BirdSO Invitational 2021

[RobertYL]

Event Supervisor Review

Hi! My name is Robert Lee and I had the pleasure of supervising for the inaugural BirdSO invitational. I was the event supervisor for Machines C with Jessica Shah and the co-event supervisor for Fermi Questions B/C with Andrew Zhang and Caleb Chiang. Thank you to all of the teams who competed and congratulations for surviving through a slew of tough tests (from what I've heard). Best of luck in the upcoming regional, state, and national competitions!

Machines C

Statistics:
Mean: 41.4 (27.6%)
Median: 37.3 (24.9%)
St. Dev.: 18.4
Max: 89.8 (59.9%)

Graphs:

Machines_C-Distributions.png (38.97 KiB) Viewed 7147 times

More in-depth statistics and graphs pertaining to sections and specific questions can be found at this link.

Thoughts:
Overall, I am content with the results, but as always, scores could have been better separated at the bottom. This test was a bit too hard for the field, but I think teams can learn a lot from this test. Congratulations to the top teams!

• Section A (Multiple Choice) had a good distribution with teams scoring around an average of 50%. The questions generally had a high correct rate and was generally written to be simple one step calculations. The correct rate of questions 7 and 15 were definitely outliers as the former was a multiple select question and the other used silly EE units (a conversion table is helpful to have on hand!). I do think the amount of kinematics questions may have been higher than necessary for a Machines test, but other than that, nothing of note.
• Section B (Free Response) was where many teams hit a wall, in terms of points. I hoped that teams would be able to get the first two questions as I tried to make them more accessible and only used simple machines. I'll try to include drawings next time as the description may have been unclear. This section had the points allocated a little top-heavy, with many of the points locked up in questions 4 and 5. Regarding question 3, the device design, I was surprised at the low number of submissions (around ~30 in a field of 120 teams) but the teams who did submit scored an average of around 10-12 points. One point that I want to stress is the difference between an inclined plane and wedge. The applied force on the two are different: a mass is pushed parallel and up the inclined plane whereas a wedge is pushed into a mass and it moves upwards (or apart). The direction of forces on the two machines are fundamentally different, or else they wouldn't be put in two categories. I saw this confusion happen at GGSO where planes were substituted for wedges and, surprisingly, at BirdSO where the opposite occurred. Questions 4 and 5 were definitely tougher questions aimed for the most prepared of teams; however, not a single team really made any significant progress on them, so I hope teams can attempt them on their own and learn something new!

Thank you to Jessica for helping out with the test writing process and bringing Snowy back! Also, thank you to the two volunteers who took the test beforehand and helped double check the exam and its answers.

Test Folder:
The exam and all other material can be found in this folder. The folder includes a solution guide walking through each of the questions in section B. I hope teams find it useful!

Fermi Questions C

Statistics:
Mean: 92.4 (30.8%)
Median: 91 (30.3%)
St. Dev.: 30.4
Max: 156 (52.0%)

Graphs:

Fermi_Questions_C-Distributions.png (39.9 KiB) Viewed 7147 times

More in-depth statistics and graphs pertaining to specific questions can be found at this link.

Thoughts:
The results turned out as we expected with a surprisingly nice, even distribution of scores. The test was generally divided into three (unlabeled) sections of easy (1-15), medium (16-45), and hard (46-60). Teams faired pretty well in the first two sections, as can be seen in the detailed statistics for each individual question, but had a pretty steep drop-off in the last section. We hoped that there was a good distribution of difficulties and variety of topics so teams didn't feel too overwhelmed by too much of one thing. Unfortunately, no one got the last game theory-esque question . When grading the tests, we noticed many teams didn't answer using Fermi numbers, but rather in scientific notation or in full numbers (lots of counting digits) and these teams were tiered. Remember! Answers must be Fermi numbers! Regardless, thank you to all of the teams who competed and took our test. Even the ones who just took it on a whim and didn't even study, because that's really what the event is about: estimating and fudging numbers.

Thank you to my co-event supervisors, Andrew and Caleb, for being amazing test writing partners, coming up with some crazy questions, and putting up with my last minute problem writing.

Test Folder:
The exam and all other material can be found in this folder. The folder includes short write-ups for each question which we hope teams find helpful.

Fermi Questions B

Statistics:
Mean: 63.7 (25.5%)
Median: 65 (26.0%)
St. Dev.: 29.5
Max: 118 (47.2%)

Graphs:

Fermi_Questions_B-Distributions.png (37.75 KiB) Viewed 7147 times

More in-depth statistics and graphs pertaining to specific questions can be found at this link.

Thoughts:
The division B test was pretty much the division C test with 10 of the hard section questions taken out. Once again, many teams did not answer in Fermi numbers and had to be tiered because of it. I'm not sure how well Fermi questions can be run in division B, but I hope teams had fun with it!

Test Folder:
The exam and all other material can be found in this folder. The folder includes short write-ups for each question which we hope teams find helpful.

Test Feedback

If you have feedback for either test, feel free to leave it here! I would appreciate it a ton, since feedback helps a lot with gauging what I need to adjust in my tests. The test codes are as follows:

• Machines C: 2021BirdSO-MachinesC-Screw
• Fermi Questions C: 2021BirdSO-FermiQuestionsC-Chickadee
• Fermi Questions B: 2021BirdSO-FermiQuestionsB-Crane

[Giantpants]

Evenin' friends,

I had the honor of supervising Geologic Mapping at BirdSO! So you all know the drill for this one haha

Since I felt like my Geologic Mapping tests were getting kind of standard-ish (BEARSO and SOLVI), I tried to take this one to the next level. With lots of math and challenging thinking questions, I set out to make this my hardest Geologic Mapping test yet, and with respect to that goal, I undoubtedly succeeded. Did it make the test quality better? Personally I think so! The difficulty of the test combined with the thinking and general knowledge required to get answers is something I really like to see in tests, and I think my test did at least a decent job emulating that. As usual, I've gotten a lot of nice feedback already, so I'm very glad so many of you enjoyed it!

Distributions will be attached at the end of the post. The distribution was largely normal, and although scores were really low, the top teams were still able to break away and really display their skill at the event, particularly 1st and 2nd place who did awesome and achieved some super impressive scores! Particular congrats to Mountain View Piplup on an incredible win! And of course, congrats to everyone on a great job participating.

Quick stats:
Mean: 32.26 (22.56%)
Median: 30.21 (21.13%)
Maximum: 75.83 (53.03%)
Standard Deviation: 12.31

So uh as you can see scores were kind of low. But that's okay! As I said, I tried to place a heavier emphasis on thinking based questions and some geophysics in addition to the normal structural geology and depositional environment and map reading stuff that is standard to my tests.

Not too much else to say other than good job to teams on giving some thoughtful answers! For some of the question series (particularly the ones about Kayla and Tim's depositional environments as well as carbonate platforms!) teams had some great insights, so definitely be sure to go back and think about the answers on the key (which of course is being attached later).

Also, thanks for all the funny answers! Challenging teams to leave funny answers was definitely a great choice! And a little easter egg is the names of the people in the test, they're all event supervisors! See if you can find out which event supervisors are included.

Final Thoughts

If you want to leave feedback for my test (which I would appreciate a great deal!) feel free to at this link, and give as much or as little as you want! It's always helpful to hear what competitors thought of the exam, for next time of course! It would be really really appreciated, and I hope to get more responses (thank you to everyone who already submitted!). My contact information is included too, so feel free to reach out with requests for help on problems, or just to talk.

As always, the litany of thanks right here:

Thanks to all my dear event supervisor friends, who I spent days with in calls writing tests. Y'all made the process so much more fun!! Thanks to HugoTroop, ArchdragoonKaela, waterlubber, RobertYL, Umaroth, sophisSyo, zgs11, malikaow1004, jaspattack, TallMrGreat, axolotl, glaucophane, lavarball, and all the other incredible event supervisors who made the competition so much fun for me!

And of course, thanks to the incredible directors for designing what honestly has been such an incredibly professional and well run competition. sciolyperson1, builderguy135, Name, 404ic, legitimately, yoshiketchup, Klebb, Scythe, and all the other directors all worked so hard to organize such an incredible tournament, and I and everyone else. They were so fun to hang out with throughout the competition, and so kind and helpful!

As always, here is a link to my test, some pertinent maps, and distributions! I hope you find them useful for future preparation.

And if you don't feel like clicking that, here are the distributions anyway.

Thanks to everyone for participating in BirdSO! I had such an incredible time supervising and writing for the competition, as well as spending time with my dear friends the directors and event supervisors and I had such a fun time writing Geologic Mapping tests this year, so thank you all for the opportunity to help out the community in this way.

PS: To the 53.7% of yall who put "Geologic Mapping," I thank you for being on the side of justice.
To the 37.9% of you who put "GeoLogic Mapping," I don't know what to tell you other than that you are wrong (and even if the SO Inc. website says it, that website is wrong too! plus in other locations it says Geologic so... not a reliable source lol)
and to the 8.4% of you who put "Geological Mapping," who even are you haha

[Giantpants]

Hi all!

I was the event supervisor for Cybersecurity this past week at BirdSO along with 404ic, who wrote the programming/hands-on portion of the eventI had a lot of fun writing the test, and I hope you had a lot of fun taking it as well! This test was largely on the difficult side, with over half of participants putting it a “10” on the 1-10 difficulty scale. Unlike previous Cybersecurity tests I’ve written, I decided to go for a five-part exam, splitting up not only the cryptography portion, but also the web architecture portion into a multiple choice and a short answer portion.

We decided to make Cybersecurity an open-internet event, partially because there was a certain irony in only allowing cheat sheets in an event about cybersecurity, and partially because we thought it would be a good way for teams to be introduced to the event, especially in an invitational. As a result, I asked for a lot more questions which asked you to explain your thinking, which rewarded people for their in-depth knowledge, rather than simply googling. At the end of the test portion, I also included a google-a-thon, seven trivia questions that likely required some amount of googling.

This test was also incredibly long, at 94 questions (91 on Scilympiad). The most seriously answered questions by a team was approximately 70. This is arguably too long, but the purpose of a Science Olympiad test is to differentiate teams, not to give teams a passing score; by rewarding teams who were able to google quickly and proficiently, the length clearly helped the test serve its purpose.

The distribution scores ended up being not as right-skewed as I originally thought it would be, being very normal, other than an outlier at the top. All of the distributions can be found below.

Below are my thoughts on each of the sections/large questions on the test:

Section 1: Cryptography - Multiple Choice
At 2 points each, this section had 15 multiple choice questions. Many teams scored extremely high, with two teams getting a 14/15.

Section 2.1: Cryptography with Python
Being at the beginning of the test, I was extremely surprised by the lack of correct answers in this section. Every single question in this section could be solved with, at most, one function in Python, and a print statement. In the question where I asked teams to compute a large number to the power of another large number (mod 10^10), teams wrote answers saying that the question was “impossible”. This question was as simple as using Python’s exponentiation function, pow(), and inputting three arguments - the base, the exponent, and the modulus.

Section 2.2: The Vigenere Cipher
This question went extremely in-depth into the Kaisiski attack, one of the most common attacks on the Vigenere cipher. This was also a low-scoring section, being very math-heavy.

Section 2.3: Linear Congruential Generators and One-Time Pads
This section included the first two unsolved questions on the test, both involving scripting of some kind. I was hoping the 20 points would incentivize teams to solve it, but it was left unsolved.

Section 2.4: Vulnerabilities of the RSA Cryptosystem
The first two questions had surprisingly very, very few solves. This section gave an implementation of RSA in which the prime generation was obviously messed up (p and q are consecutive prime numbers) but only four teams earned points on the question that asked teams to identify the issue with the implementation. Again, no one solved the scripting question, which was honestly somewhat disappointing.

Section 2.5: Signatures and Authentication
Out of all the questions teams could have solved, they solved the ones in the math-heavy questions in this section, and I have no idea why. It wasn’t worth significantly more points, it was more difficult, and the questions required a lot more thinking.

Section 2.6: “Fun” with Finite Fields and Elliptic Curves
This section was the second math-heavy section on the test. I don’t understand why people went for 2.5 and not 2.6, but… sure I guess.

Section 2.7: The 𝔄ℭ4 Cipher
This section only had one scripting question that asked teams to break a stream cipher. No one solved this question

Section 3: Web Architecture - Multiple Choice
Like the cryptography/multiple choice section, this section also had 14 questions worth 2 points each, one being thrown out. A lot of teams earned a lot of points on this section (top score 28/28, 3 teams), and the distribution was extremely normal, other than one bar at 0.

Section 4.1: Web Development
Free points! This was arguably one of the easiest free-response sections on the entire test, with about ⅓ of teams solving each question in this section.

Section 4.2: Databases and SQL
Not much to say here, other than the fact that no one solved the question about SQL injections. The other questions had a good amount of solves.

Section 4.3: Cross-Site Scripting
A lot of teams earned points here as well. The last question in this section asked for each teams’ favorite type of cookie. By far the most popular cookie was the chocolate cookie/chocolate chip cookie - 22 teams answered this.

Section 4.4: Packet Sniffing
Another easy section - a lot of teams earned a lot of points here as well. Not much to say here.

Section 4.5: History of the Internet
This was the trivia section that asked competitors to research online. The questions here had an average of 27 solves, significantly more than any other section.

Section 5: Programming/Hands-On
While a lot of teams didn’t attempt this section, a lot scored extremely high - two people scored 70/100 points. This section ended up being extremely worth going for, as one could earn a lot of points in very little time.

Statistics
Mean: 72.5 (18%)
Median: 66 (17%)
Maximum: 197.33 (49%)
Standard Deviation: 38.3

Below are the score distributions of each of the large sections on the test:








Congratulations to samy-oak-tree and his partner for pulling away significantly from anyone else, earning 45 more points than the second-place team! I was extremely impressed by the amount of points they got on every section on the test, scoring the highest in the web architecture/short answer section and the second highest in the cryptography/short answer section.

Feedback
If you took the test, please take a few minutes and leave your feedback here! The test will be released soon™ along with the rest of the tests on our front page in Scilympiad.

Final Thoughts
On behalf of all of the tournament directors, thank you to everyone who helped out with BirdSO, including, but not limited to Giantpants, ShrewdPanther46, axolotl, CrayolaCrayon, Name, Agneg05, Galahad, lavarball, Arv101, Yoshiketchup, BennyTheJett, Klebb, Jaspattack, kh.aotic, BrownieInMotion, 404ic, bernard, Person, HugoTroop, Malikaow1004, SciolyPerson1, Lumosityfan, Mr.Epithelium, ArchdragoonKaela, F1shy, Hmmm, Lei0, Longivitis, Froggie, waterlubber, bearasauras, Pkota, pb5754, peroxisomes, RobertYL, sophisSyo, Umaroth, Basilosauridae, Legitimately, glaucophane, Willpan99, Will0416, Reverse, and Zsg11.[/b] This is obviously not a full list, and I'm sorry if I couldn't find your username on Scioly.org, but we're so glad you could join us this year. To the competitors, thank you for participating. We hope you can join us next year!

[YeagerTheCat]

Huge thanks to all the ran the BirdSO event and for all the detailed analysis for the events themselves post tournament.

Happy to see Mousetrap and Gravity run (trial or otherwise). Be interested in the break down for those...

[zsg11]

Hi everyone!

I, along with Camille, was one of the Co-Event Supervisors for BirdSO Forensics!

We wrote this test to be impossible to finish, so we were seriously impressed with the top teams! This test was a 529 point gauntlet of IDs, trivia, and application questions, which all culminated in a 129 written analysis.

I had a ton of fun writing this test, as I got to research and learn about topics and their Forensic applications, such as Ballistics, Palynology, and Mass Spec! I'm sure my search history writing this exam set off more than a few red flags, because what sane person is googling "Forensic Ballistics" and "How long does it take a body to decay" at 4 in the morning?

My personal goals in writing this exam were to be able to easily distinguish between 120 teams (we only had 3 sets of ties!) and to address topics that most other Forensics tests don't. I've rarely seen a Forensics test with actual applications to Forensic Science (i.e. "In recent years, DNA Analysts have been developing technology to predict the phenotypes of individuals based on small samples of their DNA. Considering the current biases within the American Criminal Justice system, how might this be beneficial?"). I think one thing lacking from many Forensics tests is the real-world applications, and I really hope you enjoyed those types of questions! Additionally, this was the first Forensics test I had ever co-written and I really enjoyed this experience because it allowed me to bounce ideas off of someone else and put extra time and focus into the sections that I wrote.

Now, for the statistics:

The average was a 23% on this test.
The max score was a 56% an the min score was a 0.9%.
The STD was a 13%.

Despite the low average, I was honestly impressed with how teams did on this test. I thought that it did a good job of differentiating teams while also having many "free" points on the test.


I don't normally do tournament ES reviews because I am a very busy person, but BirdSO was one of my favorite tournaments to help out with and I wanted to take a minute to thank all of the tournament directors for making this invitational possible. Writing/proctoring for BirdSO felt less like work and more like a group of friends hanging out and I am so happy that you selected me to write for one of my favorite events! I'm not going to mention everybody by name/username, but I did want to extend a special thank you to Giantpants for letting me fall asleep to his dramatic recitation of the entire Revenge of the Sith movie script on vc. This is certainly the weirdest way I've ever fallen asleep and I thank you, sir

Anyways, yeah! I really hope that you enjoyed my Forensics test:)

~Zoe
________
Contact information:
Camille: camillezhang2002@gmail.com or Camel#4485
Zoe: zsg5037@psu.edu or flowersforzoe#1145 (discord is preferred for me)

[Giantpants]

but I did want to extend a special thank you to Giantpants for letting me fall asleep to his dramatic recitation of the entire Revenge of the Sith movie script on vc. This is certainly the weirdest way I've ever fallen asleep and I thank you, sir

War! The Republic is crumbling under attacks from the ruthless Sith Lord Count Dooku. There are heroes on both sides, evil is everywhere.

In a stunning move, the fiendish droid leader General Grevious has swept into the capital and kidnapped Chancellor Palpatine, leader of the Galactic Senate.

As the Separatist Droid Army attempts to flee the besieged capital with their valuable hostage, two Jedi Knights lead a desperate mission to rescue the captive Chancellor...

Knowing the entire Revenge of the Sith script by heart is a unique trait of mine but one that I love to pull out every now and then for the meme

[Giantpants]

but I did want to extend a special thank you to Giantpants for letting me fall asleep to his dramatic recitation of the entire Revenge of the Sith movie script on vc. This is certainly the weirdest way I've ever fallen asleep and I thank you, sir

War! The Republic is crumbling under attacks from the ruthless Sith Lord Count Dooku. There are heroes on both sides, evil is everywhere.

In a stunning move, the fiendish droid leader General Grevious has swept into the capital and kidnapped Chancellor Palpatine, leader of the Galactic Senate.

As the Separatist Droid Army attempts to flee the besieged capital with their valuable hostage, two Jedi Knights lead a desperate mission to rescue the captive Chancellor...

Knowing the entire Revenge of the Sith script by heart is a unique trait of mine but one that I love to pull out every now and then for the meme

This is where the fun begins

[Longivitis]

Hello, y'all!

I had the opportunity to supervise Game On at BirdSO! Initially, the original topic I was planning to use was speed and velocity, but I decided to switch it to aerodynamics at the last moment because I felt that it would allow more creativity and birds are aerodynamic. I also gave teams an extra bonus point if their game was related to birds.

Some Stats
Mean: 60.55
Median: 64.01
StDev: 25.16
Max: 98

What many teams missed out on easy points on were on commenting their code (Game Mechanics Part H) and adding sounds to their game (Game Play Part C). These two sections are relatively easy to implement and only take a minute or two of your time, and every point counts, especially when most of them come from me just scanning your code. Custom sprites are also your friend since they give you more points even if you draw a simple circle.

The trickiest section would have to be play balance (Game Play Part D) since a good number of teams did not implement multiple levels or have sprites' speed/movement change throughout the gameplay. It's important to read the rubric and understand what exactly it's trying to ask for, and I go a little more in-depth in what I was looking for when tests and keys come out.

I also really want to stress that everything in the game file must be either from the built-in Scratch library or created by you or your partner DURING the testing period. During an in-person tournament, we would simply have you guys use the offline editor with the internet disabled, but since it's online it's harder to enforce. Let this be a reminder for any future tournaments that you guys will participate in.

Overall, I really loved playing all of your games, and I would like to express my thanks to the tournament directors and the other event supervisors on making BirdSO an amazing tournament!

[vehicleguy]

Orni: 10/10 Really well made as expected. A lot of the stations required critical thinking, analysis, and application. Couldn't ask for much more.

Sounds: 10/10 I'm not that qualified to talk about sounds since I'm bad at it, but it seemed like the best sounds test I took all year. Wide variety of topics and questions, and some of them really made my partner and I think.

Chem and circuit were also good, but I soloed them so I barely got to experience each one.

[Jjshan26]

Astronomy (3): 9/10 Can't really speak for the DSO section, but astrophys was quite fun! Much appreciation for the ~common theme~ of all the questions lol. My eyes are opened and I will never watch JJK the same way again

Designer Genes (1): 9/10 Whee this was probably the second most panicked event because I didn't realize how long it was. The first part of the test was pretty standard designer genes stuff with a couple interesting combo heredity questions, but the second half I was absolutely scrambling with like 15 min left. Also read like three wikipedia pages and learned about heritability while I was taking the test so that was definitely an interesting experience (loool the wonders of open internet).

Protein Modeling (15): 8/10 Test had a lot of interesting molecular stuff that was clearly too big brain for me Jmol was also pretty unique with the ending questions; albeit there was some memorized content, other stuff was doable through logic, which is always fun. I personally wasn't a big fan of putting a paragraph from a paper at the end of the test though, that was pretty painn.

Fermi Questions (4): 8/10 I'm actually really glad that I got to take another Fermi test before I graduated--it's been a long two years since rip fermi As expected, my memory of random facts has unfortunately kind of died, so there weren't that many direct freebies on the test. I personally prefer questions that are BSable over those that require you come in with really niche knowledge anyway, and most of the test did fit this bill, so kudos to the writers. ngl 46-60 were kind of insane tho not sure how we're supposed to know (10!)!? and yea there were a couple of obscure units questions but overall had a really fun time. Thanks

Science Quiz Bowl (3): 9.5/10 I'm kinda biased here because I've just never done any buzzer competitions before so this was super hype and nervewracking for me, especially finals. It was a lot of fun bsing the entire chemistry section of the written exam with my partners and anxiously deciding whether I should buzz in the discord chat (neg gang). My dearest condolences to my parents for clowning myself on live zoom

Overall (4): 10/10 While it definitely is still a long shot from the experience of being in person, huge thanks to all the TDs and ESes for putting together such a fun tournament!! Though I must say, you sure have increased the friction of making appeals my laziness has taken precedent oops