Compound Scores

According to the latest publication by Peter Leo and collaborators they have created a metric called Compound Scores, it relates w/kg to absolute w to the probability of success of one athlete compared to another. It is a metric with a lot of future, you could even make a curve like the power curve but with the compound score metric.

Attached is a link to the preview of the publication:


Tx. I saw that and was wondering if someone would ask for it. I will have a look at the paper.

In the screen shot you have:

Screen Shot 2022-07-04 at 21.31.25

What is the 3’ and 5’?

MMP 3’ and MMP 5’

PD Curve with Compound Score curve

The paper uses compound score calculated from MMP 5’ right? I could easily add that under vo2max here:

Screen Shot 2022-07-04 at 21.43.22

As you can see things aren’t going too well for me after my partial knee replacement :frowning:


It would be a good idea, even create a power curve but with compound score.

I after my knee surgery in 2017 from 320 ftp lost more than 80 watts, the medical intervention did not go well and I have not yet recovered, after covid my ftp dropped to 190 w, I have been left with sequelae in the vagus nerve.

Ack I will stop complaining about my 10% then … I might even get that back eventually. Sorry about your knee.


What is helping me the most is the work in the gym, I am improving my metrics again, I recommend the gym.

1 Like

Its live:


wow, Spectacular David, this metric is going to have a lot of impact.

1 Like

I know its less (if at all) relevant, but any chance you can put this on the running power tab also.

Great to see this included so quickly but not entirely sure how to interpret/ use it yet. Study seems to show some correlation between the likelihood of placing on the podium in a U23 Continental Race but not much more at this preliminary stage unless I am missing something.

Perhaps provides a way to build further evidence for the study. With someway of allowing athletes to enter race results against their A, B, C assigned races etc and an extension of the weighting table against the categories there could quite quickly be a large data set which could be reviewed.

Could be worth exploring with the authors David. Just thoughts.


I think you need big data and a categorization of the numbers, we can see how this metric evolves, although in my opinion basing it only on 5m does not seem the most correct, it would be interesting to be able to visualize it in a curve and also visualize it with fatigue.


Sure I have enabled it for other sports with power.