Age based formula only. They stop the stress tests at this point. I could probably go over this, but I'll be worried about exerting the heart too much.

There are corrected formula.

"HRmax was univariately explained by the formula 211 − 0.64·age, and we found no evidence of interaction with gender, physical activity, VO2max level, or BMI groups. ... Previously suggested prediction equations underestimated measured HRmax in subjects older than 30 years."

https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1600-0838....

That formula also doesn't have much better performance, check out figure 2 in this study to get a feel for how much variance is left over, no matter how good your model is, age is just not enough:

https://www.frontiersin.org/journals/physiology/articles/10....

As-it says in the Abstract I linked:

"HRmax predicted by age alone may be practically convenient for various groups, although a standard error of 10.8 beats/min must be taken into account."

Shall we say 150—190 95% CI for me? (Rarely do I make enough effort to reach 155bpm.)

Not sure I understand what you are trying to say, do you find the prediction to be helpful? 150-190 BPM is such a huge range, what useful information can be derived from it?

It provides perspective and a target training effort.

I've done a good session if I push to 155 bpm. I don't need to worry about finger-sticks for lactate accuracy, because there's no-risk I'll do twice-a-day twice-a-week threshold sessions.

If I'm not even in the low-end of that huge bpm range, I quit and take a recovery day.

Those error bars are too wide for the formula to be useful for training based on heart rate percentages.

I've already said that I have found it to be useful:

https://news.ycombinator.com/item?id=43083815

:so please explain why I haven't :-)

You are doing nothing that requires the formula to be precise. A 10bpm standard error, meaning the max is very possibly 20bm off where you think it is, means that you can't reliably distinguish between efforts that someone doing serious training would want to be distinguished.

> You are doing nothing that requires the formula to be precise.

Exactly.

My guess is that neither are most other readers and commenters here. (Any evidence to the contrary?)

Meanwhile, what do those error bars tell use about "training based on heart rate percentages." :-)

> someone doing serious training

Care to say what you mean?

As-it says in the Abstract you linked: "The 202.5–0.53∗age formula developed in the present study was the best in the studied population, yielding lowest mean errors in most groups, suggesting it could be used in more active individuals."

Let's say it again — "suggesting it could be used".

And "A simple formula predicting HRmax based on age only may be used when the exact HRmax is not needed or is difficult to obtain."

For me the difference between 202.5 − 0.53·age and 211 − 0.64·age is less than 2 bpm!