# Zwift power curve is bad

(Johnathan) #1

How can the same values be so far apart on the graph? I know it’s never been perfect but this is bad guys.

(Patrick) #2

It’s worse than that. I often have my “this ride” curve completely cross over the top of the all-time best curve and actually be higher. It’s also not uncommon for me to have a higher value at, say, 5 minutes than I do at 4 minutes. Both of these situations should be impossible by definition.

1 Like
(Johnathan) #3

This is what the CP should look like.

(Lin) #4

No, both are very possible. In the first case, perhaps “this ride” is actually better than your “all-time” best. This assumes of course, the all-time best does not get updated until the following ride. In the second case, this could be the result of intervals:

That said, there is definitely something wrong w/the screenshot @Johnathan_F_IndoorSp provides in his original post.

1 Like
(Johnathan) #5

Excuse the crude drawling but this shows how it is possible to have a higher 5 minute power than 4 minute power on a CP curve.

(Patrick) #6

Good call on point #2. I see now how you could achieve a higher average over a longer period of time through intervals - that was a failure of imagination, assuming a relatively smooth power time history. To achieve the anomaly, just have two big intervals spaced far apart with a significantly lower effort in-between.

I think #1 is a genuine glitch, though, probably related to the OP, as I’ve seen it update “correctly” when I out-perform my previous best and the lines sit right on top of each other.

1 Like