[NE FIRST] District Championship point cutoff

[Paul T.]

after week 1 of competition we thought the point cutoff was going to be around 60 but now were starting to reconsider. With my team only having 22 points (we counted, not official) (won 5 qualification matches - 10 pts got picked next to last - 2 points won the quarter finals 10 pts = 22) meaning we would need 38 ponts (at least) to get to the district championship. I wanted some other opinions of what the cutoff may be.

Thanks
Paul

[Jay O'Donnell]

Doesn't exactly answer your question, but this might give you some idea of where you stand. http://www.chiefdelphi.com/forums/sh...d.php?t=127672

[loyal]

From the start I always thought the minimum points would be 50 or 60. So my guess would be 55 points then you are out. I really think that's what it will take to squeak in.

[M. Lillis]

So 60 teams get into Boston, do Rookie All-Stars and District Chairman's winners take away from that? If so, it is really only the top 42 point-getters (That's a minimum, because there are definitely going to be some top-60 teams that double qualify (chairmans/all-star + point ranking)). My original thought was that it would take about 70 to get in. But if awards take Boston slots, then I would venture to say that it may be closer to 80 points to get in.

[dag0620]

Originally Posted by M. Lillis

So 60 teams get into Boston, do Rookie All-Stars and District Chairman's winners take away from that? If so, it is really only the top 42 point-getters (That's a minimum, because there are definitely going to be some top-60 teams that double qualify (chairmans/all-star + point ranking)). My original thought was that it would take about 70 to get in. But if awards take Boston slots, then I would venture to say that it may be closer to 80 points to get in.

54 teams total will be attending. 9 of those slots will be allocated to District Chairmans award winners. Everyone else will get in via points.

With that said, typically, teams that win Chairmans normally would have qualified via points anyway, so it makes little impact on the amount of points you would need to get in.

[M. Lillis]

Originally Posted by dag0620

54 teams total will be attending. 9 of those slots will be allocated to District Chairmans award winners. Everyone else will get in via points.

Ok, that makes sense.

[MikeE]

tldr: probably 57 or 58 pts but we'll know far more this time next week

The answer to this question really depends on two things we don't know yet:

• how much correlation we see between the points a team gets in both their scored events and
• how many points are "scored" by teams in their third (or fourth!) event.

A high correlation will tend to reduce the Championship threshold since a large mass of points will be allocated to a small number of high performing teams therefore decreasing the points awarded to the mid-tier teams.

Points being allocated to teams in their third event and therefore discarded will also reduce the cutoff threshold by suppressing the total number of points awarded during the district event phase of the season.

We're not quite half way into the season (4 of 9 district events complete) so it's a little early to have a solid prediction; while 9% of teams have completed both scored events over 15% haven't played in any event yet. A graph of districts by first, second and third event teams


Choosing the top 54 robots of 163 teams in New England means we are selecting the top third of teams. In each of the 4 events so far the top tercile (I had to look up that word too) is between 26 and 36 points, averaging 32, so a simple answer is that championship threshold will be twice that.

My current estimate is that the threshold will be slightly lower, probably 57 or 58pts, but in a few days the Week 4 events will be concluding. At that point every team will have played at least one event and a third of teams will have competed in both events, so we should have a far better idea of where the threshold will fall.

If anyone is still reading, here are some unofficial stats for the events so far

Event		NHNAS	NHDUR	CTGRO	MAWOR
Week		1	2	2	3
Teams		39	36	33	39
Rounds		12	12	12	12
Matches		78	72	66	78
DQs		10	1	1	2
Total QualPts	462	430	394	468
Mean QualPts	11.85	11.94	11.94	12.00
Total SeedPts	236	236	236	236
Total ElimPts	210	210	210	210
Total AwardPts	86	86	86	86
Total Points	994	962	926	1000
Mean Points	25.5	26.7	28.1	25.6
StdDev Points	16.7	17.1	15.7	16.2
Max Points	75	72	57	58
Min Points	4	5	6	2
Median Points	23	24	25	19
Top Tercile	33.5	26.5	36.0	31.8

Some other musings on district points from last year's FiM & MAR season can be found here particularly on the substantial advantage of attending smaller events.

[M. Lillis]

Originally Posted by MikeE

how many points are "scored" by teams in their third (or fourth!) event.

Are you implying that points accumulated after the first two events count toward Championship? Because I believe it is the first two district events. Maybe I'm just not reading your words right.

[Andrew Schreiber]

Originally Posted by M. Lillis

Are you implying that points accumulated after the first two events count toward Championship? Because I believe it is the first two district events. Maybe I'm just not reading your words right.

They are "scored" in a sense. However, they don't count towards qualifying. However, those teams are 'taking points' from teams that they could count for effectively removing points from the overall system. A good team competing at 3+ event acts as a sponge concentrating points to them and lowering the threshold for everyone. (Those points are effectively lost)

[M. Lillis]

Originally Posted by Andrew Schreiber

However, those teams are 'taking points' from teams that they could count for effectively removing points from the overall system.

Wow, it makes sense now. Thanks!

[MikeE]

Originally Posted by Andrew Schreiber

They are "scored" in a sense. However, they don't count towards qualifying. However, those teams are 'taking points' from teams that they could count for effectively removing points from the overall system. A good team competing at 3+ event acts as a sponge concentrating points to them and lowering the threshold for everyone. (Those points are effectively lost)

While a team competing in it's 3rd/4th event is likely reducing the points available to the teams at that event, it also has the effect of decreasing the cutoff threshold for district championships so it benefits teams not at that event.

So from a purely strategic point of view a team should go to the earliest and smallest events possible, all other things being equal.

[Andrew Schreiber]

Originally Posted by MikeE

While a team competing in it's 3rd/4th event is likely reducing the points available to the teams at that event, it also has the effect of decreasing the cutoff threshold for district championships so it benefits teams not at that event.

So from a purely strategic point of view a team should go to the earliest and smallest events possible, all other things being equal.

Thank you for reiterating exactly what I said.



And there are multiple reasons to go to 3+ events. For example - One of our events is geared towards training the next generation of folks without worrying about hurting our chances at qualifying. It's a fun chance that we feel is well worth the extra cost.

And for those who said we shouldn't be allowed to be picked - That would actually hurt the higher seeded points as teams that do 3+ events tend to have more on field experience and perform well. This would mean that #8 would be able to pick the best 3+ event team which would defeat one of the main benefits of being #1 (being able to pick whoever you want)... Plus that's like saying teams shouldn't be allowed to play in multiple events...

[Nathan Streeter]

Originally Posted by MikeE

Choosing the top 54 robots of 163 teams in New England means we are selecting the top third of teams. In each of the 4 events so far the top tercile (I had to look up that word too) is between 26 and 36 points, averaging 32, so a simple answer is that championship threshold will be twice that.

My current estimate is that the threshold will be slightly lower, probably 57 or 58pts, but in a few days the Week 4 events will be concluding. At that point every team will have played at least one event and a third of teams will have competed in both events, so we should have a far better idea of where the threshold will fall.

It looks like your analysis looks only at the points awarded at events... not the rookie and 2nd-year points also. This will be 10 points for each rookie and 5 points for each 2nd year team... which, if there were 10 of each in NE would actually only amount to 150 points. Averaged out, this would probably affect the threshold by less than a point.

Given where the cut-off was in MI in 2013 (I don't recall off the top of my head; I think it was around 52-54) the prediction of 57 to 58 seems reasonable given the various changes from FiM 2013 to NEFIRST 2014 (rookie/second year points, different percentage of teams making DCMP, higher point values for many awards in 2014, etc.).

As Jess said though, the cut-off won't be solid... a fair number of teams will likely decide not to go to DCMP for a variety of reasons, even if they are above the threshold. This will shift the effective threshold downwards by some to-be-determined amount.

Agreed that predictions should be far better after this week, when all the teams will've played their first event and a larger percentage will've played their second event.

[Deke]

Originally Posted by Nathan Streeter

It looks like your analysis looks only at the points awarded at events... not the rookie and 2nd-year points also. This will be 10 points for each rookie and 5 points for each 2nd year team... which, if there were 10 of each in NE would actually only amount to 150 points. Averaged out, this would probably affect the threshold by less than a point.

Given where the cut-off was in MI in 2013 (I don't recall off the top of my head; I think it was around 52-54) the prediction of 57 to 58 seems reasonable given the various changes from FiM 2013 to NEFIRST 2014 (rookie/second year points, different percentage of teams making DCMP, higher point values for many awards in 2014, etc.).

As Jess said though, the cut-off won't be solid... a fair number of teams will likely decide not to go to DCMP for a variety of reasons, even if they are above the threshold. This will shift the effective threshold downwards by some to-be-determined amount.

Agreed that predictions should be far better after this week, when all the teams will've played their first event and a larger percentage will've played their second event.

Last year in FiM the cutoff was 76, that was with ~205 teams and an average of 47 points per team. This year there are 278 teams with an average of 54 points a team because of the new award and rookie and soph point bonuses. It looks like they are projecting low 80s as the cutoff. So most teams would need to at least advance to the finals once and win one round of the elims in the other district to advance. Otherwise they would need alot of qual wins and high pick points with award wins to advance to state.

[MikeE]

Originally Posted by Nathan Streeter

It looks like your analysis looks only at the points awarded at events... not the rookie and 2nd-year points also. This will be 10 points for each rookie and 5 points for each 2nd year team... which, if there were 10 of each in NE would actually only amount to 150 points. Averaged out, this would probably affect the threshold by less than a point.

Given where the cut-off was in MI in 2013 (I don't recall off the top of my head; I think it was around 52-54) the prediction of 57 to 58 seems reasonable given the various changes from FiM 2013 to NEFIRST 2014 (rookie/second year points, different percentage of teams making DCMP, higher point values for many awards in 2014, etc.).

As Jess said though, the cut-off won't be solid... a fair number of teams will likely decide not to go to DCMP for a variety of reasons, even if they are above the threshold. This will shift the effective threshold downwards by some to-be-determined amount.

Agreed that predictions should be far better after this week, when all the teams will've played their first event and a larger percentage will've played their second event.

That's correct, I didn't include the impact of age bonus. From the teams that have competed so far 1st & 2nd year teams distribute in roughly the same way as veteran teams.

There are approximately ~8800 points allocated from events and just 225 from the age bonus so the overall impact on the threshold is small. Of course it will elevate some young teams who are close to the threshold, but that is expected and was the stated purpose of the age bonus.

It will have a much bigger impact in Michigan where almost 40% of teams are rookies or in their second year.

Applying the NE ratio of teams attending Championship to last year's Michigan points data (rescored with the new unified point model) would have given a threshold of 58pts. That's one of other data points that went into my 57-58pt estimate.