What an instructor can see — and what they can't
An instructor who runs a biaX cohort gets four panels and two tables. Every figure is a group statistic. There is no screen, anywhere in the console, that shows one student's trades, P&L, or emotion tags — not to the instructor, not to us on request.
The screens below are a static copy of the real console, filled with a constructed cohort. Nobody on this page is a real student and no figure is anyone's record. The layout, the wording, the rounding and the withholding rules are transcribed from the shipped screens.
Seats
Seat accounting is always visible — it is not a behavioral statistic, and an instructor with three students still has to see what they are paying for.
Sample — Autumn options cohort
Participation
As of Nov 1, 2026, 21:00 UTC
Twelve seats are held but eleven students were measured this week — a student who recorded nothing is not counted as a measured zero. The date line is there because this is not live: cohort statistics are computed once a day.
Weekly consistency
Eight weeks, oldest first. Two of them have no bar at all.
The first two weeks are drawn as an empty track with no bar inside it, not as a bar of height zero. A zero-height bar would say “nobody traded that week.” The actual fact is “fewer than five students were measured that week, so the figure is withheld.” Those are different sentences and the chart is not allowed to blur them.
Most-broken rules
Rules the students set for themselves, scored against what they actually recorded.
- Daily Loss Limit 45% · 5 of 11 scored
- No Impulsive Averaging Down 36% · 4 of 11 scored
- Max Trades per Day 25% · 2 of 8 scored
Three rows, not eight. The catalog has more rules than this, and the rest are not at zero — they were scored on too few students to report, so they are removed from the list rather than shown at 0%. An empty list here never means “nobody broke a rule”; the console says so in those words.
Students
13 · 12 / 25 seats
| Name | Joined | Seat | Actions |
|---|---|---|---|
| Alex | Sep 7, 2026 | active | Remove |
| Jordan | Sep 7, 2026 | active | Remove |
| Priya | Sep 8, 2026 | active | Remove |
| Marco | Sep 8, 2026 | active | Remove |
| Yuna | Sep 9, 2026 | active | Remove |
| Student | Sep 9, 2026 | active | Remove |
| Dana | Sep 14, 2026 | active | Remove |
| Omar | Sep 15, 2026 | active | Remove |
| Lena | Sep 21, 2026 | active | Remove |
| Sam | Sep 28, 2026 | active | Remove |
| Chris | Oct 5, 2026 | active | Remove |
| Nour | Oct 12, 2026 | active | Remove |
| Riley | Sep 7, 2026 | left | — |
Four columns, and there will never be a fifth. No “last active,” no streak badge, no sort-by-activity — because in a cohort of five, sorting by activity tells you which of the five the group average is about. The one student with no display name shows the word Student; the console does not substitute an email address or an ID. The row marked left stays visible, because who released a seat is seat accounting, not behavior.
Invite links
How students get in. One link, a use count, an expiry — no email collection, no roster upload.
| Link | Status | Uses | Expires |
|---|---|---|---|
my.biax.app/join/SAMPLE-NOT-A-REAL-CODE |
Active | 12 / 25 | Nov 14, 2026 |
| — | Revoked | 3 / 10 | Oct 20, 2026 |
The revoked row shows no URL at all, not a greyed-out one — a link that can't be used shouldn't be copyable by accident. Revoking is not retroactive: students who already joined keep their seats, and the console says that on the confirm step.
What this console never shows
The part a product tour usually leaves out. These are enforced on the server, so they hold no matter what the instructor clicks.
- Any individual student's trades, P&L, holdings, or emotion tags. There is no screen for it and no export. An instructor who wants to see one student's record has to ask that student to show them.
- Anything at all, until five students have been measured. Below that, all four Participation figures render as “—” and the panel says Aggregates appear once 5 students have joined. Seats and the roster stay visible; only the behavioral statistics are withheld.
- A week the server could not measure. It gets an empty track and a “—”, never a bar of height zero.
- A rule scored on fewer than five students. The row is removed entirely rather than shown at 0% — and an empty list is labelled Not enough students have been scored on any single rule yet, not “no rules broken.”
- Whether the numbers are live. They aren't, so the panel carries the timestamp it was computed at. Before the first run it says Not calculated yet rather than showing zeros.
- An activity column on the roster. Deliberately absent. With a small cohort it would let the group average be read back to one person, which is the whole thing k = 5 exists to prevent.
How exact is this?
The panel order, column set, wording, rounding and withholding rules are transcribed from the shipped educator screens — the same source the app renders from. The threshold is five measured students, group figures round the way you see them here, and the “fewer than five” cases are handled by removing the row or the bar rather than by printing a zero. This page is static: nothing here calls the API, and there is no cohort behind it.
The Autumn options cohort is a constructed sample — not a real class and not real fills. Thirteen rows written to exercise every state, including the empty ones. No person on this page exists and no figure is presented as anyone's record.
k = 5 is not a display setting. The server computes the cohort rollup and withholds below the threshold, so it holds for every instructor on every cohort, and lowering it for one class is not something support can do.
Everything above describes trades that already closed. It is not a recommendation, a rating of any student, or a forecast. biaX connects to a brokerage read-only and never places an order.
Running a class or a trading group and want a cohort? A cohort is $99 a month for thirty seats, each seat a Pro subscription for that student. Email us, or read the support FAQ.