What's New

Adding a class to the catalog is right for something you’ll run all season and pure overhead for a Saturday-only exhibition. An event can now run a one-off class defined at the event itself, with no catalog rows created and no clutter in the class list for every other event on the schedule.

Two fee bugs were producing wrong quotes. Fee rules were being matched against the vehicle’s own type rather than the type of the class it was entering, which gave the wrong fee for anything running outside its usual category. Applicability is now judged by the class entered. Separately, a season fee that had been charged but never paid was treated as settled and never appeared again — unpaid season fees are now quoted until they’re paid.

On the cleanup side: two registrations for the same vehicle can be merged into one record, dropping a participant from a class is now an explicit action recorded in the audit log, and in-progress class results can be reset and restored as a unit rather than unwound one placement at a time. Standings sorting was corrected for competitors running multiple classes on the same day.

Every event ends with paperwork somebody assembles by hand — what the office needs on a clipboard, which vehicles are entered in more than one place. Those are becoming generated documents, scoped to one event.

A reusable layout and stylesheet for printable reports landed first, so each document comes out consistent and actually fits a page. The split vehicle registrations report is built on it, and the shared layout makes the next one a small addition rather than a fresh design.

A TV showing a results slideshow is useless in the moment you need it most — weather coming in, a track hold, a vehicle that needs moving. Broadcast lets you write a message, pick which displays it hits, and override their slideshows. It started emergency-only and now runs in two modes, because “concessions close at nine” deserves the screen too, just not all of it.

The display pages themselves were hardened for unattended TV kiosk operation, so a screen you set up in the morning is still right in the afternoon. Slideshows switched to a constant linear scroll, so a class with forty entries reads at the same pace as one with six, and the current results view had its typography reworked for reading at distance. Participants who haven’t pulled yet are now correctly excluded from posted results.

Support that starts when a customer notices is support that starts too late. The new event:monitor command runs on a schedule, checks live events against their time thresholds, and raises an alert when one looks stalled.

Operational alerts route to Slack through a dedicated alerter service. Queue health monitoring now accounts for both backlog depth and worker heartbeat freshness, with a grace window so a brief blip doesn’t cry wolf, and Redis connections gained explicit timeout, read-timeout and TLS settings so a slow network fails fast and visibly instead of hanging.

Fairgrounds are exactly the places where phone service goes to die, which is a problem when the results everyone wants are on a phone. The app caches results and schedules on the device, so the last known standings are still there when the bars hit zero at the back of the pit.

Fans get live results as classes are scored, pulling order by class, event schedules, driver and vehicle profiles, real-time standings, and dark mode for night pulls. Everything comes straight from the platform — if your event runs on Tracker Pull it’s already in the app, with nothing extra for organizers to publish. The pull order display picked up a QR code linking to it.

Grab it from the App Store or Google Play.

An organization running across more than one timezone had to mentally offset every posted time. Events now store their own timezone. Confirmation emails became an event-level toggle for organizations that handle their own communication, and every notification the platform sends now renders in the branded layout rather than plain defaults.

Camping sites were modelled as one site, one registrant, which isn’t how families and teams actually use them. Sites now carry a maximum number of registrants and an owner phone number, with site search moved onto the registration form. Fee applicability picked up an explicit event-scoped opt-out row, so exempting a single event from a standing fee rule no longer means editing the rule itself. When a schedule shifts, event class dates can be moved in bulk instead of one edit per class.

Reviewing a bulk import used to mean one page showing unmatched people, unmatched vehicles, class errors, fee questions and optional corrections all at once. It now walks through dedicated stages — Blockers first, then Persons, Vehicles, Updates, Fees, and a Summary — each with its own create-new guard so resolving a near-match can’t spawn a duplicate record. Name and hometown corrections are opt-in per row, fee defaults refresh against what you just resolved, and stage navigation no longer marks a past stage complete while it still has pending items. Duplicate detection runs during import so repeat rows get flagged rather than written twice.

Managing a paid registration was rebuilt alongside it. Fees are estimated live as you build the entry, including for vehicles that don’t exist yet, so you see the number before taking the money. Person and vehicle search now spans every organization rather than just the active one, and search sits on the registration form itself. Seasonal payments got their own routes and management screens. Recalculate Fees is back on the edit page, and a validation error no longer wipes your selections.

Underneath, both the public and admin paths write inside a transaction with race-safe pull order assignment, and a database constraint enforces one active registration per vehicle per event. A new vehicle created from the admin form is written atomically with the registration. Confirmation email failures are non-fatal — the entry saves regardless — and closed registration windows, rate limits and permission errors render friendly public pages instead of raw error codes.

Made it easier to browse through and match people and vehicles that could not be automatically matched. Class name parsing improvements, and an improved UI that identifies required corrections versus “nice to correct”.

Ensured that system formulas are only editable by global administrators. This ensures that users cannot edit system-level (shared) formulas used across all organizations. Organization scoping is now applied correctly to users who are members of multiple organizations, including global administrators.

Less per-class clicking when you set up the season. Configure your default scoring formula at the organization level and every new event class inherits it — override only the exceptions. Weightless and exhibition classes get their own slots so they don’t sit awkwardly inside the standard rules.

Importing dozens or hundreds of pre-season registrations is the kind of thing that should be a coffee break, not a weekend. The expanded importer now handles event class lists in the same flow as participants, with optional toggles for seasonal fee allocation and exhibition formula assignment per row.

The validation pipeline runs in three stages — parse, validate, preview — before anything is written, so you see and fix issues before they become rows in the database.

Operational confidence work. If results aren’t sending or emails aren’t going out, we know before you do.

Every admin account can now enroll any TOTP authenticator app (Google Authenticator, 1Password, Authy) with recovery codes issued at enrollment and regenerable on demand. New accounts must verify their email before they can act. Every admin write is recorded in an audit log with who, what, when, the field-level diff, and the request context — searchable and filterable from the admin panel.

When a competitor calls asking where the registration confirmation went, you can pull up the message timeline and answer them — no more digging through provider dashboards. Hard bounces and spam complaints are added to a suppression list automatically so your sender reputation doesn’t quietly degrade.

Maintenance mode flips on from inside the admin so visitors see a “back shortly” notice instead of a half-deployed app. The health dashboard reports the status of every infrastructure component on one screen — green means run the event.

The data is going somewhere — now you can see where. Useful for sizing infrastructure before a big event, spotting integration issues, and proving to sponsors that the audience is real.

The single-file upload flow was the bottleneck. Bulk upload + approve queue means you can dump 200 shots from the event, walk through them in the admin, and publish what’s ready.

Every pulling organization scores a little differently and that’s been the friction blocking adoption. The Formula Builder makes the rulebook executable — write your scoring once in a small DSL, version it, and let standings compute themselves from there.

Past seasons stay frozen against their formula version. New seasons get a new version. Rollback is safe. Tie-aware ranking, weightless and exhibition class support, and context-aware DQ handling are all first-class.