[FRCTop25.com] Final Rankings Show!

[JohnSchneider]

Originally Posted by AdamHeard

In 2011 a coaches poll was run side by side with the public poll. I forget who all the coaches were, but this concept worked and both top 25's were displayed.

It'd also be awesome to implement a pure statistical top25 w/ FIM style points that are standardized to cover different # of matches played.

If a public, coach, and statistical ranking were done, a combined ranking of the three could also be done.

In CFB they just take the ranking from the three systems and average it then use that as your rank.

With the coaches poll in CFB they try to get a wide sampling of coaches from different parts of the country, different skill levels, different play styles etc to get a good sample. This is probably best for FRC too because a lot of coaches don't have enough time to watch every team in every region, but they will tend to chart their own region well and you'll get a relative comparison with more well known teams

The data portion seems pretty easy to do with the provided twitter data and someone entering awards.

[Patrick Flynn]

Originally Posted by AdamHeard

I did vote... In the coaches poll... and that was cancelled.
I love the concept of the show, my fault is with the voting process. Even if everyone informed votes, they will be drowned out tenfold by the rest of chiefdelphi. This was very evident in the 2011 rankings comparing the public vs. coaches poll.

You didn't vote this week which in turn gives you no right to complain about how the votes turned out this week. That is like complaining that we picked the wrong president but you didn't vote. If you give up your right/ ability to vote you also give up your right to complain about it.

The problem with the coaches polls are that these "experts" can't be experts every week some weeks their competing other weeks their working at their own jobs. Its just as hard for them to see all the teams compete as it is anyone else. Many times the coaches polls were skewed based on what regional they were at and what teams they saw play.

Originally Posted by connor.worley

Somehow, I don't see this making the rankings more insightful. For the same reasons we hold Looking Forward posts and the like above normal chatter.

This wouldn't make them more insight full never has the claim been made that these top 25 should top your pick list. Its just a voter generated list of some teams that are playing this years game really really good. If you have some time to look at these teams we could all learn something from them.

More voters would help to balance out the regional bias. If we currently have 70% of the votes coming for the east coast, more overall voters would result in more people from the west coast or any other region. This would help to remove these "favorite" regions. Everyone is mentioning these "regional bias" but most people are failing to mention that Texas has the second most teams in the Top 25, and there are no NE teams.

Ranking teams based on performance numbers is easy to do, it takes some work to general all the necessary data but after that all you have to do is sort the teams. The point of the show is for fun, its just a fun ranking that isn't just a set of numbers read off a spreadsheet. Sure we could use OPR but i think everyone would agree that if we had done this last year the Top25 would have been really skewed.

[Justin Montois]

Originally Posted by Andrew Schreiber

Tell ya what, let's make that happen.

Mike/Justin, any chance you could contact the folks you had doing the voting in 2011 and collect lists from them? (or pass the list my way and I'll do it)

I'll have the TwentyFour guys work up a FiM style point list, an OPR, and a CCWM list. (Yes Drost, that means you too)

Let's see how they all come out, I'm curious.

The reason why we stopped this was because the "coaches" simply stopped participating. By week 3 we were getting nearly 100 voters for the public poll and 3 -5 for the "coaches" poll. We started with over a dozen coaches. Which one was a better representation of the Top 25 teams according to the FIRST community? The public poll. Simple as that. And that's the model we're going to continue using.

I'm fine with people not taking the rankings seriously. Mike already stated the reason why we do the show is to bring the FIRST community together every week throughout the competition season. The rankings are secondary and merely give us a format for the show.

The BCS in college football is a computer generated model taking into account a number of factors. There are always people who complain about it's inaccuracies. People complain about the coaches poll in college sports. People complain about the media poll in college sports.

Anytime you try and "rank" anything, people will complain and try to knock it's credibility just because they have a different opinion. It is IMPOSSIBLE to rank over 2000 teams with a high degree of accuracy. That's why we have always claimed it was for fun and not to take it too seriously. All we ask is that you vote. If you don't want to vote that's fine, we'll have fun doing the show with or without you.

[Kevin Leonard]

It really doesn't matter what the rankings are- they could be ranking cereals for all I care (Oh wait they did that ). As long as they're bringing the FIRST community together, what the rankings are don't really matter. That being said, the rankings do help to bring to viewer's attentions some great teams and robots that don't normally get much recognition- like 4451 and 2791 who each had stellar performances at their regionals, but didn't carry much notice like the big names did.

[Ivan Malik]

Let me preface this with you guys at FRCTop25 do an amazing job. I have voted in the past and stopped because I have a huge bias towards Michigan. I see the show as an opportunity for lesser known teams from the area to get some much needed PR on the world stage, and therefore rank the big ones (like 1114) low. I stopped when I learned how few voters there were and didnt want my agenda to influence things that much.

Personally I think the reason you are seeing so many people knocking the voting is because the show is currently the most visual, highest audience show about FIRST around. You guys are leading the charge in a whole new avenue for FIRST; many people don't want that to be deluded by "inaccurate" rankings. Is there anything wrong with what you guys currently do, no, but what the majority of people's perception is the best teams will make the list. You are being viewed as a some-what official source, like looking forward is, for who is good. No matter how much you guys say it is just for fun, people have built it up so much in their minds that the show has taken on a whole new meaning.

IMHO the best course of action is to show/say how many voters there were that week and where the majority of voters are from to get rid of the regional bias (whether it exists or not) that people see. This will also show viewers that their votes count.

Your audience is a huge group of nerds, they like numbers and statistics. As the show grows comparing the numbers and the votes could be a nice feature, but should never be the cornerstone. Way to dry and one dimensional.

You guys are providing an amazing service for the FRC community in an area that is in its infancy. If you arnt recognized for being responsible for putting entertainment value on a primarily education based competition, then it's a criminal offense.

[JohnSchneider]

Originally Posted by Justin Montois

The BCS in college football is a computer generated model taking into account a number of factors. There are always people who complain about it's inaccuracies. People complain about the coaches poll in college sports. People complain about the media poll in college sports.

The BCS is only bad when either you aren't in the top 25 or your main rival is

[Andrew Schreiber]

Originally Posted by Justin Montois

Anytime you try and "rank" anything, people will complain and try to knock it's credibility just because they have a different opinion. It is IMPOSSIBLE to rank over 2000 teams with a high degree of accuracy. That's why we have always claimed it was for fun and not to take it too seriously. All we ask is that you vote. If you don't want to vote that's fine, we'll have fun doing the show with or without you.

You mistook my rant about democracy in general for saying your rankings were crap. Your rankings tell us which robots captured public imagination, it shows which robots are inspiring people. This is pretty cool actually. What I'd like to know is what inspires people the same as a group of seasoned veterans are inspired by? Is it limited to teams that have had success on the field? Is it biased based on region? Week played? Event? What makes them inspirational?

Our job is to inspire. My goal is to try to determine what makes these robots inspirational.

[TheMadCADer]

Originally Posted by ErvinI

For reference's sake, here's a comparison between their Max OPR Rank and their FRCTop25 rank:

Top 25---- Max OPR Rank (Average OPR Rank)

25. 1334--- 22 (50)
24. 2169--- 19 (8)
23. 624--- 7 (24)
22. 1806--- 24 (12)
21. 11---- 46 (21)
20. 1718--- 45 (39)
19. 2590--- 9 (18)
18. 341--- 284 (266)
17. 1310--- 65 (37)
16. 111--- 21 (29)
15. 868--- 10 (6)
14. 1477--- 11 (49)
13. 359--- 18 (16)
12. 148--- 20 (46)
11. 67--- 15 (32)
10. 610--- 33 (38)
9. 1717--- 6 (17)
8. 33--- 16 (10)
7. 469--- 3 (4)
6. 987--- 4 (3)
5. 118--- 5 (5)
4. 254--- 13 (23)
3. 1986--- 8 (7)
2. 2056--- 2 (1)
1. 1114--- 1 (2)

Max OPR reference: http://www.chiefdelphi.com/forums/sh...7&postcount=65

Additionally, here's pretty much the same thing, but with CCWM instead of OPR.

Top 25----    Max CCWM Rank (Average CCWM Rank)                  

 25. 1334---       39  (100)	
 24. 2169---       32  (12)
 23. 624---        7   (19)
 22. 1806---       18  (8)		
 21. 11----        33  (26)
 20. 1718---       44  (36)
 19. 2590---       41  (38)
 18. 341---        265 (254)
 17. 1310---       344 (337)		
 16. 111---        20  (32)
 15. 868---        34  (11)
 14. 1477---       6   (33)
 13. 359---        8   (13)
 12. 148---        40  (60)
 11. 67---         26  (35)
 10. 610---        27  (23)
 9.  1717---       14  (22)
 8.  33---         17  (25)
 7.  469---        9   (5)
 6.  987---        4   (6)
 5.  118---        5   (7)
 4.  254---        66  (39)
 3.  1986--        3   (4)
 2.  2056---       1   (1)
 1.  1114---       2   (2)

Source.

Also, I think there is an error in the data for 341 and 1310. Those numbers just don't make sense (average > max).

[EricDrost]

Originally Posted by Andrew Schreiber

Your rankings tell us which robots captured public imagination, it shows which robots are inspiring people. .

Way to steal exactly what I told you two hours ago.