Yep, I think a single effort is enough rather than a best 3 - although that effort only bends the curve or potentially pushes up your MAP.
James, ‘bends the curve’ is exactly the correct description, you have to do your best numbers at all durations to seriously alter it, and if that was the case you probably deserve an upg 
1 Like
Cheers @xflintx you the funking man, and have to have some pretty thick skin to be the human shield for Zwift HQ on here!
13 Likes
I’ve gone from mildly interested in these events to really excited to see how it plays out. Selfish I know, but it feels miles more suitable for me now. I’m thrilled that the feedback has been acted on to address the outliers, and I hope it’s sorted things for the large majority.
I do more than enough moaning on here so when things are cleared up, fixed and/or improved then credit is due. Even moreso when it happens swiftly. @xflintx , @James_Zwift and others, thanks for listening. Likewise @gloscherrybomb for assisting, and the other folks who have articulated the issues sensibly on the whole. I appreciate it. 
9 Likes
PS: I think everyone agrees this is just the first step in what could develop into results-based racing and who knows what else. It’s not the perfect solution and that’s absolutely fine.
1 Like
The absolute best and easiest way to predict how well someone will do in a bike race is to look at how well they have done in previous bike races.
Ranking is reasonably easy to implement, document and describe.
Categorization via ranking levels is reasonably easy to implement, document and describe.
People can easily see what category they will be in, and (almost as important) who will be in the same category.
If it was easy to predict winners from power curves we would not actually need to race. Just watch what people do day to day and post a daily list of the best “racers” of the day.
4 Likes
We have had years of this “testing a new system” , when all this time they could have just implemented PEN enforcement while they played endlessly with refinements and there would be no complaints at all right now at more test events. .
Before you build a house you lay the foundations and make secure the area. Its not a matter of either/or its simple a matter of maximum value from minimum effort which is how you build these things .
What should have been a good news story here looks like its unravelling because the feedback that has overwhelming been given til date has been to all intents fundamentally ignored . Thats why some might be a little disappointed and annoyed at the request to “give more feedback” … we already have for years and its been ignored … again 
1 Like
Thanks Flint. Cat Changed for me from A to Cat B for the Monday Race. that looks like a good Matching for an upper C Rider.
2 Likes
And of course this . What is with all this endless and ultimately doomed to failure attempt to predict race outcomes by sensor metrics and output numbers . I dare say with enough compute power it can be made “better” , but surely to goodness the easiest way to rank people at how well they race is to … rank there race results . Obvious right ? .
3 Likes
I’m confused, who is trying to predict race results?
The purpose of splitting riders in to pens is to group riders by ability to keep the racing interesting and exciting. The strongest rider nearly always wins. There is an extremely high correlation between physiological metrics and outcome, in real life and on Zwift. Splitting riders by ranking is also a great way to do things, but it presents a different dynamic, not necessarily a better one. That’s why I am ultimately pushing for empowering the race organiser to determine pens as they see fit.
This development, now that it is understood, brings 2 HUGE improvements to Zwift Racing. It is now impossible to sandbag, and it is now, to all intents and purposes, no longer practical to cruise.
Let’s test this and get it embedded, and then hopefully the next steps will be pen definitions for race organisers, and a new ranking system.
3 Likes
Conceeded that Race Results has cons . However it has one massive pro when it comes to categorisation. Results are black and white and easy to understand . It also is the basis of sporting competitiveness . Do better in a race , get promoted , do badly in a race ( or series of races ) get relegated . Its IMO fair , easy to understand and confirm and measure … its why many if not most sports use it . ( Including most non age specific competitve cycling race leagues )
2 Likes
It is true that i can put down way more watts then others if i join a lower level race. But that doesn’t mean I do. I am talking about long distance races I like to join in the B category. I just stick to the power <4W/kg and I am not going over it. So usually I will be in the second or third group. So it is a bit different than your e-bike example since I am not beating anyone. As I said, irl I join lower cat races too and just sticking somewhat in the back. Cyclists in that position in the back are usually struggling anyway and not competing for the win. So I will not influence a lot from that position.
The tweaks seem more reasonable for me. Now I don’t need to jump up 2 Cats to try it out.
Appreciate the quick turnaround to incorporate feedback.
3 Likes
See the other thread, they’ve changed it. 
1 Like
The companion app still shows the ZP w/kg limits for each of the 4 pens. Is that intentional or will it be eventually tweaked to reflect the new criteria?
I am excited to start racing again now that the pen enforcement is fixed.
6 Likes
I find two REALLY positive things about this change:
-
Zwift has listened to feedback, and responded to it.
-
Zwift has been ABLE TO CHANGE - very quickly - a system of pre-race enforcement.
Literally within 24 hours, a system has been changed. Now THAT’S promising.
11 Likes
No, the purpose of splitting riders into pens is to group them into equally successful riders.
Think of a ladder tournament at your local tennis club. You climb to the top by winning. Not by hitting the balls faster than your opponent.
While hitting balls faster is one indication of possible future success, actually winning is a better predictor of future success.
4 Likes
Categorization by rank is also what is the primary method used IRL.
So, again, meets people’s preconceived notions of how it works. People in the A category have won (or podiumed, or top 5 etc) more than the people in B. People in B more than C, etc.
Start people in a category based (roughly) on the w/kg, move them up or down if they are in top or bottom X%.
Not difficult to implement. Not difficult to police. Not difficult to describe. Easy for people to predict where they will end up based on the results of any race. Up, down or stay.
No magical formulas required to estimate future success based on power curves or VO2Max etc.
3 Likes
No
-
How are you determining the category that I belong in?
- We are using a 2 month historical look at your best values for both estimated VO2 max and FTP (CP/W`) as it is captured in Zwift for all activities, excluding workouts. This is the same for every rider. Riders with an ability to sustain higher short duration power and with a lower FTP might see their available categories increased as a result.
I read that as any Zwift ride, aside from workouts, including solo rides.
@Gerrie_Delport_ODZ Based on those rides, I cannot fathom how you have been put in A. However, do you have other solo rides on Zwift w/better power numbers or have you you really only been on Zwift for 11 group rides over the last 60d?