Most Competitive Region

Is the most competitive region defined by the current competitiveness or historic success?

If you use historic data and go by number of awards one, the older teams have an advantage because they have had a longer window to win awards.

If you go by percent of awards won versus years competing the new teams have an advantage because they have not had as many opportunities to lose. A region competing with a handful of teams for only one year but who has had great sucess for that one year will have an advantage over a region who has competed for two years but has only been successful for one of those years.

What defines competitiveness? Should the competitiveness within the region also be considered? Would averaging the difference between the winning and losing score and comparing that to the average scores for each region at that region’s state’s / national estimate the internal competitiveness?

Should the competitiveness within a region also be considered?

OK, thanks to StackOverflow I have an answer and moving forward…

I had the wrong oeprator. $set will insert one, $push puts additional items in.


db.awardout.find().forEach(function(award_team){
if(award_team != null)
{
db.teams.findAndModify(
{
   query: {number: award_team.team},
   update: { **$push**: {accolades:  {sku: award_team.sku, acc_category: "Award", acc_name: award_team.name }] } },
   upsert: true,
   });
   }
   });

Sample here has Green Eggs with their three accolades from 2014 now safely loaded…


/* 1277 */
{
  "_id" : ObjectId("552071f69ad9c19e80996432"),
  "program" : "VRC",
  "robot_name" : "Fred VII",
  "organisation" : "Green Egg Robotics",
  "region" : "Massachusetts",
  "country" : "United States",
  "number" : "44",
  "team_name" : "Green Egg Robotics",
  "city" : "Oakham",
  "grade" : "High School",
  "is_registered" : "0",
  "accolades" : {
        "sku" : "RE-VRC-14-0110",
        "acc_category" : "Award",
        "acc_name" : "Tournament Finalists (VRC/VEXU)"
      }], {
        "sku" : "RE-VRC-14-0110",
        "acc_category" : "Award",
        "acc_name" : "Robot Skills Winner (VRC/VEXU)"
      }], {
        "sku" : "RE-VRC-14-0110",
        "acc_category" : "Award",
        "acc_name" : "Think Award (VRC/VEXU)"
      }]]
}


Now to load more years worth of awards and then get rankings/elimination match data loaded. Then on to aggregating this up to team’s region and country.

We can look at all those things once we get the data organized!

We can load up not only the worlds data but all the data on roboevents and then look at the teams over the entire season or what happened in the region prior to coming to worlds.

Working backwards from success at worlds to what happened before is one good way of looking at this because you can back test theories based upon what you have seen before. Is competitiveness/success at worlds caused by factors happening down in the region? Is dominance at region level a predictor of what will happen at worlds? Is it number or types of awards won at the regional level the important part? Is it the entire season or just the qualifying tournament for worlds?

Wow now I know i am not the only one who thinks BC is good this year. That is kinda cool to me in way.

I need to load more award data up from previous events.

California had the most accolades in awards at high school division. But they also sent the most teams too.

So is strength of region the sheer number of winners or is it likelihood of producing the highest percentage of accolades per team (can be over 100%)?

If I had to guess, New Hampshire did not send all that many teams last year and will have one of the highest award to robot present ratio.

So interim results of just last year high school accolades per region and country are…

{ "_id" : { "country" : "United States", "region" : "California", "event" : "RE-VRC-14-0110" }, "count" : 11 }
{ "_id" : { "country" : "United States", "region" : "Texas", "event" : "RE-VRC-14-0110" }, "count" : 10 }
{ "_id" : { "country" : "Canada", "region" : "British Columbia", "event" : "RE-VRC-14-0110" }, "count" : 6 }
{ "_id" : { "country" : "New Zealand", "region" : "", "event" : "RE-VRC-14-0110" }, "count" : 5 }
{ "_id" : { "country" : "Canada", "region" : "Ontario", "event" : "RE-VRC-14-0110" }, "count" : 5 }
{ "_id" : { "country" : "United States", "region" : "New Hampshire", "event" : "RE-VRC-14-0110" }, "count" : 5 }
{ "_id" : { "country" : "United States", "region" : "Arizona", "event" : "RE-VRC-14-0110" }, "count" : 4 }
{ "_id" : { "country" : "Puerto Rico", "region" : "Not Applicable or Not Listed", "event" : "RE-VRC-14-0110" }, "count" : 4 }
{ "_id" : { "country" : "United States", "region" : "Massachusetts", "event" : "RE-VRC-14-0110" }, "count" : 3 }
{ "_id" : { "country" : "United States", "region" : "Kentucky", "event" : "RE-VRC-14-0110" }, "count" : 3 }
{ "_id" : { "country" : "Puerto Rico", "region" : "N/A", "event" : "RE-VRC-14-0110" }, "count" : 3 }
{ "_id" : { "country" : "United States", "region" : "Colorado", "event" : "RE-VRC-14-0110" }, "count" : 3 }
{ "_id" : { "country" : "United States", "region" : "Pennsylvania", "event" : "RE-VRC-14-0110" }, "count" : 3 }

Here’s the code… You roll up the number of accolade sub document records you have grouped by country, region, and event. (just one event loaded now)

db.teams.aggregate(
    { $unwind : "$accolades" },
    { $group: {
        _id: {country:"$country",region:"$region",event:"$accolades.sku"},
        count: { $sum : 1 }}
        },
        {$sort: {count: -1 }},
        {$limit: 300}
        ]);

At the end of Friday British Columbia is proving to be the top region with a 74 winning percentage. They did send only two clubs though. Wisconsin is behind them with a 67 winning percentage. Florida is next with a 63 winning percentage. Utah has a 62 percent.

Also China is doing very well in the high school division already having won 174 matches for a 63 winning percentage.

source:
http://vex.us.nallen.me/extras/worlds_regions?RE-VRC-15-1717=1

I love this statistics. I really thought this is great.

But at the same time, I guess it is also a classic case of statistics can only tells us that much. The context and exceptions are normally not taken into account.

I will need to say that singapore percentage win of 57% is including the wallbot… Which has been pretty successful in the defensive work but not in scoring.

If we are not including all the wallbots in the calculation (which I really think we shouldn’t, since we are looking at win percentage), then singapore will have a decent percentage of 66.7%.
This will look more decent :smiley:

In my opinion, the top three regions for the past couple of years…

  1. New Zeland
  2. Texas, USA
  3. Canada

This seems reasonably accurate. Although I think Texas should be in there somewhere, because of their number of high point matches and the fact that they did have two teams at the Worlds finals.
BC definitely has some seriously tough competition. (I compete in BC) We had a qualification limit of 8 teams for the worlds, unfortunately due to some large school district cuts 2 of the teams were unable to go. That left our two top robot clubs: Shawnigan Lake Shockbots and Gladstone Robosavages as the only two clubs going to the worlds.
The playoffs at the BC provincial championship included scores of 100 and 101 points. Not quite as impressive as some other competitions, but certainly up there.

– Also, CHHS BRUINS. I agree New Zealand is probably the top robotics country. Its accurate to put Canada in there, but this is a huge country, really its Ontario, BC and some teams from Alberta. I don’t know if there’s much else up here for Vex teams. Someone please correct me if I’m wrong. :slight_smile:

I guess with worlds over I should get back to this and get a new data set in there.

I want to try a few different views of the data and back test them. I fear a few stellar teams from a small region might sway the results when looking at number of teams per area coming to worlds versus number of teams doing great.

Then looking further back into the region itself. Do we look at the state/national championship only as you can be up and down within a season?

One idea I want to test is dominance of score differential in state qualifying matches being a predictor of success in worlds. (but the NZ data not showing scores just winners in some bracket matches may throw this off) The next test is type of accolades prior to worlds as a predictor (Excellence or design awards or skills winners or top 30 in the world in skills)

In addition to BC’s 128 teams and Ontario’s 203, there were 40 teams in Alberta last season, and their top teams are good – I saw them in action twice this year. Plus six teams from Saskatchewan, three from Manitoba and one from Quebec. :slight_smile:

Watch out, the maple leafers are coming.

There were actually six teams from British Columbia at VEX Worlds.

2V, in Science, went 7-3
6264E, also in Science, went 7-2-1
6264G, in Tech, went 8-2
2X, also in Tech, went 9-1
2Z, in Engineering, went 7-3
6264, in Math, went 5-5

So the total record for BC teams was 43-16-1, or about 72%. Unfortunately, for the first time in several years, none of the teams made it out of their division.

It was also unfortunate that two of the BC teams that qualified were unable to attend this year, 1136B and 7793R have both had strong records at worlds in the past and in competition this year. We had a couple other strong teams who had never attended worlds before who would have liked to fill in for them, but the qualifying structure doesn’t allow for it.

Jason

Actually, Texas had three teams, not two, in Worlds finals: 5408F, 9090C, and 118.

I realise that this is trying to find the regions which are best overall, but i think if you define competitiveness as the relationship dynamic between teams then i think you have to count the UK up there. If you are going for high scorers and best robots you may as well forget about us. :slight_smile:

The running joke around the VEXU division was that Jack’s garage is the strongest region.

2915
AURA

2918A who won the High School Excellence Award also played practice games with us in Jack’s garage (only on one day but it totally counts…?)

In fact, teams who practice in Jack’s garage took out 5 of the 8 “Big Red Trophies” across the High School and VexU divisions at Worlds 2015, 9 out of 16 in the past 2 years, and 11 out of 24 in the past 3 years, as well as a bunch of finalist titles…

OYES - 2014 VexU Programming Skills, Robot Skills, Excellence Award
2915C - 2014 High School World Champions
AURA - 2015 VexU Programming Skills, Robot Skills, Excellence Award
2915A - 2015 High School World Champions, 2013 High School World Champions, 2013 High School skills champions (2 Skills finalist titles as well)
2918A - 2015 High School Excellence Award

Take that BC :stuck_out_tongue:

Does 7682 practice there too? They won the tournament level Innovate Award.

This made me want to send my teams to Jack’s garage… especially since Singapore is now out of APAC… :cool:

We need to all show up at his door. He uploaded a few quad copter videos of him flying above his house so maybe that will help us narrow down his exact address.

Maybe this is why BNS didn’t do as well as we wanted to this year. Can you fit one more VEXU team in the famous garage? :stuck_out_tongue: