pic: Sum of team numbers versus match number in Curie

[Kris Verdeyen]

loading photo ...

[artdutra04]

http://www.chiefdelphi.com/forums/sh...ad.php?t=66983

Joe Ross identified the problem, and found that the easiest solution would be to randomize the order of the team list before inserting it into the match scheduling algorithm.

[Uberbots]

Originally Posted by artdutra04

http://www.chiefdelphi.com/forums/sh...ad.php?t=66983

Joe Ross identified the problem, and found that the easiest solution would be to randomize the order of the team list before inserting it into the match scheduling algorithm.

yeah, but that doesnt adress the problem. What happens is that you are essentially paired with the same group of teams. if the data isnt randomized before the algorithm is run, then something like this chart would happen. you would still notice a pattern in the sequence if it were randomized before sorting.

I dont think there is any viable algorithm to fairly balance all the teams with adequate spacing... there are just too many constraints

[Vikesrock]

Originally Posted by Uberbots

yeah, but that doesnt adress the problem. What happens is that you are essentially paired with the same group of teams. if the data isnt randomized before the algorithm is run, then something like this chart would happen. you would still notice a pattern in the sequence if it were randomized before sorting.

I dont think there is any viable algorithm to fairly balance all the teams with adequate spacing... there are just too many constraints

Any algorithm that guarantees that teams are a certain number of matches apart will show such a grouping. The groupings will decrease and become less noticeable as the minimum match spacing becomes smaller.

The algorithm currently in use tries to optimize the number of different teams you play with and against with a specified minimum match spacing. Randomizing the teams before putting them into the algorithm would still cause the "pool" effect, but the teams you are pooled with would not be determined by number. If the teams you play with and against are different (algorithm) with no attention payed to team number (randomize before algorithm) there isn't a whole lot left to complain about.

[waialua359]

Originally Posted by Vikesrock

Any algorithm that guarantees that teams are a certain number of matches apart will show such a grouping. The groupings will decrease and become less noticeable as the minimum match spacing becomes smaller.

The algorithm currently in use tries to optimize the number of different teams you play with and against with a specified minimum match spacing. Randomizing the teams before putting them into the algorithm would still cause the "pool" effect, but the teams you are pooled with would not be determined by number. If the teams you play with and against are different (algorithm) with no attention payed to team number (randomize before algorithm) there isn't a whole lot left to complain about.

I hope it happens next year. Point well made.

[martin417]

I noticed right away that ~90% of our matches were with and against teams on the same row as us in the pits. The other 10% were with / against teams on the adjacent row. 0% were with teams on the first row. Since the pits are arranged by team number, it is obvious that team number plays a huge part in the alliance picking algorithm.

[Fireworks 234]

We came across the same thing. Our team played only 8 different teams with a team number below 1000. And 6 of the 8 were in our first 2 matches.

[Andy Baker]

There are many things frustrating about this. One that people may not think of is this: rookie or 1-2 year-old teams had minimal chance to play with the veteran teams. Part of the championships is playing with and against the best. If I were on a rookie team, I would relish the chance of playing in a match against teams who have been around for many many years. Sadly, this year's algorithm did not allow for much of this.

Case in point: in each of the TechnoKats' (team 45) matches, there was not a team number above 1000. This is not right.

Andy B.

[Kris Verdeyen]

Originally Posted by Andy Baker

in each of the TechnoKats' (team 45) matches, there was not a team number above 1000. This is not right.

IIRC, in one of the Technocat's alliances, there wasn't a team number over 67.

[Ty Tremblay]

You can have a random schedule without having a randomly timed schedule. The team numbers need to be randomized, not the schedule.

[Elgin Clock]

Just by looking down the lists of teams in the match schedule, you could see this pattern. 95% of teams in a block of say 5 matches were under team number 1000, and then 95% were over 1000 in the next 5 matches, and so on.

There were many matches where you (if your team number was 300 or less) were the youngest team on the alliance. I know on Archimedes specifically, Team 228 was the highest number in one match. Their team is ten years old this year! It was like a flashback to an old school match from New England circa: 1999 all over again! lol

Andy Baker just said it best I think. The rookies really never got to play with the older teams (and potentially learn from them), and the older teams really never got to play with the younger ones (and potentially help them in their FIRST career on the field by showing them the ropes).

That kinda saddens me. Helping the program grow isn't just getting new teams every year, it's also about equally pairing them up with some older teams so they can be trained/molded/advised by older teams on the field and learn what FIRST is exactly all about in that way.

[smurfgirl]

The teams in our matches at the CT Regional and on Archimedes were also skewed. At 1124, we were nearly always the lowest team number on the field. We mostly played with rookie and sophomore teams. We had really low qualifying points, because we were generally playing with/against teams who were inexperienced and could not put up a high score.

[Ross340]

Originally Posted by Kris Verdeyen

IIRC, in one of the Technocat's alliances, there wasn't a team number over 67.

We were on the other side of that one.

bad news bears. haha

[Rick TYler]

How many of you have read the paper describing the mathematics and reasoning algorithm of this year's new match scheduling software? You can read it here: http://www.idleloop.com/matchmaker/. For the record, this software was written (for free) by two mentors from team 1318, and was meant to address the criticisms from past years. In fact Tom and Cathy wrote it because they despised some of the "we seem to have the same partners in every match" problems. It seems to have fixed that. What I'm guessing might be missing in the algorithm is the assumption that "team numbers determine robot quality." The algorithm is trying to minimize the number of times a given team sees another specific team as an opponent or alliance partner, to make sure the red/blue alliance experience is close to 50-50, to smooth out match separation (to assure no back-to-back matches), and what Cathy and Tom call "round uniformity." Making sure the team numbers are smoothly distributed was not a factor in the algorithm.

If you have suggestions why not email Tom and Cathy at the address on their Website? Adding a factor for "balanced mix of rookies and older teams" would probably not be impossible, and going to someone who can fix the problem would probably be more productive than complaining on CD. For what it's worth, the current match scheduler looks to me like it produces much better results than older solutions. I also think someone ought to thank Cathy and Tom for their contribution of this software.

[Vikesrock]

The algorithm is not intended to be random, as has been mentioned many times before a random schedule would be very very bad.You may play 2 or more matches in a row then not play for a few hours.

The team number issue with this year's algorithm has already been discussed here:
http://www.chiefdelphi.com/forums/sh...ight=algorithm