Top competitions sorted by average rp

Sorting by average rp of the event after quals:

The israel dcmp starts more 5 days and will include the best teams from the top 2 events.
I strongly suggest to watch it if you have the time because the event will behave like alot of strong events that will come in the following weeks

  1. ISR District Event #3 - 2.0629999999999984
  2. ISR District Event #4 - 1.8796874999999988
  3. Central Valley Regional - 1.7831578947368407
  4. Arkansas Regional - 1.763793103448275
  5. FIN District Princeton Event presented by Toyota - 1.7560714285714276
  6. NE District Rhode Island Event - 1.7463333333333324
  7. Ventura County Regional - 1.706086956521738
  8. NE District SE Mass Event - 1.671666666666666
  9. Hueneme Port Regional - 1.6443181818181827
  10. FMA District Hatboro-Horsham Event - 1.643529411764705
  11. Orange County Regional - 1.63404255319149
  12. ONT District Georgian Event - 1.6210714285714276
  13. Midwest Regional - 1.5960869565217397
  14. ISR District Event #1 - 1.5867999999999993
  15. PNW District Glacier Peak Event - 1.584999999999999
  16. PNW District Clackamas Academy Event - 1.5817647058823519
  17. St. Louis Regional - 1.578837209302326
  18. FIM District Jackson Event presented by Consumers Energy Foundation - 1.5755263157894726
  19. ISR District Event #2 - 1.573448275862068
  20. Southern Cross Regional - 1.5508108108108114
  21. Northern Lights Regional - 1.547966101694914
  22. Great Northern Regional - 1.5433962264150938
  23. FIM District Kettering University Event #2 presented by Ford - 1.5035897435897427
  24. FIN District Mishawaka Event - 1.4972727272727264
  25. FNC District UNC Asheville Event - 1.488620689655172
  26. Utah Regional - 1.4795454545454532
  27. Lake Superior Regional - 1.478999999999999
  28. FIM District Calvin University Event presented by Amway - 1.477749999999999
  29. FIT District Dallas Event - 1.472580645161289
  30. FIM District Milford Event presented by GM Proving Grounds - 1.4639473684210518
  31. South Florida Regional - 1.459999999999999
  32. PNW District Wilsonville Event - 1.452142857142856
  33. CHS District Bethesda MD Event - 1.4483333333333324
  34. FIM District Kettering University Event #1 presented by Ford - 1.4446153846153835
  35. NE District Granite State Event - 1.4394871794871789
  36. FIM District Lansing Event - 1.4319444444444436
  37. ONT District Newmarket Complex Event - 1.429999999999999
  38. FIT District Channelview Event - 1.4230555555555546
  39. FMA District Mount Olive Event - 1.4227027027027017
  40. CHS District Blacksburg VA Event - 1.4226470588235285
  41. FIM District St. Joseph Event presented by Whirlpool Corporation - 1.4115384615384607
  42. Canadian Pacific Regional - 1.4088571428571426
  43. FIT District Waco Event - 1.4009756097560964
  44. NE District Waterbury Event - 1.3789473684210516
  45. FIM District Escanaba Event presented by Highline Fast - 1.371842105263157
  46. FIM District Wayne State University Event presented by Magna - 1.3689743589743582
  47. Oklahoma Regional - 1.3641176470588237
  48. PCH District Anderson Event presented by Magna - 1.3581481481481474
  49. FIT District Belton Event - 1.3392105263157887
  50. Regional Monterrey - 1.3125000000000002
  51. Week 0 - 1.275
  52. PCH District Dalton Event - 1.2472
  53. PCH District Albany Event - 1.1985714285714284
  54. FNC District Johnston County Event - 1.1879999999999995
3 Likes

FRC is done mostly by high school students. Seeing RP expressed to 17 figures reminds me that I was chased around the chalkboard by my high school Physics teacher, and later by several professors and TAs at Georgia Tech, about the wisdom of expressing experimental results using the correct number of figures. The idea is that only a certain number of figures in any given result are necessary to convey the precision and accuracy with which that result can be confidently stated, and that extending the calculation to further figures conveys no further significance.

So, for all the high school students out there who use math to analyze competition results, what is the number of significant figures to which RP should be expressed? Let’s assume that each team plays 12 qual matches, which is the case for District events.

31 Likes

I feel a better indication of overall depth of field would be average match points per event. Especially since many week 1 events weren’t effectively chasing hybrid nodes for the RP, but rather going for higher scoring.

7 Likes

I’d use 3 decimal places out of habit on the grounds that “anything past thousandths is just irrelevant noise”. (inb4 someone argues the same for rounding to 2 places, which probably wouldn’t be that bad either)

8 Likes

It would be 2 because the 12 is the smallest number sig fig wise we have.

4 Likes

Given numbers like 12 matches or 2 RP are considered to have infinite sig figs as they are given and not measured.

If you want to get picky (and who doesn’t?) then we cut off by the add/subtract rule of least number of digits after the decimal place when adding up a team’s total earned RP (e.g. 24RP for earning 2 RP every match and having 12 matches). Then use the multiply/divide rule when averaging and that is keeping the least number of sig figs between what you are multiplying or dividing. Therefore since all of these are given numbers, they all have infinite number of sig figs and we can take them out as far as the calculators will allow.

And considering sig figs are no longer a part of any curriculum in Michigan or across AP classes, I would frankly be impressed by anyone correctly using those.

6 Likes

I know this may be sacrilegious, but I tell my AP Physics students (after they’ve taken AP Chem) just use 3 or 4 digits and you’re fine. :laughing:

11 Likes

Significant figures tries to solve two problems, and honestly kinda sucks at both.

As a means of propagating uncertainty through calculations, it’s fine sometimes, but it has to introduce several special cases for known numbers. It’s also super imprecise, only giving you he order of magnitude of the uncertainty. It is convenient for giving the relative uncertainty, whose order of magnitude is the number of sig figs.

The other thing sig figs tries to do is trim numbers down to a nice size for humans to read. And in some contexts, it makes sense to do this based on the uncertainty. But there’s also many cases where it doesn’t, like here. A human’s intuition for the right number of digits to display is usually going to be better than the output of a single, one-size-fits-all rule. (Caveat: sig figs may help in learning this intuition, but still shouldn’t be used after that)

So yeah, if you care about uncertainty, then use actual uncertainty propagation equations. Otherwise, just use 3-4 digits.


Another note, I used some software to check and confirmed that the ranking score of teams (playing up to 12 matches), rounded to two decimal places like on TBA, is sufficient to determine the exact reduced fraction of that ranking score.

For events, three decimal places is probably best.

9 Likes


https://xkcd.com/2170/

37 Likes

I was just commenting to some friends that when reviewing Dean’s List submissions ahead of an event, I am shocked how many decimal places some GPAs are reported to. I don’t even understand, in some cases, how the math possibly reported them.

4 Likes

You’re not a student any more, Andrew. However, you are correct, and that is an important thing, especially for a mentor.

1 Like

I’m thinking a more interesting number would be the RP needed to be in the top 8 (or 4 or 16).

Why not all?

Event Name Week Team Count   4th   8th 16th
NE District North Shore Event 3 40 2.92 2.58 1.75
ISR District Event #3 2 30 2.75 2.50 2.17
Miami Valley Regional 3 50 2.90 2.50 2.20
Central Illinois Regional 3 32 2.75 2.42 1.75
Central Valley Regional 2 38 2.75 2.42 1.67
FIT District San Antonio Event 3 43 2.83 2.42 1.83
San Francisco Regional 3 42 2.80 2.40 1.90
ISR District Event #4 3 32 2.58 2.33 1.83
Finger Lakes Regional 3 55 2.56 2.33 2.00
Hueneme Port Regional 1 44 2.45 2.27 1.82
NE District Rhode Island Event 2 30 2.67 2.25 1.42
FIM District Traverse City Event 3 38 2.50 2.25 1.75
FIM District Belleville Event 3 40 2.67 2.25 1.75
Ventura County Regional 2 46 2.56 2.22 1.89
Northern Lights Regional 1 59 2.56 2.22 2.11
Orange County Regional 2 47 2.40 2.20 1.90
Great Northern Regional 2 53 2.50 2.20 1.90
Midwest Regional 2 46 2.64 2.18 2.00
PNW District SunDome Event 3 27 2.25 2.17 1.67
FMA District Springside Chestnut Hill Academy Event 3 31 2.58 2.17 1.67
ONT District Western University Engineering Event 3 33 2.75 2.17 1.75
Arizona East Regional 3 37 2.50 2.17 1.67
FIM District Standish-Sterling Event 3 39 2.25 2.17 1.83
Los Angeles Regional 3 44 2.50 2.10 2.00
Orlando Regional 3 54 2.50 2.10 1.70
St. Louis Regional 2 43 2.36 2.09 1.82
FIN District Princeton Event 2 28 2.25 2.08 1.83
Arkansas Regional 1 29 2.58 2.08 1.67
NE District SE Mass Event 1 30 2.42 2.08 1.58
FMA District Hatboro-Horsham Event 1 34 2.42 2.08 1.67
NE District Western NE Event 3 36 2.83 2.08 1.75
FIM District Jackson Event 1 38 2.58 2.08 1.67
FIM District Detroit Event 3 39 2.25 2.08 1.50
FIM District Kettering University Event #2 2 39 2.42 2.08 1.50
ISR District Event #2 1 29 2.25 2.00 1.58
FIT District Fort Worth Event 3 34 2.42 2.00 1.58
PNW District Glacier Peak Event 1 34 2.17 2.00 1.58
FIM District Milford Event 1 38 2.42 2.00 1.50
FIM District Muskegon Event 3 38 2.83 2.00 1.58
FIT District Waco Event 1 41 2.25 2.00 1.42
Lake Superior Regional 1 60 2.33 2.00 1.89
PNW District Wilsonville Event 2 28 2.00 1.92 1.25
FMA District Robbinsville Event 3 31 2.33 1.92 1.50
South Florida Regional 1 34 2.42 1.92 1.50
PNW District Clackamas Academy Event 1 34 2.25 1.92 1.67
FIM District Lansing Event 2 36 2.08 1.92 1.50
FIM District Kettering University Event #1 1 39 2.42 1.92 1.58
FIM District Calvin University Event 1 40 2.33 1.92 1.58
CHS District Alexandria VA Event 3 40 2.08 1.92 1.50
Utah Regional 1 44 2.25 1.92 1.50
Southern Cross Regional 2 37 2.36 1.91 1.55
ONT District Georgian Event 1 28 2.25 1.83 1.58
FNC District UNC Asheville Event 1 29 2.17 1.83 1.50
FIT District Dallas Event 1 31 2.25 1.83 1.50
FIN District Mishawaka Event 1 33 2.25 1.83 1.50
CHS District Blacksburg VA Event 1 34 1.92 1.83 1.50
Heartland Regional 3 36 2.17 1.83 1.42
CHS District Bethesda MD Event 2 36 2.17 1.83 1.50
NE District Waterbury Event 2 38 2.33 1.83 1.50
FIM District Wayne State University Event 2 39 2.17 1.83 1.50
FIM District St. Joseph Event 2 39 2.25 1.83 1.42
Regional Monterrey 1 40 1.90 1.80 1.30
FNC District Mecklenburg County Event 3 27 1.92 1.75 1.25
FNC District Wake County Event 3 27 2.42 1.75 1.33
PNW District Bonney Lake Event 3 31 2.50 1.75 1.42
ONT District Newmarket Complex Event 1 34 2.33 1.75 1.42
Canadian Pacific Regional 1 35 2.08 1.75 1.42
FIT District Channelview Event 2 36 2.00 1.75 1.42
FMA District Mount Olive Event 2 37 2.25 1.75 1.50
FIM District Escanaba Event 1 38 2.00 1.75 1.42
NE District Granite State Event 1 39 2.38 1.75 1.63
Brazil Regional 3 42 2.30 1.70 1.40
ISR District Event #1 1 25 2.08 1.67 1.33
PCH District Gwinnett Event 3 27 2.08 1.67 1.25
PCH District Anderson Event 2 27 2.17 1.67 1.17
ONT District Humber College Event 3 28 2.00 1.67 1.25
CHS District Portsmouth VA Event 3 29 1.92 1.67 1.42
Oklahoma Regional 2 34 1.82 1.64 1.36
Magnolia Regional 3 29 2.54 1.62 1.38
Regional Puebla 3 33 1.92 1.58 1.33
FIT District Belton Event 2 38 2.00 1.58 1.42
FNC District Johnston County Event 2 30 1.67 1.50 1.17
PCH District Dalton Event 2 25 1.67 1.33 1.00
PCH District Albany Event 1 21 1.58 1.25 0.92
10 Likes

Wither, buddy, you are a mad lad! Thanks for all the handy dandy numbers (especially for SunDome, definitely fun to see!)

2 Likes

Sorting and plotting your excellent data, to see the trend. Strong events are definitely getting stronger, and Week3 is definitely stronger than Weeks 1 and 2.

7 Likes

Silly us, picking the 2nd toughest week 1.

4 Likes

Silly us, picking the toughest Week 3.

6 Likes

I’m not sure how useful these numbers are. Stronger events do get more ranking points on average, but these aren’t always concentrated in the top teams. Especially because a deep event could result in the top teams winning less, and thus having a lower RP/match.

Silly us, doing FRC with three full-time students.

9 Likes

Well math says this can’t happen.

The average winning RP for every event is going to be 1.0000000000000000 (double precision, RW)

More meaningful would be to subtract off that 1 to get the bonus RPs at each event. :slight_smile:

2 Likes