers opting out for a 6v0. Ties are great, but near ties result in the winning team getting approximately 300% the loser score. (20:19 = 58+c:20) High scoring matches will cause giant boosts in rankings.
********************
Winners: W+2L+C
Losers: 2L
Tie: 3T or 2T
Promotes high scoring from both sides. Winners should score for opponents if scoring for opponents is less than 2x as hard as scoring for themselves during a blowout. Defense should be minimized for small values of C. As C increases relative to scoring possibilities, the value of the win increases. As the value of a win increases, there is a higher incentive to use defense to create a win. At a near tie, Winners will get about 50% more points than losers plus a constant. (20:19 = 58+c:38). 3T means we should try for a tie vs. 2T which just says you will not be quite a screwed from a tie. High scoring matches will still cause giant boost, but a high scoring loss will not kill your qualification possibilities.
*****************************
Winners: W+L+C
Losers: 2L
Tie: 2T
Promotes high scoring from both sides. There is no incentive for Winners to score for their opponents as points are equal. There is also no disincentive for your opponents to score. Defense should be minimized for small values of C. In theory 6 machines scoring should have a higher total than 4 scoring and 2 defending. As C increases relative to scoring possibilities, the value of the win increases. As the value of a win increases, there is a higher incentive to use defense to create a win. At a near tie, Winners will get about the same as losers plus a constant. (20:19 = 39+c:38) High scoring matches will remain relatively balanced. C is really the only incentive to Win, but scoring lots is a huge incentive to seeding. having the "ideal" match is significantly less critical to seeding high. (this one is my favorite as tweaking a single constant C can give you the field behaviour you want to see).
************************************************

Using a scoring delta (W-L) in the denominator such as loser ss= W/(W-L) incentives a close match from the losers perspective, but also incentives defense from the winners perspective. There is a huge swing in points difference between loosing by 1 vs. loosing by 2. Any of those formulas are also difficult to understand.

Re: Am I the only one who LOVED the seeding system this year?
 

I disagree with you Joe (partially anyway).

There are many rules in sports (the safety in footbal, the infield fly rule in baseball, team orders in Formula 1, etc.) that are confusing the first time you see them. After you see it once and the announcer explains the situation, it's not very confusing any more. Now I agree that FIRST should keep the rules as obvious as possible without needing an explaination, there will always be rules that are in the background that the spectators don't know about (or need to know about) until the need to explain arises. Certain penalties are a good case in point: "why did team XYZ get a penalty when they crossed that bump into the defensive zone?" - that would also be confusing the first time a spectator saw it, but I won't advocate removing that rule from the books based on confusion alone.

Re: Am I the only one who LOVED the seeding system this year?
 

I like seeing defense in matches. To me, it makes them more exciting. The best teams can still maneuver around the defensive robots. Of course, this is a personal preference, and might be related to my region. I also like games which have multiple strategies involved in them and recognize the importance of scouting, but something about changing the objectives of match play because of the ranking system just doesn't feel right to me. I don't hate this year's ranking system, but I haven't fully bought into it either.

Re: Am I the only one who LOVED the seeding system this year?
 

Okay, how about this: let's take the AP poll, the Coaches poll, six computer rankings (drop the highest and lowest of course) and then add them all up. That should take all the controversy away...or maybe not I seem to recall some other sport having trouble with rankings.

Re: Am I the only one who LOVED the seeding system this year?
 

I like the way that a scoring delta (W-L) in the denominator highly incentivises close matches, as that seems to be what FIRST wants to do, while at the same time disincentivising scoring for your opponent. This would totally remove any incentive to play 6v0. However, its a more complex SS algorithm, and therefore harder to understand. I like IKE's third one. Its simple, and results in good payoffs for both. It doesnt, however, incentivise a close match.

Re: Am I the only one who LOVED the seeding system this year?
 

Are win-loss records stored anywere? Someone mentioned that more matches are played now so i was wondering if there is someway to compare standings using the old system and new system. Of course i am not that smart and just throwing that idea out there for someone to fill thier off-season time.:)

Re: Am I the only one who LOVED the seeding system this year?
 

FIRST used point based seeding prior to 2003, with W-L in eliminations. They went through a few different formulas, probably some of which were posted in here (I don't remember). People complained no matter what the forumla was (both about the formula and about the difference in play of elims and quals), so in 2003, FIRST used point based ranking in both the seeding and eliminations rounds. In the eliminations, you would play 2 matches and which ever alliance had the higher seeding score combined would move on. This had the consequence of the alliance that won the first match would not play to win the second match, but play to keep the scores low. If the second matches scores were low, they would advance based on the combined score. If they played all out in the 2nd match and lost, they might not advance.

After that, FIRST changed to W-L for both the seeding and elimination rounds, until this year.

Re: Am I the only one who LOVED the seeding system this year?
 

In 2007 and 2006, it was illegal to score for your opponents. It would not have been that difficult for referees to spot obvious attempts at scoring for the other side in Breakaway, but the GDC chose to leave this mechanic in the game.

I've been thinking about it, and here is my stab at a "Unified Theory of FRC Seeding Algorithms"...

The variables
WPoints = unpenalized points scored by the winning alliance
LPoints = unpenalized points scored by the losing alliance
WPenalties = winner's penalty points
LPenalties = loser's penalty points
WBonus = bonus seeding points for winning
TBonus = bonus seeding points for a tie
WSS = winner's seeding score
LSS = loser's seeding score
TSS = seeding score for both alliances in case of a tie (WP = LP)
C1, C2, C3, C4, C5, C6, PW, PL, PT = Constants

The formulas
Winner is defined as the alliance with the most points minus penalties.
WSS = C1*WPoints + C2*LPoints + WBonus - PW*WPenalties
LSS = C3*WPoints + C4*LPoints - PL*LPenalties
TSS = C5*WPoints + C6*LPoints + TBonus - PT*WPenalties

(for the last example, let WPenalties be YOUR penalties only)

Examples
Looking at the above equations, let:
C1 = 1
C2 = 2
C3 = 1
C4 = 0
C5 = 1
C6 = 2
PW = 1
PL = 0
PT = 1
WBonus = 5
TBonus = 0
This is the Breakaway scoring system.

Another example, W-L-T:
C1 = 0
C2 = 0
C3 = 0
C4 = 0
C5 = 0
C6 = 0
PW = 0
PL = 0
PT = 0
WBonus = 2
LBonus = 1

So clearly we can play with these constants to generate anything from score-doesn't-matter-as-long-as-you-win to full on coopertition and anything in between.

What is the optimal choice of the above constants? Most importantly, which should be zero and which should definitely be non-zero (regardless of game)?

C1 = 2
C2 = 1
C3 = 1
C4 = 2
C5 = 1
C6 = 2
PW = 1
PL = 1
PT = 1
WBonus = Dependent on score distribution; could be either 0 or non-zero
TBonus = 0

What would peoples' objections to the above consist of? Each team always benefits from additional scoring for itself AND for their opponent scoring regardless of the current score (both winner and loser get a contribution from both alliances' scores). There is still some incentive for defense, as given a finite number of game elements, each team is always better off scoring for its own alliance (so to parlay into Breakaway terms, punting the balls from your opponent's offensive zone down to yours would help your overall score). Each team is penalized for the infractions they commit (a flaw in Breakaway IMO, as the loser penalties disappear). High scoring and close matches are necessarily going to generate more seeding points than blowouts and low scoring affairs.

Re: Am I the only one who LOVED the seeding system this year?
 

A number of locations have Win/Loss, data but the easiest and most complete data for this type of calculation is probably Ed Law's Scouting Database. Another good option (although it doesn't have Champs data) would be the 1114 Scouting Database

Re: Am I the only one who LOVED the seeding system this year?
 

I would think that rewarding exceptional performance would be a good thing, but there seems to be a cultural stigma against having lopsided outcomes. I believe that encouraging someone to either do less than their best or otherwise artificially create a more aesthetically pleasing final score is wrong and has no place in a seeding system. Sure it can be humiliating to lose by a wide margin (been there), but it is also an exceptional motivator.

And while I'm on my soapbox... I really dislike using pre-penalty scores for anything. No one should benefit from a penalty. Then again a penalty should be reserved for behavior that you really want to inhibit and even then should be proportional to the damage incurred. I'm not much for calling penalties on robots that happened to be dragging chains as I saw multiple times this year. Sure it is against the letter of the rules but where is the advantage? Besides it seems to me that the team is already being penalized.

Re: Am I the only one who LOVED the seeding system this year?
 

If we're looking at a new formula, or open to new ways of ranking teams, why not go to a tiered system.

It could work... roughly... something like this.

All teams start out in tier 3, with randomly assigned partners and opponents.

The teams that win move up to tier 2. The teams that lose move down to tier 3. Teams that tie or don't play in the first round (due to the total number of teams not being divisible by 6) will have to be dealt with, possibly by coin toss, or by remaining in tier 3.

In following rounds teams move up... or down... depending on how they do, but always play against teams from within the same tier.

Ranking could take place in a number of ways... you could sort by tier, then subsort by w-l-t or some other measure, or you could make a win in a higher tier worth more points than a win in a lower tier and sort by points.

So yes, it would take a bit of thinking to work through the bugs and come up with an algorithm that works. I'm not really worried about the fine details at the moment, as they can - likely - be worked out.

The general idea, however, is that stronger teams will play with and against stronger teams, while weaker teams will play with and against weaker teams. After two or three rounds the top tier matches will begin to look like elimination round matches. In lower tier matches teams that wouldn't normally have a chance to effect the outcome of a match would get to have some control of their own destiny. Almost no one would go home undefeated, and almost no one would go home without a win. (Even more so than now.)

So take this as a brainstorming suggestion... not a concrete proposal... and consider the concept rather than the mechanics.

Does the concept have appeal?

Jason

Re: Am I the only one who LOVED the seeding system this year?
 

Jason,

That is very similar to a Swiss system of match pairing.

The biggest issue I see in implementing this type of system for FRC are that teams won't know when their next match is until at most an hour or so before it occurs (if you use rankings after the previous round to pair the next round, ignoring the current round). Another issue is that the system becomes much more complex when you add in the various desirable factors of FRC pairing such as reducing repeat pairing, balancing alliance colors, minimum match spacing, etc.

It's a cool idea, and perhaps one worth looking into, but it's certainly not without drawbacks.

Re: Am I the only one who LOVED the seeding system this year?
 

Thanks for the link to the Swiss System... I knew I'd seen tournament structures like this before, but didn't know the name.

I agree with the complexities, though...

Jason

Re: Am I the only one who LOVED the seeding system this year?
 

I think that you should get points for the points your robot scored. This is to make sure the good teams are at the top. That is what was bad in Atlanta for us. We had 2 bad matches where the other robot shut down on us and we were 8th.

Here is my plan:

W= W+2L+3P
L=W+3P

P is the points your robot scored.
W is winner's points.
L is loser's points.

Re: Am I the only one who LOVED the seeding system this year?
 

Overdrive is the only game in recent memory that was scored in a manner that I would consider it feasible for field personnel to accurately track this and even then it's questionable. You would also need extra volunteers to fill these roles.

Verdict: Probably not going to happen

Re: Am I the only one who LOVED the seeding system this year?
 

love the idea, but one big problem/potential complaint is what about defenders? By this defenders would receive less points. Now this could be a gamble teams would have to consider (play strong defense and hope to get picked for elims). But that's not entirely bad.

I liked this years system in that the top teams were at the top. In the past I didn't care about rank cause it didn't matter until late Friday. This year it was cool competing with other teams trying to get the top seed.

Re: Am I the only one who LOVED the seeding system this year?
 

That's pretty stupid for most teams, since that makes the assumption that any action taken by a robot that doesn't directly result in a scored point doesn't contribute to the alliance. Feeders were kinda pretty big this year.

Re: Am I the only one who LOVED the seeding system this year?
 

By your metric, we would have been much lower in Archimedes. Our specific strategy was to feed balls to the home zone. Feed until surplus, then snipe from the midfield. If surplus is gone, re-stock home zone. If we would have used your metric, we would have been fighting with our own alliance members so that we could be the ones to score the balls. Instead of 3v3 or 6v0, it would have turned into 1v1v1v1v1v1.:confused:

Re: Am I the only one who LOVED the seeding system this year?
 

Basically, it comes down to doing a regression fit on which of a set of score scenarios should be most heavily rewarded.

Consider:
A. Winning 20-0
B. Winning 15-5
C. Winning 11-9
D. Winning 10-0
E. Losing 0-20
F. Losing 5-15
G. Losing 9-11
H. Losing 0-10
I. Tie 10-10
J. Tie 5-5

Which of the above should be worth more than others?

Should winning 20-0 be worth more, less, or the same as winning 11-9? 15-5?

Should winning 10-0 be worth more, less, or the same as losing 9-11 or 0-20?

If I lose 0-10 or 0-20, should my score change at all?

If you can come up with a concrete answer to the above, you can generate a linear scoring function that will behave the way you want it to.

Re: Am I the only one who LOVED the seeding system this year?
 

Very interesting post Jared; really gets the brain churning...

When I go for an "offense is best" approach I end up with something like:

A=B=C>I>G>D>F=J>E=H=zero

but I'm finding that I have an innate bias to favor wins over ties over losses regardless of the score, so relationships like G>D are troubling. Also, I find that I could easily change my mind if the scoring mechanism were different than in Breakaway.

Alternatively, we could just seed everyone based on their least squares contribution like 1114 does in their database. Of course, that might be a bit difficult to explain to casual observers and there are linear algebra issues with small numbers of matches.

Re: Am I the only one who LOVED the seeding system this year?
 

Id say you might be the only one who liked it. :p

Re: Am I the only one who LOVED the seeding system this year?
 

I think that the idea of using pre-penalty points for seeding came from some strange cases that can come up when you allow one team's bad behavior (intended or accidental) to impact another's seeding score.

I can't think of a particularly onerous case in this year's formula (I mean more onerous than the fact that you had incentives to score for your opponents), but I suppose there were some cases they were trying a dis-incentivize.

It is possible that this may be a case where the cure is worse than the decease.

Joe J.

Re: Am I the only one who LOVED the seeding system this year?
 

I agree.

Consider the following matches that I grabbed from FRC-Spy. I don't really know the details behind these matches but I think they illustrate my point about penalties and the use of pre-penalty scores (and if not there are plenty out there that do):

In Archimedes Q-142, the Red alliance is penalized 14 points which turns a nominal 13-10 win with 38 seeding points into a 0-10 loss with 10 seeding points. Okay, that is a pretty big penalty, but it gets worse. The Blue alliance instead of getting 18 seeding points for a 10-0 win gets 41 seeding points (mostly due to the pre-penalty contribution). So the Red alliance gets penalized again with a huge difference in seeding points that I will submit aren't earned. I will further speculate that the penalty was most likely due to DOGMA penalties induced by a ball that missed the return sensor. If indeed this was the case then was a 56 point seeding swing an appropriate response?

The second example is a more typical case. The Red alliance wins 12-5 post-penalties but is effectively given a 2xPenalty bonus for Blue's mistakes which for all I know could have been as innocuous as having a piece of overhanging polycarbonate film or running up on a ball and continuing to move towards the goal. As Joe pointed out the bonus could be warranted if Blue (1) knew they were going to lose and (2) intentionally caused the penalties in order to reduce Red's seeding score. Personally, I think very highly of all the FIRST community teams and would like to think that this behavior is beneath our consideration (in the particular example I've used, the match would have effectively been 12-10 except for the penalties so what is the likelihood that Blue caused these penalties intentionally?).

Re: Am I the only one who LOVED the seeding system this year?
 

You're right regarding the DOGMA causing Red to win the match on Archimedes. We (2062 and the rest of the alliance) were overjoyed at the win but dissapointed how it resulted with seeding points.

Re: Am I the only one who LOVED the seeding system this year?
 

It was very likely either that or a double-counted ball in the goal (which happened in the last qualification match at the Boilermaker regional, with an even larger number of automatic penalty points). The other option is that someone "helpfully" tossed a ball past the alliance station wall, but nobody would have done that, right?

A few matches later, another mass of DOGMA penalties happened. People who were paying attention at the time decided it was because two balls went down the rails effectively in contact with one another and got counted as a single ball in the return sensor. That match was replayed.

I have no idea what you mean. Penalties against the losing alliance do not affect the winning alliance's seeding points, just as penalties against the winning alliance do not affect the losing alliance's seeding points.

Re: Am I the only one who LOVED the seeding system this year?
 

Alan, I'm not sure whether my interpretation of the effect, my bookkeeping, or a little of both is confusing so please forgive me as I start from the obvious. The manual says that the winner's seeding points will result from:

Since the coopertition bonus is based on the losing alliance's pre-penalized score, I am choosing to represent this (in the second example given) as:

RedSeedingPoints = RedFinalScore+2*(BlueFinalScore+Penalties)+5

Hence the effective 2*Penalties statement. I realize that strictly speaking there is a max(BluePrePenaltyScore-Penalties,0) operation that isn't quite captured in this interpretation.