So I went to try the Zwift games last week and noticed I have a ZRS of 949! WTF?? I’ve not raced for many years (and even then, it was occasional) and I definitely never won a race. So how does an 80kg 55 year old with FTP barely over 240 get such a high ZRS that I would have to race in A category races? Seems to me Zwift has a division by zero problem. I’m not going to sign up for A races so I can ride by myself. Ridiculous.
Is it possible you did a strong effort at a lower in-game weight? If you look at your power stats on your profile on the Zwift website you can see if any of the power numbers look high to you.
You can email zwiftpower@zwift.com to get help with this. The Fitness section on my.zwift.com will have your 90 day 5 minute power PR that determines seeding. Check the date on it and compare to your activity history.
ZRS seed score (zrCS) is based on best 5min Watts and Watts/kg over the past 90 days.
If you changed your weight significantly for something and did a hard 5min this could happen.
ZWIFT ZRS v3 by VirtuSlo says that at 80Kg, seed score of 949 puts your 5min effort at ~482W.
But, if for some reason their weight was lower when the value was set then the power needed for that rank would be much lower, which is why it’d be good for them to check both the power and w/kg of their best 5min effort on the Fitness section of the Zwift website.
Just take the 5 min PR from the Fitness section of the profile, divide the Watts by the W/kg and you have the weight to verify that it is not something unexpected
1 Like
Any update?
Thanks for all the suggestions. This is my power info: clearly I’m not racing with Pogacar:
Silence from Zwift support so far besides an unhelpful comment that they are aware there are bugs, but I can “in the mean time just keep racing as usual”. 
1 Like
~3.2W/Kg for 5mins would give you a seed score of ~275, lower if you were lightweight.
1 Like
This is mine at 79kg, that had me at a ZRS of 395
5 min 299w 3.78w/kg
1 Like
That is odd. I would not expect a big racing score increase based on that 5 min number. Mine is 301W / 3.91W/kg and I’m down in the 400-450 range.
Yeah, you should not be in the 900s. Something is odd.
Not sure where @Mathew_Bittleston is looking?
I looked this morning and it was definitely what he said. Looks like it has been corrected
Yes, this morning it was still in the stratosphere. I told support about it last Wednesday… Looks like they changed it this afternoon. No email or anything though! Thanks everyone for your input.
OK Zwift support got back to me, and here’s the reason: Don’t let your niece play on Zwift!
During mid-December my niece came to visit and she wanted to play with zwift. She must have changed the weight while she was using it, but I’m sure her power/weight ratio was nowhere near Peter Sagan; Zwift probably saw that weight, and then applied it to all my numbers internally(?) even though the numbers they were showing were based on my current weight. So: If you change the weight, even for a day so a kid can use it, it will mess up the ZRS. Anyway, “solved” for me. Hilarious how these algorithms have no “That’s ridiculous” checks in them.
Well, if the new PB was set while the weight was low then it should affect the ZRS, even if you reset the weight higher after. But if there were no PBs set (from a w/kg perspective) at the lower weight then it shouldn’t make much difference - So, what probably happened is you did a ride afterwards with the lower weight before putting the weight back to normal.
Though in either case it should probably show up in your PBs on the Zwift site, so not sure why that didn’t show up when you looked. Wonder if it was already taken out by the time you got to the screenshot.
Yes this is normal and expected. If you want to introduce someone to Zwift it’s best to have them create a new account and use the trial period before deleting the account. The thing I wish Zwift would do is make it impossible to make an extreme weight or height adjustment. It should just say “sorry, no”.
I don’t think their algorithm is that sophisticated, but I went back to look at the rides recorded around that time: She was generally 30-50W and the rides showed some maximum spikes up to 277W. She’s got some power 
The first ride I did after that was a couple of weeks later in January, and I don’t see any crazy speeds, and I’m pretty sure I checked the weight after she was done anyway.
I very much suspect the algorithm is just looking at the lowest weight over a 3 month period, and then looking for the highest power over the same period. I doubt they’re bothering to calculate P/W at each ride (the mathematically correct approach).