2019 Chezy Champs Zebra's Dart Data Analysis

Similar to my previous threads regarding Zebra’s Dart at IRI here and here, I will be sharing some statistics from the 2019 Chezy Champs offseason event here. All data was obtained using the Zebra’s Dart UWB sensors, and analyzed using open-source code available here. Thank you to @Matt_Boehm_329, @Tom_Boehm_329, and everyone who helped make this possible. Let’s dive right in.

Most Defense in Quals
  1. 498 - 82.5%
  2. 6443 - 82.5%
  3. 1671 - 57.3%
  4. 2928 - 50.5%
  5. 972 - 48.9%

Compared to IRI, the teams that played defense played it more exclusively. Correspondingly, there were fewer teams who played defense more than 50% of the time.

Most Defended in Quals
  1. 4414 - 45.9%
  2. 1619 - 41.7%
  3. 846 - 36.9%
  4. 1678 - 36.4%
  5. 254 - 35.9%

Defense seems slightly lighter here compared to at IRI. 1114 was defended 56% of the time for example.

Elims Defense/Defended

Most Defense in Elims (Min 2 Matches)

  1. 2928 - 94.3%
  2. 498 - 93.5%
  3. 2930 - 92.2%
  4. 3647 - 88.1%
  5. 6443 - 83.4%

Most Defended in Elims (Min 2 Matches)

  1. 3476 - 71.2%
  2. 1678 - 65.1%
  3. 3309 - 62.0%
  4. 4414 - 54.3%
  5. 973 - 51.8%

Similar trends as in qualification matches, with even more extreme percentages due to smaller sample sizes.

Zebra Power Rating (ZPR)
  1. 971 - 50.6
  2. 254 - 47.3
  3. 114 - 45.0
  4. 1619 - 44.8
  5. 1678 - 40.7
  6. 2910 - 40.6
  7. 5026 - 39.4
  8. 1983 - 38.9
  9. 5818 - 38.0
  10. 6443 - 37.7

ZPR, a new quantity I calculate, uses a similar equation as OPR (Offensive Power Rating), but attempts to factor in time spent on defense and being defended. A ZPR represents the value a team contributes in a match without any defense played on them. ZPR also rewards teams playing defense more than OPR does. 2056 and 1114 had higher ZPRs at IRI than any team at Chezy Champs, although I have not fully investigated if ZPR can be compared cross events.

Average Moving Speed
  1. 3309 - 5.75 ft/s
  2. 2910 - 5.67 ft/s
  3. 696 - 5.64 ft/s
  4. 1678 - 5.63 ft/s
  5. 2659 - 5.61 ft/s
  6. 971 - 5.54 ft/s
  7. 3476 - 5.49 ft/s
  8. 846 - 5.48 ft/s
  9. 6443 - 5.39 ft/s
  10. 973 - 5.32 ft/s

This represents the average speed among the times a robot is moving at least 2 ft/s. Teams generally are nearly stationary (velocity under 2 ft/s) roughly 40% of a match.

95th Percentile Speed
  1. 2910 - 12.0 ft/s
  2. 4414 - 11.8 ft/s
  3. 5199 - 11.7 ft/s
  4. 649 - 11.3 ft/s
  5. 2928 - 11.2 ft/s
  6. 2046 - 11.1 ft/s
  7. 1678 - 11.0 ft/s
  8. 1619 - 10.9 ft/s
  9. 2659 - 10.8 ft/s
  10. 3309 - 10.7 ft/s

This statistic gives insight into the top speed teams hit. Surprisingly, the six fastest teams at IRI (including 2910) all had higher 95th Percentile Speeds than teams at Chezy Champs. I wonder if it had anything to do with styles of offense and defense.

There are some interesting trends that can be explored further. I’m also looking into other uses of this data, and am open to questions, suggestions, and feedback. Thank you for reading!

Edit 1: Formatting
Edit 2: Typo

29 Likes

1671 had some killer defense, but they didn’t get a chance to play in elims. Could the 2nd most defended bot in elims be a typo for 1678?

super cool analysis, nice to see us up there with the greats.

1 Like

Thanks for finding that typo! I’ve edited the original post.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.