Chief Delphi - Some statistics on year-to-year consistency, 2005-2007

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)

- General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)

- - Some statistics on year-to-year consistency, 2005-2007 (http://www.chiefdelphi.com/forums/showthread.php?t=56631)

Re: Some statistics on year-to-year consistency, 2005-2007

Quote:

Originally Posted by Joe Ross (Post 613426)

I'm somewhat surprised that Regional Winner #4 is ranked last, unless one of my assumptions is wrong.

For a team to be #4 on the alliance, they would have to be a replacement team, the highest seed to not be picked. The lowest this team could be would be 25th, if the top 8 seeds all choose within the top 24. (Typically it seems to be more in the range of 16 or even lower). If they have a performance of .7, then 70% of teams perform better then them. 25/.7=35 for a typical regional where a replacement happens. Since almost all regionals are larger then 35 and replacement teams are almost always ranked better then 25th, that number doesn't seem right.

Quote:

What's regional winner #4 in that graph? It looks like it's ranked dead last.

The team in question (there was only one #4 regional winner) was team 1216, who came 20/46 (0.43) at one regional (presumably the one they went #4 regional winner at) and 34/34 at another (1.0). So their math ends up being an average placing of 0.71, which is why the regional winner #4 is so low. Since there is only one sample point, it is artifically low and I probably shouldn't have included it in the list.

All the other awards except for regional finalist #4 have at least 24 data points, and in the case of the judges award, as many as 52.

Being that I'm bored now, I'm going to update it to include 2005 and 2006, then I'll see about doing some year-to-year relationships.

Re: Some statistics on year-to-year consistency, 2005-2007

Quote:

Originally Posted by Bongle (Post 613560)

Oops, I forgot that a team's rating came from multiple regionals.

Midwest and Peachtree had a 4th champion as well (1850 and 1848)

Re: Some statistics on year-to-year consistency, 2005-2007

Quote:

Originally Posted by Joe Ross (Post 613570)

Oops, I forgot that a team's rating came from multiple regionals.

Midwest and Peachtree had a 4th champion as well (1850 and 1848)

My team-data algorithm ignores teams after 1705 for now, because it was initially made just to do year-to-year comparisons, and there were no teams after 1705 in 2005. So anything involving team seeding performance only uses early teams.

Anyway, I did the award-count ranking, and the result kinda surprised me. 1305 is the king of the awards from 2005-2007, with 15.

Anyway, here's the list of the top 10 award winners (THAT I HAVE DATA FOR, the archived 2005 data from FIRST is spotty*) from 2005-2007:

Code:

1305        15

48        13

111        13

375        13

494        13

1114        13

118        12

188        12

71        11

103        11

1305's achievements are as follows:
2 x 2005 Regional Winner #2
2 x 2005 Regional Engineering Inspiration Award
2 x 2005 Motorola Quality Award
1 x 2005 Underwriters Laboratory Industrial Safety Award
2 x 2006 Kleiner Perkins Caufield & Byers Entrepreneurship Award
1 x 2006 Regional Finalist #1
1 x 2006 Regional Engineering Inspiration Award
1 x 2007 Website award
1 x 2007 Regional Finalist #1
1 x 2007 Regional Finalist #2
1 x 2007 Regional Chairman's Award

*An example of the spottiness: I have 961 awards for 2006, 1082 for 2007, and just 392 for 2005. I should probably ignore 2005 because it completely ignores most teams that played in the first 2 weeks of march, where almost 100% of those regionals are missing award lists. If I ignore 2005, then the top awards-given list looks like:

Code:

111        11

375        11

494        11

234        10

1114        10

103        9

114        9

188        9

469        9

1714        9

1305 drops to a still-impressive 21st since they lose their very plentiful 2005 season.

Here are 111's accomplishments in 2006 and 2007:
2 x 2006 Innovation in Control Award
1 x 2006 Regional Winner #1
1 x 2006 Regional Winner #2
1 x 2006 Regional Chairman's Award
2 x 2007 General Motors Industrial Design Award
1 x 2007 Motorola Quality Award
2 x 2007 Regional Finalist #2
1 x 2007 Regional Winner #1

Also attached is the updated awards-vs-performance rankings now using 2005 and 2006 data. I like how regional winners #1, #2, and #3 just edge out regional finalists #1, #2, and #3 in seeding on average. When you consider that most of the data is from 2006 and 2007 where the serpentine draft is in use and assume that the higher-seeded alliance tends to win (hmmm... this gives me an idea), you would think that the #3 pick for the winner would tend to be a lower-seeded team than the #3 pick for the finalists.

Re: Some statistics on year-to-year consistency, 2005-2007

Quote:

Originally Posted by Bongle (Post 613560)

Being that I'm bored now, I'm going to update it to include 2005 and 2006, then I'll see about doing some year-to-year relationships.

If you are really bored and want to go completely crazy with this, most regional's award pages will go all the way back to 2003 (if they were around back then) just by changing the 2007 to 2003 (like this for Philly http://www2.usfirst.org/2003comp/events/PA/awards.htm)

It stops there right? Wrong, because we have the old FIRSTStar system. If you wipe all the dust off, it will give you an awards history for every team from 1998 (first year team # set) - 2002 all in one Excel page:eek: . You might have to ignore some of the awards that aren't given anymore like #1 seed and Outstanding Defense. You also might have to take into account the fact that there were only 199 teams back then. Overall, there is 10 years of awards data available if you want it :)