Random Pairings???

[DragnButt]

Can anyone explain to me how in a field of 69 teams (at ypsilanti), teams ended up playing with or against the same teams in four out of six of the qualifying matches? I thought the qualifying rounds were supposed to be randomly matched up, but after this weekend... that's obviously not correct.

I found that it didn't give the teams a good opportunity to showcase their robot if the teams they're paired with didn't complement each other.

[Elgin Clock]

Random... HA!!! Think again... I think FIRST needs to make a new number generator for next year's competition and then have someone go over it and look for patterns that Scott is going to mention in the next post here.

[Scott Shaw 237]

I totally agree with you Elgin. At the New York City Regional we, (237) were against team 19 four out of 11 times and with the same alliance 3 times! If it was random, then something really big went wrong. (Oh yeah, this is out of 30 teams)

[157#1Driver]

I dont know about anyone else, but once you do something, dont you check it to see if it actually worked? If I saw that someone was paired with another team 3 or 4 times our of 6-7 matches, I would def say there's something wrong with the program

[srawls]

http://www.chiefdelphi.com/forums/sh...threadid=19012

[Koci]

The reason why it is not completely random is because otherwise the matches would not be spaced out evenly for teams. FIRST does not want a team to EVER have back to back matches, or something similar to that. They try their best to make it as random as possible, but they have to keep other considerations in mind too. They're certainly not "out to get you," like some people seem to think.

[sevisehda]

Read the previous posts to understand the problem. For a simple review of random.


Just because theres a 50% chance of a coin landing heads doesn't mean if you flip it twice it will be heads once and tails once. There will always be a certain chance of something happening. It's actually more probobable you'd play with a bot twice than not playing any bot twice.

[Elgin Clock]

Originally posted by Koci
The reason why it is not completely random is because otherwise the matches would not be spaced out evenly for teams. FIRST does not want a team to EVER have back to back matches, or something similar to that. They try their best to make it as random as possible, but they have to keep other considerations in mind too. They're certainly not "out to get you," like some people seem to think.

I understand your point, but I just don't remember it being this non-random in previous years.. I don't know maybe it was and I just didn't notice....

[Gadget470]

It's definitly not random, by almost any means. The algorithm needs a fixin'

Consider our matches this weekend..

First match: w/ 313 vs 1017 & 857
Second match: w/ 5 vs 184 & 292
Third match: w/ 240 vs 5 & 27
Fourth Match: w/ 247 vs 240 & 440
Fifth Match: w/ 5 vs 184 & 815
Sixth Match: w/ 240 vs 5 & 1017


So for 6 matches, with potential to see 18 new teams, we saw 11. 7 Repeat appearences.

Our sixth match was almost a pure repeat of our third match (less a swap of 27 for 1017.. who we had already been against)

Also.. I assume coincidence.. in our fourth match, our opponent 440 was directly next to us in the pits, and the other opponent 240 was next to our alliance 247.

2 of the teams we saw went on to elims. My team and I were not happy with how closely related all our matches were. Every match we had someone that was a repeat in a later or earlier match. We were with 5 twice and against them twice also.. they are a great bunch of people but come on now.. 4 matches involving 470 and 5? That's not anywhere near "random"

</rant>
</tired personal views>

Team 226 was against 111 once, and with them twice. Both times we were with them we were against 469. And we were with 469 twice to, and both times we were with 469 we were against 1015. And this was at Great Lakes with 68 teams. Oh well, It worked out good for us because we won (Alliance of 302, 67, and 226) and are now qualified for nats.

[Judy]

My team also found the pairings kinda odd. I found a lot of the times that teams you had paired with, you then went against, and it seemed that almost, if not, everyone had a team they were with more than once. I also know that for my team atleast, we were paired with a lot of the higher numbered rookie teams. We would have liked to try out competing with lower numbered teams as well, because a lot of them hadden't seen us in action when it came for the selections. *sigh* aw well, hopefully the Candian Regional will be a lil more random. Good Luck everyone!!
~Am (team 783, the ONLY Canadian team at the GLR!)

[Brandon Martus]

http://www.chiefdelphi.com/forums/sh...threadid=19012