I’m a fan of mass start as well, and keep EVR as mass-start longer races.
1 Like
Probably signed up before a score increase
Or the pen descriptions are wrong.
There were a bunch of Pinarello RS events where the bulk editing tool didn’t work right, and the original high-end cat definitions were used (where D used to be set at 520-605). So could be the definitions are incorrect, so description doesn’t match. @James_Zwift ?
Is that the next high end race?
My score is just over 600 and it will allow me to enter that “500-550” pen too.
There’s now one signed up in the “550-600” pen with a score of 683.
And the event ZRS range is showing as 0-450.
Hopefully ZHQ allocate some more resources to smoothing out the event creation process, seems like it’s a huge manual effort and prone to errors if not done perfectly.
Sorry to be so late to respond! Using CP and W’ in the CS formula hasn’t come up before, no. Do you feel like it can provide something to the equation?
Nope. Score has been unchanged since the last big change. Signed up yesterday for a race tomorrow.
It’s this one:
It looks like the descriptions are outta whack.
Using CP and W’ in the seeding formula could offer an alternative way to evaluate both anaerobic and aerobic abilities, rather than relying on 30 sec and 10 min power.
You could argue that they provide a more complete picture of overall performance.
Example formula:
Score = ((w1 * CP_norm) + (w2 * W_prime_norm)) * 1000
Where:
CP_normis the normalized Critical Power.W_prime_normis the normalized W′ (W prime).w1andw2are weighting factors such thatw1 + w2 = 1.
1 Like
I thought Danny was referring to ZR.app CS formula 
I’m not sure if Zwift considered those metrics before I arrived on the Racing Score project but it’s not been part of any conversation I’ve been in
1 Like
Sorry! I was a bit quick there.
If anyone would like to be part of more detailed and nuanced conversation about specific improvements that could be recommended to Zwift, I have created a channel and threads on the ZR.app discord and am always happy to hear all opinions and thoughts there.
(No obligation of course. I just find it an easier way to narrow conversation to more specific aspects of RacingScore)
2 Likes
No problem Tim, it´s ok … and thanks in advance for your reply.
To clarify (and I apologize for my basic English): I wanted to know if in the research process to identify which model to use for seeding in ZRApp, was there any kind of analysis based on a formula that combined CP (raw & W/kg) and W´? (and if the answer is negative your opinion about the idea).
From the little I know (my knowledge is basic), the choice fell on an adaptation of the CS 5 min, based on a paper that described the CS 5min as a good predictor of performance for U23 cyclists.
I am interested in it because I think W´ is a relevant factor for the outcome of events such as Zwift races.
But only W’ (imho) couldn´t serve as a seeding formula without being combined with CP (only looking at W’ would be like a chair without a leg).
The two elements, together, seem to me to be an excellent predictor of an athlete’s ability in events were some kind of separation or surge/sprint is very important to decide a race … but also a good aerobic engine.
But that being said, I would like to congratulate you for the AMAZING JOB with ZRApp!
Great idea and it was a great tool to break the “ice age” of Zwift using 0,95 * 20min and after that CE.
Without talking about private conversations that you had with ZHQ, can you tell us why the 10min for the “aerobic” part of the seeding?
To catch more max efforts, since a lot of people don´t do (true) long max efforts?
To “harvest” the data of people doing “The Grade”, and that way catching more max efforts?
Appreciate the kind words. I am sure you could put together a decent model with CP and W’ but it’s not something I’ve analyzed deeply 
1 Like
I wasn’t involved in the data science that led to the formula in use with 30s and 10m.
1 Like
Ate podium bonus related to ZP wins or Zwift side?
As far as I know nothing in Zwift depends on ZwiftPower because ZwiftPower is the past, not the future. But it does raise the question of getting a podium bonus from Zwift results in a points race where Zwift doesn’t know the actual result, or in a race that has late join enabled (shhh did I say that)
But have you explained to them the reasons why they are terrible choices and that whoever did it needs to go back to stats 101?
1 Like