Hello! We are the Strategy department from Team 3128, The Aluminum Narwhals, from San Diego, CA. As we did last year, we will be sharing our thoughts about the game and what our department has been doing each weekend of build season!
Offseason Review:
This offseason, we prioritized teaching our new members everything about Strat from basic fundamentals to scouting systems. We went to three offseason competitions: Battle at the Border, Tidal Tumble, and Beach Blitz (and were finalists!). During these regionals, we wanted to give our new members competition and leadership experience during competitions. Furthermore, we participated in the Scouting Palooza and were the Data Visualization Award and Overall Best Team Award winners. Altogether, this offseason was one of the most productive that our department has seen so far. For more on the department, see our recently published whitepaper.
Week 0/1 questions:
How many game pieces will be left on the field?
Will there be enough game pieces left on the field from missed speaker shots to have some cycles be half-field vs. full-field cycles?
Can Human Players score high notes in the Speaker?
Do amp points scored in auto count toward amplification during teleop?
How many robots can score in the trap?
What will the ratio of trap and speaker robots be?
When should the hp press coopertition button?
What will the most promising structure of an alliance be? (ex. Amp-defense-speaker, speaker-speaker-amp)
How difficult will harmonizing be? Will robots collide with each other?
Pitscouting Points:
Drivetrain Type (Tank, Swerve, Meccanum), Robot Weight, Motor Type, Vision Capabilities
These data points supply technical information about the robot which allows us to more precisely estimate a robot’s capabilities. For example, having a heavier robot weight or more motors would allow it to perform better at defense.
These data points allow us to better understand the robots capabilities which is important for making early match strategies when other data hasn’t been collected.
Scouting Data Points:
Leave, Auto Amp, Auto Speaker, Auto Miss A, Auto Miss S, Tele Amp, Tele Speaker, Tele Miss A, Tele Miss S, Average Cycle Time, Trap, Spotlight, Climb, Park, Defense Timer, Forced Misses, Oof Time, Penalties
These data points were created for our scouting app to improve the quality and consistency of the data that our scouts collect during the competitions.
Once the game was released, Strategy created a list with all the possible tasks the robot can do and prioritized the tasks, inspired by 4481’s Skills List. Soon after, we made a kahoot for the rest of the team to play, so that they understood the game better, and shared our priority list to get their input on it.
Afterwards, we decided on what is feasible to complete for a Week 1 and Week 3/4 through a functional requirements spreadsheet that we will continue to update throughout Build Season. In making our functional requirements, we specified how fast and how reliably the robot should be in doing the task, so that technical departments could better decide on how to reach those goals. We made sure to not specify how the robot should be designed to perform the task to give the technical departments the flexibility to create what they think will be the best way to perform those tasks.
Later, we brainstormed as many different robot archetypes as we could, and analyzed how different combinations of these would work together in a match. This list enabled our cycle analysis on what cycles we predict different robot archetypes will complete with theoretical point counts.
We created a Defence Manifesto that highlights strategies specific to defence.
We created an Auto Priority List for our Controls department listing what auto paths we believe will be most beneficial and feasible to have for Weeks 1, 3, and 4.
We also analyzed Crescendo in the lens of past FRC and FTC games, finding parallels to Ultimate Ascent’s game pieces, Deep Space’s Cargo Ship scoring, Rapid React’s endgame climb, among others.
We also created a Drivers Guide for Drive Team with strategic priorities, as well as important rules to follow.
Predictions about game play:
Autos are very important, can get ~30-40 points/auto which can significantly alter the outcome of a match
Defence looks much more viable in this game as opposed to Charged up, as a lot of teams will look to reduce cycle times and shoot from a distance to score in the speaker. Because of this, we predict blocking shots will be quite prevalent and with minimal protected zones, defence seems like it will be quite effective.
The Ensemble (climb) RP looks to be quite difficult as without a trap mechanism you either need to have 3 robots climb (at least 2 on 1 chain), or have 2 robots on the same chain with a spotlight. If you aren’t going for a trap mechanism, you would need a well trained human player for spotlighting.
The Melody (scoring) RP seems to be much easier to obtain than the ensemble RP. We expect lots of scoring this year and feel that scoring 18 (15 with coop) notes will be met in a large number of matches.
We expect that the ability to score in the amp is very important this year as the 2.5x score boost represents a large number of potential points.
With a carrying capacity of 1 note, the ability to score accurately will be very important in this game as missing a note can cause a large time loss which will have a large impact due to quick cycles we believe will be common this year.
Overall, Crescendo seems to be much easier mechanically than Charged Up (minus trap) and we expect to see a higher level of play across all teams, especially seeing how successful kitbot and the Ri3d bots seem to be.
This is some really good analysis, thanks for sharing!
One aspect I find interesting from a game design standpoint is that the Amp kind of has two difficulty modes:
Easy Difficulty: if you’re a lower resource team, and want to make a typical “low goal robot”, it’s totally doable with a static robot.
Medium Difficulty: However, scoring in the low goal becomes more challenging if you are building a robot that is short enough to go under the stage. It either needs to accurately launch into the Amp, or extend upward in a way to score it.
I’m curios what the typical low resource robot will do in this game, Amp or Speaker, or both. Certainly the kit bot being a Speaker launcher probably sways it in that direction.
Hello! We hope you all are having a great build season so far! This is what the strategy department has been up to this week:
Scouting Systems:
We are almost done with our scouting app, and have made progress on our backend and data visualization app.
What Strategy has been doing:
We simulated a rule removal paper game, where we strategized for different robot archetypes based on the designs we observed from Open Alliance teams! In addition to being a good strategy exercise using realistic robot capabilities based on real-life teams, the rule removal aspect created a fun and lighthearted atmosphere where we could come up with nonsensical and insane ideas (G101…)
We have created a spreadsheet for cycle analysis that predicts both point totals for different robot archetypes and average theoretical cycles that consider driver experience and speed (inspiration from team 95 The Grasshoppers!)
We have also refined our functional requirements, modifying our requirements to ensure that we create a robust robot, over one that is multifaceted but inconsistent.
We are working on a video series that briefly describes parts of what we do in strategy! We have just finished drafting the script for the first few videos and hope to film next week!
Good analysis, but i have a few questions, What was spotlight? What does Oof Time mean? And what does park have to do with this game? What do you mean by motor type?
Spotlight is the term for a human player successfully landing a high note on the microphone during endgame, as named in the game manual. Similarly, robots can park in endgame rather than climbing by entering the area under the stage. We believe these data points to be valuable to strategizing (ex: if we have seen a human player spotlight in the past, we may benefit from putting that same human player at the amp). Oof Time measures the amount of time a robot has had an aspect of it fail on the field (ex: drivetrain malfunctioned, arm stopped working). If we observe that a robot has a lot of Oof Time, we can make a more holistic picklist. As for motor type, during pitscouting, we ask about the type and number of drivetrain motors a team uses, mainly to understand the general speed the robot may perform at.
Hello! This is the third weekly update from team 3128’s strategy department. Here is what we have been up to:
Scouting Systems:
We are even closer to being done with our scouting app, and have made progress on our backend and data visualization app.
What Strategy has been doing:
We have done several human games this week, modeling different alliances and testing how different robots work together. Here are our findings! Human Game Takeaways
We have been coordinating a scouting alliance for each of our regionals, and have developed a handbook detailing how to operate our systems.
We checked in with our team’s Mechanical department to ensure that their subsystems meet our functional requirements, and have refined them accordingly.
We have continued working on our strategy themed video series. Due to unexpected circumstances it has been delayed but we still hope to film next week!
Hey team 3128! In an effort to increase interoperability between scouting apps, we recently announced The Purple Standard (TPS), a unified, community-driven standard for FRC scouting data (see https://chiefdelphi.com/t/the-purple-standard-a-unified-and-community-driven-standard-for-frc-scouting-data/449394 and https://thepurplestandard.com), paving the way for collaborative scouting across apps (which benefits smaller teams and rookies). TPS is modular and completely customizable based on the data you want to collect in your scouting app. Would you be interested in integrating TPS into your app to join the wider network of collaborative scouting data?
This is nearly irrelevant. As of 2024 R502 (frctools.com) says that you are allowed at most four drivetrain motors and there aren’t really any robots that will have less than four (swerve has four tank has four and mechanum has four)
Drivetrain Type (Tank, Swerve, Meccanum), Robot Weight, Motor Type, Vision Capabilities
My wording was a bit misleading. Last year, we asked about the number of motors on a robot’s drivetrain, but in light of this new rule, we altered the question to ask about motor type, rather than number of motors, as shown in our pitscouting points. Thanks for catching that!
Hello! Here’s what 3128’s Strategy department has been up to this week:
What Strategy has been doing:
We scouted matches from the Kettering Mid-Build Season Competition this week! Here are some of our takeaways:
Compared to Charged Up, the playing field seems more leveled with the relatively high skill ceiling that kitbots have.
Missed pieces at the source (on alliances with robots without ground intakes) can be capitalized on.
We have edited some videos from our Strategy video series and we hope to have them posted by the end of the week!
We are planning a Week Zero Watchparty with some of our Scouting Alliance teams this Saturday, where we’ll be scouting both Week Zero and Blue Twilight Week Zero Invitational! Thanks to both of these teams for putting on these events, we’re looking forward to posting some of our strategic takeaways after seeing gameplay!
Scouting Systems:
We worked on our data visualization app, and continued programming features into it.
We finalized the first version of our scouting app!
We also implemented security into our scouting systems and expanded our scouting alliance!
Lady is one of our archetype names! We refer to an archetype that scores in the trap and amp as “Tramp” and “Lady” just stuck when referring to a robot that’s its inverse, an archetype we’ve defined as one that can score in the speaker and climb.
Hello! We hope you’re having a good week! Here is what Strategy has been up to this week:
Strategy:
Our Strategy video series will be posted imminently!
We also made a picklist template prior to our first regional by identifying things we would be looking for in an alliance, specifically:
Compatible auto paths
Robots that complement our weaknesses in each part of a match
Scenarios in which we would prefer a triple offence alliance to an alliance with 2 offence and 1 defence bot
We had a great time scouting Week 0 and Blue Twilight Week Zero today! Here are some of our takeaways:
Having a good auto can give an alliance a huge advantage, similar to how important autos were in 2018.
Due to how auto paths are more abstract this year than in 2023, compatibility with autos might be a greater defining factor than a high number of auto pieces scored.
Scoring in the trap might be harder than expected. Because it requires a complicated mechanism and likely would eat into the number of cycles able to be made, it might not be worth it unless a robot has a very reliable, fast trap mechanism.
Defense wasn’t as frequently played as we had initially expected, as the alliances were more focused on outscoring one another.
We were also surprised by how amping could completely change the outcome of a game. If robots were synchronized enough to score multiple pieces in an amped speaker, they could compensate for a subpar auto compared to an opposing alliance. We feel that coordination regarding amping has the ability to make or break matches.
Scouting Systems:
We have continued working on our data visualization app. We’ve also updated our pitscouting form to include a picture of drawn out auto paths! With how variable auto paths can be in this game, we believe this data will be vital to picklisting and match strategy.
We tested out the final version of our scouting app with our scouting alliance, who joined us to watch and scout Week 0 and Blue Twilight Week Zero!