The power curve visible on the exit screen resets automatically after an event is finished, even when simply continuing your ride in the same world… this didn’t used to be the case (and it also defies a bit the purpose of the functionality, since most power records are set in an event).
Would be great if you can fix this.
I noticed it simply is the scale of the event curve vs the all time curve that seems off… I had a PR today (as indicated by the stats that appear on that chart when hovering your mouse over it), but there was nevertheless a gap between the two curves…
On top of the issue described above, the power curve of a current ride sometimes resets upon finishing an event. Whether this occurs seems to depend on the world.
I’ve been in the same boat and I see you’ve highlighted all of the relevant posts.
I’d just accept that Zwift puts their cash towards getting new subscribers. There are 100s of bugs that have not been ironed out that are being ignored. It clearly isn’t a priority. You’ll see that many zwift subscribers use Zwift in combination to other services eg external training plans.
I’d sign up to a different power analysis website/service as zwift is woeful in this area and just accept that Zwift is there for the social element.
If you are just interested about events specifically, you can also see the power curves on Zwiftpower. For power data analysis in general, there are indeed many much better alternatives on the market.
It was a nice feature while it worked. There’s no free software out there which would allow for a simple comparison between an event’s power curve, and your all time zwift-specific PR curve (zwiftpower included)
So, alternatively, instead of me paying an additional monthly fee for another piece of external software which might allow to construct a PR power curve for a specific subset of rides, which could serve as a benchmark for the power curve of a specific ride… they could … and I know this sounds insane - but bear with me for a second … fix this simple little issue 
(alternatively - remove it from the feature set)
(subscription fees paid so far: 600 euro and counting)
Take of your free zwift cap and bear with me, Gerrie 
Until they actually pay you for moderating, I guess you can keep a tiny bit of objectivity and agree with me that they should simply fix this.
(yes, I know and use Golden Cheetah - No, I won’t register a new account to link to Zwift exclusively to get a Zwift specific PR curve)
I used intervals.icu
You can definitely compare event power curves to indoor (zwift) power curves.
Two screenshots taken after yesterday’s ride (Japan)…
5 second power stands at 1050w (which was also a PR):
neither value is taken into account in the power cure on the next page:
(one of the many random bugs with the power curve)
the average power is also not the same, 247 and 239?
That could be due to the average on the second screen reflecting the short period of time I coasted to take those screen shots (it doesn’t get paused immediately, only after coasting to a stand, I think)
Been nonsense for ages. I wish they’d fix it.
Golden Cheetah, with the bonus that it can spot and flag potentially erroneous data from your powermeter, allowing you to then fix it too (be honest with yourself!).
I load all my rides up Golden Cheetah so I can see my training load & power curve both current and in previous time periods. Does mean I’m mixing power data from multiple sources with all the quality issues that entails, but it does give a fair idea as to where I am and where I have been over the years in comparison.
Paid version of Strava will also create a power curve for you.
Plenty of dedicated online sites and apps that will take your money for in depth analysis as well, which shows how involved this “simple curve” can get.
I’m not bothered about analysing it, but I’m paying for Zwift so the least I expect is that a simple overlay graph with two lines is accurate. I don’t want to piss about with other websites and shouldn’t have to, let alone pay for them.
It bongles my head the amount of excuses made for Zwift by its own customers. Never known owt like it.
Except I don’t think it’s a “simple overlay”, it’s a Critical Power Curve.
Should they get it right - YES
Are there things I’d rather they put their finite resources into first - YES
In the grand scheme of what you are paying for with Zwift this isn’t even small fry, BUT it also looks very bad as it’s the display layer, the menu/ 1st exit panel with the HR & disappearing Power histogram is also a bad look.
Are there good external solutions out there - YES, for those who want to use them and possibly pay for them too.
I use Golden Cheetah because I used it before I got on Zwift.
There are lots of bugs Zwift need to fix, I’m hoping like all good IT based Companies/ Depts that they have prioritised them and are hitting with the bug hammer, they do need to sort out the UX/UI as it is customer facing, it’s the shop window and that is easy to forget when you’re up to your neck in re-working serverside code to fix a deep lying fundimental issue.
Some people just don’t remember the “fun” of indoor training before Zwift, it were proper mingin’ r kid.
this is the bug reporting section of the Zwift software forum.
(And Neo’s don’t produce erroneous data )
I wanted to report the same issue with wildly inaccurate power curves. My post event power curve shows my 1 minute effort at 249w despite averaging 337w for almost 2 minutes for a segment. The ride specific power curve shows my 1s power for the ride at something like 350w despite hitting 449w. It seems like its not even looking at the same data for the power curve and the timeline.
