[FRC Blog] 2017 FIRST Championships Allocations for Districts

[Jon Stratis]

Originally Posted by AGPapa

Under the old system the North would send ~265 teams from district areas, but under this system they'll only send 201. Given that district teams are much more competitive than waitlist teams, this will seriously decrease the quality of championship play.

I don't think that analysis applies here. Michigan (for example) sent 76 teams. if ~56% made it to the playoffs, that's only ~43 teams. How competitive are the other 33 teams when compared to the waitlist spots? Because this change doesn't really affect those 43 teams - they'll get in either way. It's the team's further down in the district points that are affected, and I haven't seen an analysis that really tells us how competitive those teams are, when compared to the waitlist teams. Do that same analysis comparing the achievement of all waitlist teams against the lower 10% of district teams (sorted by district points), and then we can talk about how competitive or not this decision is.

[AGPapa]

Originally Posted by Jon Stratis

I don't think that analysis applies here. Michigan (for example) sent 76 teams. if ~56% made it to the playoffs, that's only ~43 teams. How competitive are the other 33 teams when compared to the waitlist spots? Because this change doesn't really affect those 43 teams - they'll get in either way. It's the team's further down in the district points that are affected, and I haven't seen an analysis that really tells us how competitive those teams are, when compared to the waitlist teams. Do that same analysis comparing the achievement of all waitlist teams against the lower 10% of district teams (sorted by district points), and then we can talk about how competitive or not this decision is.

That's a very good point. Looking at the data, the bottom 10% of district teams only had 3 selected out of 22 (teams 193, 3539 and 3663). That's only 13.7%, meanwhile 16.7% of waitlist teams were selected for division playoff matches.

I'm pretty surprised about that, since all waitlist teams from district areas would have fewer district points than the points teams. Either regional waitlist teams are doing well enough to pull up the average or district points aren't doing that a good a job of sending the best teams.

Originally Posted by FarmerJohn

If you want to go to the championships then work harder, don't try and lessen the value of the championships by letting everyone in.

I don't think you're being serious, but it's important to note that nobody else is talking about how many teams are at champs, just how to pick those ones (points based or lottery). The same number of teams are getting the "championship experience" either way.

[Jon Stratis]

Originally Posted by AGPapa

That's a very good point. Looking at the data, the bottom 10% of district teams only had 3 selected out of 22 (teams 193, 3539 and 3663). That's only 13.7%, meanwhile 16.7% of waitlist teams were selected for division playoff matches.

I'm pretty surprised about that, since all waitlist teams from district areas would have fewer district points than the points teams. Either regional waitlist teams are doing well enough to pull up the average or district points aren't doing that a good a job of sending the best teams.

Well, speaking from experience... when a team is having a bad year, it might turn down a waitlist spot. On the other hand, when a team is doing well and just misses making it to champs, they're much more willing to accept such a spot.

And then you can consider which teams would apply for waitlist spots. While there are certainly exceptions, it's going to be the better funded teams, as attending champs is a rather expensive proposition. And it'll probably be the more confident teams, those that are confident they can build a decent robot and do well enough at champs to make a trip worthwhile (even if they are realists and realize the schedule or luck in playoffs can work against them).

Now, think about all the regional teams you don't see at champs. Teams that were finalists (although there's less of that now with the wildcards), semifinalists, or quarterfinalists, all of which may be pretty decent - especially when you look at the different relative strengths of regionals. Just take a look at the 2014 Lake Superior regional playoffs for an example - every set except Semifinals 2 went to a third match, with so many close games that any alliance really could have won that event that year.

So while any team can get in off the waitlist, there are enough good teams across FIRST, and enough psychology involved in spending that type of money, to suggest that we'll get some better teams than you might expect from it.

[Mr V]

Originally Posted by AGPapa

That's a very good point. Looking at the data, the bottom 10% of district teams only had 3 selected out of 22 (teams 193, 3539 and 3663). That's only 13.7%, meanwhile 16.7% of waitlist teams were selected for division playoff matches.

I'm pretty surprised about that, since all waitlist teams from district areas would have fewer district points than the points teams. Either regional waitlist teams are doing well enough to pull up the average or district points aren't doing that a good a job of sending the best teams.

Note just because a team accepted a waitlist position does not mean that they had less points than the teams that qualified on points. Two seasons ago there was a team in the PNW that received and accepted a wait list spot on the first day of DCMP. By the end of DCMP they did earn enough points to qualify but since they already had a spot their points spot was passed on to the next team.

The district points system is not designed to send the top performing robots to CMP, it is designed to send the teams that earn the most points. Yes robot performance is where the bulk of the points are earned but awards earn points and there are bonuses for rookie and 2nd year teams. At the cutoff point the range of points is often quite small and winning just one award can make a difference of earning a CMP spot or not.

[Brian Maher]

Originally Posted by Mr V

Note just because a team accepted a waitlist position does not mean that they had less points than the teams that qualified on points. Two seasons ago there was a team in the PNW that received and accepted a wait list spot on the first day of DCMP. By the end of DCMP they did earn enough points to qualify but since they already had a spot their points spot was passed on to the next team.

While this is technically true (it happened to 303, 3637, and 5404 in MAR this year), I think that for the purposes of analyzing this proposal, Antonio's assumption should be used.
Say team A is above the points cutoff and team B is directly below it. If team A receives a waitlist spot, team B qualifies for Champs. While team A technically received a waitlist spot and team B earned a spot on district points, team B only earned their spot because team A received a waitlist spot. In this regard, I think it makes more sense to categorize team B as the waitlist team than team A because team B would not have earned a spot if it weren't for the waitlist.

Originally Posted by AGPapa

Either regional waitlist teams are doing well enough to pull up the average or district points aren't doing that a good a job of sending the best teams.

My hypothesis is that districts are much better than regionals at qualifying competitive teams for Champs. I think the waitlist picks up the slack with qualifying good regional teams who missed champs because they were not on the right alliance to win/make finals and didn't win a culture award. Many of these teams, if they were in a district, could have easily qualified for champs.

[Mr V]

Originally Posted by BMSOTM

While this is technically true (it happened to 303, 3637, and 5404 in MAR this year), I think that for the purposes of analyzing this proposal, Antonio's assumption should be used.
Say team A is above the points cutoff and team B is directly below it. If team A receives a waitlist spot, team B qualifies for Champs. While team A technically received a waitlist spot and team B earned a spot on district points, team B only earned their spot because team A received a waitlist spot. In this regard, I think it makes more sense to categorize team B as the waitlist team than team A because team B would not have earned a spot if it weren't for the waitlist.

While certainly team B would not have made it if team A had not made the space available be getting a wait list spot, in the official record Team A will be shown as attending due to a wait list spot and Team B earning their spot on district points.

[Basel A]

Originally Posted by Mr V

While certainly team B would not have made it if team A had not made the space available be getting a wait list spot, in the official record Team A will be shown as attending due to a wait list spot and Team B earning their spot on district points.

In what official record? I have no knowledge of any public record of this.

[AGPapa]

Originally Posted by Basel A

In what official record? I have no knowledge of any public record of this.

I am also not aware of any official record of how teams qualified.

Here is my recreation of how teams qualified. Note that it incorrectly includes DCMP winners as "District Points" spots.

Distinguishing waitlist teams from district points teams was tough. If a district got X points spots, then I took the top X teams from that district who did not already qualify. The rest were included as waitlist teams. There was usually a pretty clear cut off point.

[logank013]

What is the predicted spots allocated for 2017 based off in the 2016 season? Because you have PNW going from 30 to 41 and then there is Indiana going from 9 to 10. Indiana has an 11% increase where PNW has about a 37% increase.

Are these increases based off of a projection of how many rookie teams will be added to the 2017 season? IN gained 0 teams last year total so is FIRST assuming IN will gain 0 teams again this year? I'm just curious how these were designated. Thanks

[Jon Stratis]

Originally Posted by logank013

What is the predicted spots allocated for 2017 based off in the 2016 season? Because you have PNW going from 30 to 41 and then there is Indiana going from 9 to 10. Indiana has an 11% increase where PNW has about a 37% increase.

Are these increases based off of a projection of how many rookie teams will be added to the 2017 season? IN gained 0 teams last year total so is FIRST assuming IN will gain 0 teams again this year? I'm just curious how these were designated. Thanks

from the blog "if 2017 were to look like 2016 with respect to team counts"

So it's using the exact same numbers as 2016, assuming no increase anywhere.

The difference you note is the difference between norths champs and south champs. districts going to south champs are going to get bigger increases than districts going to north champs.

[Taylor]

Originally Posted by logank013

What is the predicted spots allocated for 2017 based off in the 2016 season? Because you have PNW going from 30 to 41 and then there is Indiana going from 9 to 10. Indiana has an 11% increase where PNW has about a 37% increase.

Are these increases based off of a projection of how many rookie teams will be added to the 2017 season? IN gained 0 teams last year total so is FIRST assuming IN will gain 0 teams again this year? I'm just curious how these were designated. Thanks

Originally Posted by Jon Stratis

from the blog "if 2017 were to look like 2016 with respect to team counts"

So it's using the exact same numbers as 2016, assuming no increase anywhere.

The difference you note is the difference between norths champs and south champs. districts going to south champs are going to get bigger increases than districts going to north champs.

Looks like a good time to move to North Carolina.

[logank013]

Originally Posted by Jon Stratis

from the blog "if 2017 were to look like 2016 with respect to team counts"

So it's using the exact same numbers as 2016, assuming no increase anywhere.

The difference you note is the difference between norths champs and south champs. districts going to south champs are going to get bigger increases than districts going to north champs.

So because we have so many more district teams in HalfChampsN compared to HalfChampsS, it causes less "new spots" to be distributed to each of the HalfChampsN districts?

Full Sarcasm Alert!: Can FIRST do us all a favor and send 3/4 of MI to half champs south? 201-60 = 141 and 88+60= 148. That'd make for a much better district distribution and would cause the same percentage increase of HalfChamps spots allocated to each district

[Ernst]

Originally Posted by logank013

So because we have so many more district teams in HalfChampsN compared to HalfChampsS, it causes less "new spots" to be distributed to each of the HalfChampsN districts?

Full Sarcasm Alert!: Can FIRST do us all a favor and send 3/4 of MI to half champs south? 201-60 = 141 and 88+60= 148. That'd make for a much better district distribution and would cause the same percentage increase of HalfChamps spots allocated to each district

I was curious about this.

Using 2016 numbers and accounting for 3114 of 3130 teams (there are 16 hiding somewhere):

North: 1766 teams
South: 1348 teams

That's including the US, Canada, Israel, China, Australia, Mexico, Brazil, and a few other smaller countries.

The blog post mentions

the formula for 2017 will use the percent representation of each District compared to the number of FRC teams just within the geography assigned to the District’s home Championship

So there's your answer.


I wonder how much FIRST looked at team counts before drawing the lines? Or if it was more based on population? Really populous states like Texas, California, and Florida are all in the south, but they're currently lagging behind many northern and eastern states in teams/person.

[MechEng83]

Originally Posted by Ernst

I was curious about this.

Using 2016 numbers and accounting for 3114 of 3130 teams (there are 16 hiding somewhere):

North: 1766 teams
South: 1348 teams

That's including the US, Canada, Israel, China, Australia, Mexico, Brazil, and a few other smaller countries.

The blog post mentions

So there's your answer.


I wonder how much FIRST looked at team counts before drawing the lines? Or if it was more based on population? Really populous states like Texas, California, and Florida are all in the south, but they're currently lagging behind many northern and eastern states in teams/person.

If you look at the 2017 vs 2018 distribution, it looks like the distributions are not quite as drastic long term.

Using 2016 numbers, there were 22 Kansas teams and 73 Missouri teams, which are part of the North Champs in 2017, but will be part of the South Champs in 2018.

Those 95 teams skew the balance. In 2018 the distribution, using Ernst's numbers would be:

North: 1671
South: 1443

I wonder if we can move another logical grouping of 114 teams to balance out the numbers!

P.S. This assumes that team growth is uniform throughout all of FIRSTdom, which Michigan has been ruining for a while now.

[adciv]

Originally Posted by MechEng83

P.S. This assumes that team growth is uniform throughout all of FIRSTdom, which Michigan has been ruining for a while now.

Which brings up an interesting question. What will FIRST do if the existing team imbalance between North and South increases? When do we rebalance?