Your training data, analyzed — not just tracked.

MyThreshold pulls in your training data automatically, computes the same training-load metrics real coaches use, and walks through them with you in chat — every day, based on what your numbers actually show.

Request access
Fitness & fatigue, over a training block Illustrative data
Illustrative fitness and fatigue trend over a training block
Fitness (CTL) Fatigue (ATL)

What it does

Syncs your training data automatically

Every run, ride, and workout pulls in automatically on a schedule — no manual upload, no CSV export, no spreadsheet.

Daily analysis you can talk to

Chat with an AI assistant that has already read today's numbers and suggests plan adjustments based on them — you decide what to apply. Not a static template, not a canned tip of the day.

How it's different

  1. Data-driven, not templated

    No generic 12-week plan handed to every athlete. Your plan is shaped by YOUR training data, not a template that ignores what you actually did this week.

  2. Computed, not guessed

    CTL/ATL/TSB, ACWR, and polarization are computed the same way exercise-science tools compute them — never an LLM estimating a number from a vague impression of your training. See exactly how, with worked examples, on the metrics reference.

  3. Chat-native

    Ask anything about today’s numbers, in plain language, and get an answer that starts from what was actually computed — not a dashboard you have to interpret alone.

Every proposed change, checked before it's applied

Before any proposed change to your plan is applied, an independent AI reviewer checks it — and the AI advisor that proposed it has no way to approve its own work.

AI advisor

Proposes a change

Reads your computed numbers through a fixed set of read-only tools, drafts a change, and sends it as one typed request.

The only thing that crosses. There is no other channel.

Independent AI reviewer

Reviews it

A different model instance — the advisor has no way to approve its own work. It is walled off — it can't call out to check anything else, and it reads the request as data, never as instructions.

The verdict

Rejected nothing is written
Needs changes nothing is written
Approved the change is written

A blocked change goes back to the AI advisor to be revised, and starts over as a new request. The same request can never be sent twice.

Only an approved change is written

Your plan
Your analysis
Your profile

This is a quality/reliability check on how the product reasons about your training data, not a clinical safety system.

Proposed plan changes are checked by a second, independent AI reviewer before they are applied — but MyThreshold is a data-analysis tool, not a coaching or medical service. It doesn't know your full medical history, and it can't replace judgment from a doctor, physiotherapist, or certified coach.

Read how the review loop and the specialist bench work

Hard calls don't go to one generalist model

18 specialists

When a call actually matters — a quality session on tired legs, a plan that ramps too fast, a number computed from an incomplete week — it is checked by the reviewers scoped to that one domain, not by a single model doing everything at once.

Readiness and injury risk

4 specialists

Whether today's body matches the session that was planned — accumulated load, sleep, travel, and heart-rate behavior.

Session design and periodization

5 specialists

Whether a proposed session fits the block you're actually in — and whether the last one was executed the way it was designed.

Fueling and hydration

2 specialists

Whether the session is fueled for, in heat, on travel days, and after a long or hilly effort — inside your stated dietary preferences.

Schedule and calendar integrity

2 specialists

Whether the calendar the plan is written into still reflects reality — travel windows, firm days, and provisional ones.

Data integrity

5 specialists

Whether the evidence itself is trustworthy before anything reasons from it — missing sessions, stale state, a sync that failed quietly.

Every one of them is advisory-only and reasons from the same evidence packet your numbers were computed into — not from an impression of the conversation. None of them writes to your plan: a proposed change still has to pass the review above before anything is applied.

See what each group of specialists actually checks

Real output, not a testimonial

Illustrative data

MyThreshold is early — there's no wall of logos or quotes to show you yet. What we can show you is what the product actually produces. The numbers and chat comments below are illustrative, not a live account.

Your fitness (CTL) has climbed steadily for 6 weeks straight — the consistency is showing up in your numbers, not just your legs.

Fatigue spiked after last week's long run, but your form (TSB) is trending back toward fresh. Good week for a quality session.

82% / 18% easy/hard split, computed from pace + HR zones
1.18 acute:chronic workload ratio (ACWR), this week

Your training data, analyzed — not just tracked.