Summary
It would be helpful to have a guide for how people can contribute to the repository. This would include including how to set everything up locally in the README, and more detailed information on the design / architecture of the website as part of the Wiki

Stat hints don't work great on mobile. Additionally, stat hints should be displayed in the sortable table

Summary
BBQ is currently only computed by checking if a given banner is won strictly prior to the event start date. This means that most of Einstein results are slightly wrong due to banners being awarded the same day as Einstein field.

The objective of this issue is to fix the problem by allowing banners to be won on the same day as the given event provided that they were not won at the given event. This is an update to the python scripts and should have no frontend impact.

Summary
RIBS and BRIQUETTE are not calculated intuitively. This is most obvious in the case of RIBS, where later events in a season are almost guaranteed to have higher scores because 1.x seasons are considered in the calculation. It also doesn't make much sense to only count a part of the previous season. Perhaps RIBS should be a weighted interpolation between seasons where the previous season's banners all count equally to each other, but are linearly interpolated out as the current season progresses. For example:

t = normalized progress through current season
RIBS = (1 - t) * prev season banners + t * cur season banners

Team A
2024 (going into week 3, approximately halfway): 1 robot banner -> RIBS = (0.5 * 3) + 1 = 2.5
2024 (going into week 1): 0 robot banners (it's week 1 after all) -> RIBS = 3
2023: 2 robot banners

Similarly, for BRIQUETTE we could do a weighted average over the last 4 seasons that fades out the least recent season for the active season over the course of the active season. So at the beginning of the season we'd have:

BRIQUETTE = 0.25*season1 + 0.25*season2 + 0.25*season3 + 0.25*season4 + 0*currentSeason

Then over time the value from season1 would transfer to currentSeason linearly as shown above

Work Scope

• Add interpolation of seasons to RIBS and BRIQUETTE stat calculations in bbq_stats.py
• Update About page to explain this calculation

Summary
TBA API lists teams as being on Einstein field if they are competing for team attribute awards. Likewise, teams on Einstein due to the robot may not be in the running for a team attribute award. The objective of this ticket is to make that distinction in the data calculations for Einstein, as well as in the display of the teams in the team list on the front end.

Work Scope

• Improve the data update scripts to only count team attribute awards for eligible teams (and likewise for the robot) on Einstein fields
• Update the front end to distinguish between teams in robot/team attribute/both sections (if both, they should probably get special recognition because that is impressive)

Summary
In order to better highlight teams' accomplishments, awards that are won at different levels should be displayed slightly differently. The objective of this ticket is to specify the theme of the blue banners appearance on the site based on award level (district, regional/district championship, worlds)

This will help set the foundation to add stats that weight harder-to-win banners heavily, as well as possibly allow for an offseason tier of awards to be shown

Work Scope

• Award appearance based on competition level proposed and acceptable
• Changes implemented on site

Teams that have inactivity probably should not take a hit to their BBQ or their SAUCE, since they were not attempting to win blue banners during a hiatus / retirement. To reflect this, BBQ and SAUCE should be recomputed for teams that have periods of inactivity or who have ceased operations for the foreseeable future

Summary
When event data gets updated, team appearances should be added/deleted appropriately. For 2024cmptx, the teams initially were every team that made worlds, but then TBA changed it to be the 49 teams on Einstein during the event. The Appearances remained stale though so while the stats were updated correctly, the display was not.

Summary
Unit tests should be added and integrated into CI in order to ensure there are no regressions with the python data loading tools

Summary
Supabase auto creates documentation for interacting with the database. For ease of use by other developers, these docs should be linked to this repo via the README or some other means.

The percentile for a team/event (and/or rankings) would be interesting to see on the respective page.

For events:

• percentile / rank all time
• rank in its competition year

For teams

• percentile / rank all time
• percentile / rank in the last 4 years
• percentile / rank in current season

The website is not amazing on mobile. In particular:

• the nav bar is semi broken
• things don't fit in their containers
• dropdowns don't show everything (too small somwhow)

There are likely many other issues to fix

Currently all testing is done in production. The objective of this ticket is to make a staging area for data script testing.

It would probably be best to have a cloned schema on supabase named "staging".

Summary
Vercel is only free for one user, so the purpose of this ticket is to determine if github pages can be used instead to run the app

Work Scope

• Try to get a prototype working on github pages
• Add deploy and preview actions for github pages
• Remove vercel from project if these things work well

When I look at the team info page for team 868, I see a list of awards that are up to date with District State Champs in the A1 Awards section. These awards include several event wins that should contribute to the banner count. However, the "Banners won in 2018" field reads zero.

Summary
The data should be updated nightly in Github actions rather than manually from the terminal.

Work Scope

• Update the python script to accept credentials on the command line so Github actions can update the database
• Set up a workflow to run the award updater in github actions

Summary
It would be interesting to highlight the outcome of an event in terms of blue banners awarded. This can be added to the event page somewhere.

Summary
The old BBQ FIRST had rankings pages for each metric, with various filters. These were all in different pages and much of the code was duplicated.

The objective of this issue is to implement an event ranking page that filters on: year (or since year), team vs. robot BBQ, official vs unofficial event. This should show the top N events (probably 100). Ideally this would also be sortable by the user in order to show the events in ascending/descending order as desired.

Summary
The old website had a page that explained what all the stats were and why the website was created. In the new site, we should re-create that concept.

Some possibilities:

1. Make a section in the homepage an about section, and have a nav bar link that auto-scrolls the page to the relevant about section
2. Add an about page that stands alone and is linked in the homepage

Summary
The original website had the following stats:

• BBQ (we have this one)
• BRIQUETTE - basically BBQ but only counting banners since 2005, when the 3 team alliance became the baseline in FRC
• SAUCE - BBQ only considering the last 4 years
• RIBS - BBQ only considering the current and previous season

The objective of this ticket is to add these stats back to the website

Work Scope

• Update the EventData table in the database to include columns for these stats (for team and robot)
• Update the python scripts to run the correct calculations for these metrics
• Make a first order update to the front end to show the stats on the event page and event rankings page

Summary
The objective of this ticket is to build up a system that predicts each team's chances of making worlds for the current/upcoming season based on their historical results and their path to championships.

Prediction
Districts advancement would be computed via Monte Carlo based on events remaining to determine district championships attendance / final district points / awards.
Regional advancement would be computed by simulating all remaining regional events and compiling the outcomes for each team

Modeling
We could use Elo or some sort of Bayesian update model to build a system for predicting event results. We can lean on blue banner history at first since that is the theme of this website.

Database Updates
The database will need to be updated to include additional team modeling information (if any additional information is needed besides blue banners).

UI Updates
The UI needs to be updated for each team to include the predictive stats. Each event should have a list of favorites to win as well.

Summary
Each team should have a BBQ associated with the blue banners they have won per year of existence. This needs to be added to the python scripts and the database schema will likely need to be updated to add a TeamData table

Summary
If the event ranking view is set to filter years 1992-2004, no events show up. However if the upper limit is raised to 2005, events appear and are ranked correctly.

This is likely due to SAUCE not working for events prior to 2005, however it does impact all event metrics.

Work Scope

• Fix filtering logic to allow events pre-2005 to appear
• Auto-disable SAUCE if the top of the range is below 2005, and do not allow it to be selected

Summary
The search bar currently doesn't do anything if you hit enter without first pressing the down arrow. This was initially done by design, but after testing more it would be better to always have the first result highlighted unless the user moves the autocomplete focus with the arrow keys manually.

Work Scope

• Set the default starting autocomplete index to 0 rather than -1
• Test that this does not regress the autocomplete feature

Summary
If the data loader python script fails to an exception, it will simply hang in the console and need to be force killed. This is not good for GitHub actions usage as the script will sit there for 6 hours until GitHub times out the workflow. This is partially due to the supabase client not destructing gracefully, and also due to a lack of exception handling throughout the code.

The objective of this ticket is to address the problem in the short and long terms

Short term solution

• Make the github action workflow timeout shorter in order to protect against script runaway

Long term solution

• Add exception handling to joblib parallelization
• Add exception handling to any third party API calls
• Figure out how to destruct the supabase API client correctly

Summary
There's a lot of duplicate code in the team and event rankings pages. The objective of this ticket is to heavily refactor those views to be as general as possible, and to unify them with a single library of functions

Work Scope

• Refactor repeated code into a library for team and event rankings
• Create a library for handling filters and use that in both pages
• Create components for filters rather than having them be copypasta'd in both files. These components should be configurable.

Summary
The objective of this issue is to allow users to interactively rank teams based on team BBQ, total banners, robot/team banners, etc. The users should be able to filter based on those fields, as well as "since year" and country/state.

This depends on #11 for Team BBQ, but the rest of the implementation is possible without that issue

Summary
The table rows in rankings pages do not handle ties at all. The objective of this ticket is to improve table row rankings to recognize and group tied teams/events with an obvious indicator (such as T-X, where X is the rank the items are tied for) and then pick up the numbering appropriately after the tie.

For example:
1
T-2
T-2
4

Work Scope

• Process rank ties as part of event/team rankings updates
• Update event/team rankings pages to embed ranking in the list of top 100 teams/events

Summary
Create a first draft event page, similar to the team lookup page

Work Scope

• Replicate the team page + search view, but for events
• Update database schema to have an Event with BBQ related stats
• Update database schema to have a Team with BlueBanner stats
• Update award python script to update the Team blue banner counts in the new schema
• Create python scripts to aggregate event data and stats based on attending teams and their awards entering the event
• Use TBA API in javascript to associate teams with events

Championship events are incorrectly listed as Week 1 because they are not affiliated with a particular week. The objective for this issue is to add checks for championship event types and replace the week string with a championship string instead

Summary
A pie chart of how teams contribute to an event's overall banner count (for robot and team banners respectively) would be interesting to see

Summary
Currently the search bar must be team number or event code (despite what it says) and is dependent on which page it is on. Ideally, this search would be able to take in things like team/event name dynamically, and be added to the title bar.

The search bar should also show suggestions that the user can select to search immediately (with click and arrow keys)