View Full Version : FRCTop25.com- Submit Your Picks for Week 1!
Mike Starke
05-03-2011, 21:31
Hey all!
Well Week 1 Regionals are now over! There have been some excellent matches played this weekend! Hope you all had fun participating and watching the Regionals!
FRCTop25.com is now accepting submissions for your Top 25 teams in FRC through week one. Head on over to FRCTop25.com (http://www.FRCTop25.com) to make your picks! Please please please read the instructions and rules on the voting submission page (please remember you can't vote for teams who haven't played yet). You have until 9PM EST on Tuesday to cast your votes!
If you're unaware of what FRCTop25.com (http://www.FRCTop25.com) is, please refer to this thread! http://www.chiefdelphi.com/forums/showthread.php?t=92941
Thank you for all your support! Justin and I, as I'm sure the rest of you, look forward to seeing who you guys vote as the top robots through week one!
- Mike and Justin
Justin Montois
05-03-2011, 22:05
Woohoo! I cannot wait to start seeing the lists come in.
Just so there is no confusion, Each week will NOT be an isolated list of only that weeks teams. Once a team has competed, it can be included in your list from then on. IE in your week 2 list, teams that competed week 1 can still be included.
Ok, this is a feature request.
Filling out 25 boxes is tedious and I can't even think of 25 teams that I like. How about allowing users to enter just 5 or something like that.
rees2001
06-03-2011, 11:29
Ok, this is a feature request.
Filling out 25 boxes is tedious and I can't even think of 25 teams that I like. How about allowing users to enter just 5 or something like that.
Agreed, I would say to limit it to a top 10 list or 15 at most. I got through the first 10 and then didn't have enough info to make educatedish assessments on other teams.
Here's mine.
My ranking is based solely off the numbers.
Total Points Scored in Qualifying Matches.
Proves the FIRST ranking algorithm is seriously flawed
especially this year
for BAE team 40 was ranked 12th...
had our team not hit the estop button, we would have been ranked 11th.
(we capped 2, maybe 3 tubes all competition)
absolutely crazy
we would have been ranked ahead of a team with a 2 tube autonomous!!!?????
The following form submission was sent.
1: = 33
2: = 40
3: = 131
4: = 148
5: = 1519
6: = 1918
7: = 836
8: = 2137
9: = 138
10: = 176
11: = 16
12: = 69
13: = 78
14: = 302
15: = 201
16: = 126
17: = 58
18: = 2337
19: = 894
20: = 3074
21: = 118
22: = 494
23: = 811
24: = 175
25: = 67
Mentor/Student = Mentor
ChiefDelphi_Username = lineskier
Yeh. I don't get it. Not a single team from the NJ regional? I am losing faith that I will put a lot of weight in these results.
I am not judging your opinion, but rather highlighting the difficulty involved in developing the system.
Adam Freeman
06-03-2011, 11:52
25: = 67
Interesting a team that essentially sat still for 5 straight matches would even get a vote.
Its a great robot....when it moves.
Thank you.
Vikesrock
06-03-2011, 12:00
Yeh. I don't get it. Not a single team from the NJ regional? I am losing faith that I will put a lot of weight in these results.
I am not judging your opinion, but rather highlighting the difficulty involved in developing the system.
His vote was not based on opinion. He used statistics based on the final score of qualification matches to generate that list.
It's clear that the results of such a query at questionable at best. The 8th team on the list (the new one) did not make elims at Granite State and some of the great teams from New Jersey such as 2016 and 1676 didn't make the list at all.
I'm personally eager to see both the user and expert polls to see what everybody else thought of the robots that played this weekend.
Lil' Lavery
06-03-2011, 12:08
:-D eek...
that's one thing about this years game, it seems there are only a few powerhouses, and they tend to define the matches they participate in.
The consequence of this: when you look solely at the numbers, some teams sneak in there.
Just proves the need for me to speed up the development of my match by match analysis, to find a team's estimated contribution to a single match. That way point inflation doesn't occur like it did here :).
I was happy with many of the teams on the top 25 so I chose to stick with it, knowing there would be some mistakes. Also I wanted to avoid any regional bias I had.
one question... did you sit still for 5 qualifying matches or were they other types?
my numbers are based solely off qualifying matches.
By not including a single team from the New Jersey regional? :rolleyes:
Adam Freeman
06-03-2011, 12:33
one question... did you sit still for 5 qualifying matches or were they other types?
my numbers are based solely off qualifying matches.
Yes... 5 of our 6 Saturday morning matches resulted in limited or no robot operation. :(
When determining your top 25, I would suggest investigating the elimination matches also.
Yes... 5 of our 6 Saturday morning matches resulted in limited or no robot operation. :(
When determining your top 25, I would suggest investigating the elimination matches also.
good call!
its interesting though, based solely off the qualifying matches, the top 9 all were finalists.
but they were truly in a league of their own.
my stat algorithms weren't intended for a top 25 kind of use. i just wanted to see how close they got to predicting finalists. obviously needs a little bit of work...
remulasce
06-03-2011, 12:43
Woohoo! I cannot wait to start seeing the lists come in.
Just so there is no confusion, Each week will NOT be an isolated list of only that weeks teams. Once a team has competed, it can be included in your list from then on. IE in your week 2 list, teams that competed week 1 can still be included.
I think this unfairly benefits early week competitors. Take, for example, a pretty good robot in week one. They obviously are better than the rest in week one, so are rated highly. In week two, they don't compete, but they still are better than average and thus are rated highly. By week six, they have gotten five weeks of good reviews, when another team shows up with a demigod as a robot, and their drivers have the force. They are ranked really highly, and the week one teams is ranked a lot lower now that people have seen the demigod, but the week one team has been ranked highly by more people, so in total rankings, they win.
TLDR; you can't keep the voting window open longer for some teams than other teams, and you can't vote until you see all the teams.
teamnumber opoints/nummatches
217 84.0000
2056 79.3000
33 65.5000
1676 65.1111
1126 64.9000
148 63.6667
40 60.6000
131 58.0000
2016 58.0000
1218 56.6667
191 56.5000
1519 55.2000
156 54.1000
16 53.3333
836 52.8000
340 51.2000
138 50.2000
176 49.4000
2234 49.1111
3015 48.0000
69 47.9000
118 47.1111
78 47.1000
1918 45.9167
1089 45.5556
1923 45.2222
229 45.0000
1559 44.7000
126 44.0000
25 43.7778
WOW! 7 teams from NJ.
I agree about including elim matches. High seeds will get favor, but I guess it would be deserved. This would help out good teams in a regional with fewer good teams to support them in qualifying matches.
Mike Starke
06-03-2011, 13:03
These are all great concerns! Thank you so much for all of your feedback! Please remember, we have a secret list of "experts" who will also be voting. So you can decide which list you think will be accurate; between the public poll and the expert poll. Our experts know that they need to go through each regional and spend a lot of time sorting through data and teams. On some of their submissions you will also see why they voted why they did. It's in my opinion that regional biases (the reason why no NJ teams made that one list) will get balanced out by the number of voters that we will have. Some one may heavily weight Alamo, while someone will heavily weight FLR. So i think it will be well rounded by the end. We have over 20 submissions so far, with just being open for about a day. We'll see how it goes.
And please remember, this is just for fun. :)
WOW! 7 teams from NJ.
I agree about including elim matches. High seeds will get favor, but I guess it would be deserved. This would help out good teams in a regional with fewer good teams to support them in qualifying matches.
you see major point inflation when you include the finals.
not sure which is better for a top 25...
theres one thing thats still for sure, 217 is a monster
#1 overall
#1 minibot
#2 robot (behind 148 )
this is including elims...
217 91.9375
2056 86.3750
148 85.1429
1676 79.9286
16 77.3571
2016 77.0000
33 73.6111
1126 72.7500
1519 68.2143
303 68.1538
40 67.3529
176 67.0000
340 66.4667
131 66.1765
25 61.5000
1923 61.3000
1518 60.8824
175 60.2000
2137 58.2778
191 58.0833
1918 56.3333
229 54.7692
118 54.6154
3074 53.9231
156 53.9000
138 53.8889
1218 53.3333
2228 53.2000
811 52.8462
69 52.3636
Justin Montois
06-03-2011, 13:41
I'm going to echo what Mike said in that this is just for fun! Are some people going to vote based on little to no knowledge? Yes. Are people going to disagree with the eventual week 1 top 25? Yes. Just try to have fun with this and if you want to disagree with someone's list then feel free to submit your own list.
We fully expected this and will enjoy reading the discussion.
Also, if you feel as though you don't have enough knowledge to fill out 25 teams out of the over 200 teams that competed week 1 then your more then welcome to either do some more research to become more informed (looking at OPR would be a great place to start) or you could hold off and vote next week.
Thank you all for your feedback and participation.
you see major point inflation when you include the finals.
not sure which is better for a top 25...
theres one thing thats still for sure, 217 is a monster
this is including elims...
217 91.9375
2056 86.3750
148 85.1429
1676 79.9286
16 77.3571
2016 77.0000
33 73.6111
1126 72.7500
1519 68.2143
303 68.1538
40 67.3529
176 67.0000
340 66.4667
131 66.1765
25 61.5000
1923 61.3000
1518 60.8824
175 60.2000
2137 58.2778
191 58.0833
1918 56.3333
229 54.7692
118 54.6154
3074 53.9231
156 53.9000
138 53.8889
1218 53.3333
2228 53.2000
811 52.8462
69 52.3636
Yes. There is an issue with including elims. The second alliance pick. As an example, 303 was a good enough scorer, but we (1676/2016/303) used them as a defender and they jumped from off the list to 10th on the list due to the alliance score in elims. Not really in the spirit of what you are trying to accomplish with this list.
There is value in what you are doing with this data. Thank you for sharing. I am not the kind of guy to do the full research on 25+ teams to submit a fully proper entry, but using your analysis as a starting point I may feel confident enough to submit an entry. Maybe after I see data for week 2 as well.
Chris is me
06-03-2011, 14:07
Interesting a team that essentially sat still for 5 straight matches would even get a vote.
Its a great robot....when it moves.
Thank you.
Well, I only saw the eliminations at Kettering...
Yes. There is an issue with including elims. The second alliance pick. As an example, 303 was a good enough scorer, but we (1676/2016/303) used them as a defender and they jumped from off the list to 10th on the list due to the alliance score in elims. Not really in the spirit of what you are trying to accomplish with this list.
Exactly :).
Its just nice to see the evidence that elim rounds should not be used when evaluating teams stats as they are not a reflection of the team but the alliance. Also once I start calculating a teams contribution in each match, the best teams will really rise to the top, more so than they already have.
Again I only view my #1 pick as a ranked pick. Simply because 217 killed it in week 1, I can comfortably say they were the best. Everyone else should be looked at as a top 25 team, but not as #6 or #9...
It will be interesting to see how teams' second regionals go. I'm personally looking out for team 40 and 33, as they are the only teams i know of going for the double autonomous, with a reasonable chance of doing it.
Chris is me
06-03-2011, 14:11
I think using purely statistics to rank teams is kind of defeating the whole purpose of the website and polling. Just saying...
I think using purely statistics to rank teams is kind of defeating the whole purpose of the website and polling. Just saying...
Actually I agree...
But I am only one person, and I know people's opinions of robots will obviously rise to the top. I think numbers are a good place to start, and then based on that people should actually watch match videos and such. Unfortunately right now, most people can only vote based off the one or 2 regionals they watched in week 1. I'm simply trying to provide a basis for powerhouses that should be looked at :).
Exactly :).
Its just nice to see the evidence that elim rounds should not be used when evaluating teams stats as they are not a reflection of the team but the alliance. Also once I start calculating a teams contribution in each match, the best teams will really rise to the top, more so than they already have.
Again I only view my #1 pick as a ranked pick. Simply because 217 killed it in week 1, I can comfortably say they were the best. Everyone else should be looked at as a top 25 team, but not as #6 or #9...
I think there is value in the elims data to see what a good team can really do when put with other good teams, but not for the round 2 picks if that is possible. There was inflation, but also compression of the spread. That shows that the separation of the better teams might not be so much when playing at higher levels and team play is key.
I think there is value in the elims data to see what a good team can really do when put with other good teams, but not for the round 2 picks if that is possible. There was inflation, but also compression of the spread. That shows that the separation of the better teams might not be so much when playing at higher levels and team play is key.
Unfortunately, for me there's no easy way to do this. As a twitter post comes in, I update a running sum for a given team. Unfortunately the alliance selection is not posted to twitter. Maybe something they could add to the twitter feed next year. I'm still trying to get autonomous added to it too.
If they ordered the teams on the twitter feed based on alliance selections, then I could do this, however that is not the case this year.
nikeairmancurry
06-03-2011, 15:06
It will be interesting to see how teams' second regionals go. I'm personally looking out for team 40 and 33, as they are the only teams i know of going for the double autonomous, with a reasonable chance of doing it.
There are two other teams that pulled of a double autonomous... 326 and 148... and I wouldn't be surpised to see 1114, 111, 330 or 254 having one either..
There are two other teams that pulled of a double autonomous... 326 and 148... and I wouldn't be surpised to see 1114, 111, 330 or 254 having one either..
Awesome! Yeah I don't think the limit factor on a double autonomous is programming, its actually fairly simple to code it up given the right sensors.
The limiting factor is having a very well designed robot. This is why I've always shot for the design requirement "Design for autonomous"... because if it can meet the goals of autonomous, its probably simple enough for the drivers to use quite well.
It's awesome to hear that so many of top 25 teams are getting there based on focusing on building great tube hangers, rather than great minibots. Also it seems teams who chose not to pick up from the field (my team included) made a huge mistake. Is anyone aware of a top 25 team that can't pick up off the floor?
nikeairmancurry
06-03-2011, 20:39
Awesome! Yeah I don't think the limit factor on a double autonomous is programming, its actually fairly simple to code it up given the right sensors.
The limiting factor is having a very well designed robot. This is why I've always shot for the design requirement "Design for autonomous"... because if it can meet the goals of autonomous, its probably simple enough for the drivers to use quite well.
It's awesome to hear that so many of top 25 teams are getting there based on focusing on building great tube hangers, rather than great minibots. Also it seems teams who chose not to pick up from the field (my team included) made a huge mistake. Is anyone aware of a top 25 team that can't pick up off the floor?
From the week one regionals, if you couldn't pick up from the floor you were probably a runner and a late third pick during selection... Plus with so many tubes on the ground, I don't see you wouldn't... If I was on your team, I would consider changing to one if possible...
Awesome! Yeah I don't think the limit factor on a double autonomous is programming, its actually fairly simple to code it up given the right sensors.
The limiting factor is having a very well designed robot. This is why I've always shot for the design requirement "Design for autonomous"... because if it can meet the goals of autonomous, its probably simple enough for the drivers to use quite well.
It's awesome to hear that so many of top 25 teams are getting there based on focusing on building great tube hangers, rather than great minibots. Also it seems teams who chose not to pick up from the field (my team included) made a huge mistake. Is anyone aware of a top 25 team that can't pick up off the floor?
From what I saw, HYPER 69 was unable to pick up from the floor or preferred to obtain tubes from the feeder station.
Buzz Robotics 175 is able to pick up from the floor but prefers to be fed as well. We did not use the "flipper" at all during the competition.
We found it more effective to keep the floor clear of tubes in order to negate the opposing alliance from stealing the tubes we threw out.
From the week one regionals, if you couldn't pick up from the floor you were probably a runner and a late third pick during selection... Plus with so many tubes on the ground, I don't see you wouldn't... If I was on your team, I would consider changing to one if possible...
Yeah, our seasons over. Honestly one of our many poor design decisions. We live and learn. For me it was a great year to test my autonomous structure and made a lot of forward progress. I also now have a long list of requirements for my code structure.
Now its time for me to work on my ranking algorithm, its okay but could be a lot better :)
From what I saw, HYPER 69 was unable to pick up from the floor or preferred to obtain tubes from the feeder station.
Buzz Robotics 175 is able to pick up from the floor but prefers to be fed as well. We did not use the "flipper" at all during the competition.
We found it more effective to keep the floor clear of tubes in order to negate the opposing alliance from stealing the tubes we threw out.
yeah I'm thinking this was a big difference between BAE and the other competitions. Unfortunately i think the teams were weaker at BAE this year than some of the other regionals. Also the BAE strategy seemed a lot different from the other locations too.
Just for example, the best BAE teams averaged around 60, where as 217 averaged 85 points per qualifying match and 2056 almost 80. Come the championship, if you want to play with the big boys, you'll have to be able to pick up off the floor, and have autonomous too.
nikeairmancurry
06-03-2011, 21:12
yeah I'm thinking this was a big difference between BAE and the other competitions. Unfortunately i think the teams were weaker at BAE this year than some of the other regionals. Also the BAE strategy seemed a lot different from the other locations too.
Just for example, the best BAE teams averaged around 60, where as 217 averaged 85 points per qualifying match and 2056 almost 80. Come the championship, if you want to play with the big boys, you'll have to be able to pick up off the floor, and have autonomous too.
Your season is never over... I would continue to improve that robot and teach those on your team... Learning is what comes of a regional, plus it doesnt hurt to have something working for an offseason event... Use that robot as a great tool for the 2012 season...
The average may have differed greatly but there are different ways this data may be perceived. It could mean that the teams were "weaker" or that they were better balanced with respect to one another.
Like you stated, the strategies at the regionals varied and the presence of tubes from both alliances in a central location compared to that of playing conservatively will result in different results.
There is a thread about the highest score that refers to 148 and their partners scoring 135, but the opposing alliance with only a mere 8. Compare that to the first finals match at GSR with 117 over 82 and you see more of a balance. What were some of the finals scores at FLR,Alamo, NJ, and Kettering? I'd like to see some of the results.
nikeairmancurry
06-03-2011, 21:20
Quarter Final 1, Match 1 (1 vs 8) at kettering... ended 114-79
http://www.youtube.com/watch?v=iOzqNaCscgc&feature=related
probably the best two alliances in that elims...
The average may have differed greatly but there are different ways this data may be perceived. It could mean that the teams were "weaker" or that they were better balanced with respect to one another.
Like you stated, the strategies at the regionals varied and the presence of tubes from both alliances in a central location compared to that of playing conservatively will result in different results.
There is a thread about the highest score that refers to 148 and their partners scoring 135, but the opposing alliance with only a mere 8. Compare that to the first finals match at GSR with 117 over 82 and you see more of a balance. What were some of the finals scores at FLR,Alamo, NJ, and Kettering? I'd like to see some of the results.
the top NJ score:
red score:124 blue score:22 red :1860 25 1923 blue: 1089 1647 102 red mini:50 blue mini: 0
top alamo:
red score:135 bluescore: 37 red: 3481 16 148 blue: 57 2848 245 red mini:50 blue mini:25
kettering
red score:114 blue score:79 red: 2137 33 1 blue: 326 67 3322 red mini:50 blue mini:15
Tristan Lall
06-03-2011, 22:24
Top teams, or top robots? You mention both in the opening post.
the top NJ score:
red score:124 blue score:22 red :1860 25 1923 blue: 1089 1647 102 red mini:50 blue mini: 0
The other top NJ score
red score:124 blue score:36 red :2016 1676 303 blue: 1302 2180 816
The Lucas
07-03-2011, 17:13
Since everyone is talking about avg qualification score, I figured I would post Top 30 OPR (http://www.chiefdelphi.com/forums/showthread.php?t=75272). I made this using version 13 on each regional and compiling it using Excel (while Bongle is working on a new version for this year).
Rank Team OPR Event
1 33 56.7075 GG
2 148 56.5229 STX
3 217 54.4144 ROC
4 2056 52.7503 ROC
5 1676 46.3775 NJ
6 2016 45.5644 NJ
7 1218 45.5229 NJ
8 1126 43.2325 ROC
9 1519 40.1307 NH
10 131 38.6893 NH
11 16 38.4332 STX
12 40 34.7814 NH
13 836 34.2788 NH
14 118 33.748 STX
15 1918 32.5715 GT
16 25 32.5498 NJ
17 229 32.3928 ROC
18 191 31.3583 ROC
19 1429 30.0761 STX
20 2848 29.419 STX
21 1089 28.689 NJ
22 2587 28.6853 STX
23 2137 28.6095 GG
24 1923 28.4289 NJ
25 245 28.249 STX
26 340 27.9883 ROC
27 69 27.961 NH
28 138 25.7975 NH
29 3074 25.7734 NH
30 2590 25.4736 NJ
OPR seems to be a good stat this year. Even though it only looks at qualification data, the team with the highest OPR won every event. Even this even applies to NH, where 1519 had the highest OPR, was picked 3rd overall and won. Keep in mind that OPR doesn't have any way to connect between regionals, since no inter-regional matches are played.
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.