GutThe readiness algorithm you were born with

Do you feel like training today?

No account. No sensors. No data collected.
GutThe readiness algorithm you were born with

01 — The point

You already knew the answer.

Readiness apps measure sleep, heart rate variability and training load to estimate something you can report in one second.

The cheap measure holds up. A review of 56 studies found that subjective and objective measures of athlete wellbeing generally did not agree with each other, and that the subjective ones tracked training load with better sensitivity and consistency.1 In Premier League players, perceived fatigue tracked daily load about twice as strongly as HRV did.2

Readiness has no ground truth. Nobody can measure it directly, so a model has to be trained against a proxy, and the proxy is a performance test or a self-report. The system is learning to predict the thing you could have said out loud. Give it 365 highly autocorrelated samples a year and day to day noise larger than the effect it is chasing, and this stops being a problem a better architecture solves.

The problem is the days it disagrees with you. You feel good, the app says 41 percent recovered, and you take an easy day you did not need. A number with a nice interface tends to win that argument.

Reading your own body is a skill. It gets better with practice and worse when you hand it to something else.

If you build these things, there is a baseline you owe your users. One question, on its own, against your score. Run the comparison. If you have never run it, you do not know your model adds anything.

02 — Price

What the alternative costs.

Wearable or AI app subscriptionGut
SubscriptionAbout $10 a monthNothing
Hardware$200 to $400, replaced every few yearsNone
SetupAccount, app, a few weeks of baselineOne question
Data collectedHeart rate, sleep, movement, locationNone
Time to an answerOvernightNow

A wearable does plenty that Gut cannot. None of it is answering the question on the first screen.

03 — The honest part

Data is worth having. Just not for this question.

Measurement earns its place over long horizons. Training load across a season, pacing and power over a race, the slow drift of your resting heart rate across months. None of that is available to introspection, and all of it is genuinely useful.

A raised resting heart rate can flag illness a day before you feel anything. That is one signal that beats asking yourself, and it needs a number and a sense of your own normal rather than a readiness score.

Some people should not trust their gut on this. If your honest answer is yes on 27 days out of 28, your gut is campaigning rather than reporting, and something external is a reasonable check.

So the argument is narrower than it looks. Data is good at questions you cannot answer yourself. Whether you feel like training today is not one of them, and it is the one these products are mostly sold on.

04 — Why this exists

I build sports data infrastructure for a living.

I am the founder and developer of SweatStack, a sports data platform for developers building training apps. My working days are spent making athlete data easier to collect, normalise and query.

Which is most of the reason I made this. The teams I work with are good at adding data. The question that gets asked less often is whether a particular piece of it earns its place, and a model that repeats back what the athlete already knew has not earned much.

Better training apps are the point. Sometimes that means more data. Fairly often it means less.

References
  1. Saw AE, Main LC, Gastin PB. Monitoring the athlete training response: subjective self-reported measures trump commonly used objective measures: a systematic review. Br J Sports Med 2016;50(5):281–291. doi:10.1136/bjsports-2015-094758
  2. Thorpe RT, Strudwick AJ, Buchheit M, Atkinson G, Drust B, Gregson W. Monitoring fatigue during the in-season competitive phase in elite soccer players. Int J Sports Physiol Perform 2015;10(8):958–964. doi:10.1123/ijspp.2015-0004. Perceived fatigue r = −0.51 against daily high-intensity running distance; Ln rMSSD r = −0.24. The authors read both as usable.
  3. Thorpe RT, Strudwick AJ, Buchheit M, Atkinson G, Drust B, Gregson W. Tracking morning fatigue status across in-season training weeks in elite soccer players. Int J Sports Physiol Perform 2016;11(7):947–952. doi:10.1123/ijspp.2015-0490. Perceived fatigue, sleep quality and soreness tracked session load across the week. Submaximal heart rate, heart rate recovery and HRV did not change substantially.