You’re supposed to include your clothing! That’s part of the weigh-in protocol for every event that I’ve seen. Not shoes, just shorts, shirt and socks.
1 Like
My weight doesn’t change much tbh (as mentioned before I’m made of pipe cleaners), but I also thought it would look more dodgy if I was constantly fiddling with it up and down. It’s a fair comment though, I will check mine again later because it’s been a while.
Fairly sure my height hasn’t changed, so I’ll leave that one. 
1 Like
the CP seems way too high if anything (i put in my lifetime PRs of 7.4w/kg for 3 minutes and ~5wkg for 20 minutes and got a cp of 254 which is barely above the proposed B>A limit), the MAP is causing problems
1 Like
I think the best way to deal with the wattage floor is to design a smoothly varying function that goes between 4W/kg at high weight and 250W at low weight (these values can of course be changed). I just haven’t had time to design it yet. Will do soon.
1 Like
Based on curve fitting existing data or on assuming a plausible first derivative ?
Even if you change your weight every day, there is no guarantee that the stated weight is real. You can not conclude anything from weight changes or lack of changes.
5 Likes
I don’t believe for a second I’m the first to bring this up, but as we are experimenting with the formula to calculate category enforcement this week…
How are so many racers getting away with such infrequent weight changes in their Zwiftpower profile?
You know, if we could sync our smart scales to our Zwift accounts, that would be awesome.
2 Likes
True.
Let’s do away with user entry weights and as it’s a game, make everyone in Zwift an average weight…
80-85Kg sounds good? 
Or how about categories according to weight brackets, lightest riders in A, instead of W/Kg (of sorts) based categories?
Somewhere in one of the three threads dealing with sandbagging and cat enforcement, I think Flint gave a tentative schedule for rolling out cat enforcement to community races.
Can someone direct me to it.
These threads explode over night and get so long, it’s hard to find specific info.
Read my post again. I did not say anything in that direction.
As a pretty light rider (62.5), not extreme, but lighter than most, can I just say I always find the claim that light riders are categorically disadvantaged somewhat contrary to my experience (and occasionally slightly patronising).
My best results come on flat courses with bunch sprints. I can climb better than many, but I also find a greater percentage of those better than me (A+ guys) in the rankings can climb even better still, but many of those can’t sprint so well.
I never have any problem staying in the front bunch of even the best crit races and while I rarely have a race winning sprint (I don’t have quite enough when I hit the wind) I’m there or thereabouts, within a second of the winner. I’ve never been dropped on a flat course.
Ultimately if you have the strength, light weight is not an guaranteed disadvantage written into the very code of the algorithm.
3 Likes
As long as the ranking is based on those trophies how could one be achieved without affecting the other?
Speaking for myself but several others have posited the same a thousand posts back or so: I don’t care about people with inaccurate SMART trainers (Zpower is a different kettle of fish entirely) or don’t accurately post their weight. All that matters to me in a race is that I’m racing against people who are game-wise genuinely in my category. Just be a real C cat (or B cat if I ever get there) and let’s go at it.
If someone is lying about their weight so they’re faster than they would be otherwise IRL, that’s on them. Don’t care as long as we’re on a level playing field in the game.
Now, using the not to be spoken of exploit during a race … I KEELLLL YOU! I don’t think I’ve ever seen that but sh*t when I’m racing I don’t see much. Friends ask me later, “What was your power up so and so?” My answer: Ummmmm… 
2 Likes
I think that we need to use a system that ranks people in the pens based on how they have finished in previous races, so if someone is consistently winning by a large margin in a certain cat, they get bumped up. if someone is consistently doing very poorly, maybe get bumped down a cat. It’s not perfect, but it will make it more fair for very light riders and very heavy riders, while maintaining fairness for those in the middle of the weight/height range. It would also encourage people to race in the type of race that they are not as strong at so that they don’t get bumped up as frequently, which would ideally develop more rounded riders, and more interesting races. Any one else agree with this?
If you’re in A cat of course there isn’t an inherent disadvantage because you aren’t prohibited from putting out a a few more watts. If you’re in B/C/D cat you are liable to get disqualified or upgraded for riding with the big guys at the front if they are pushing hard. Not saying a 65kg rider can never win there but it’s obviously harder for them and the data bears this out (zwift insider analysis).
1 Like
I have been thinking along the same lines.
Zwift keeps track of how successful each rider is based on the result position in percent of the total number of finishing riders.
Variables involved in the system:
-
The position in percent of the total when the rider places below t percentile (pl)
-
The position in percent of the total when the rider places equal to or above t percentile (ph)
-
A counter for how many times the position falls into pl (plc)
-
A counter for how many times the position falls into ph (phc)
t = Upper and lower percentile threshold. Could be e.g. 5%
pl and ph are stored as the average for the last n days.
The rules for promotion and relegation:
-
Promotion: ph > t AND ph > 100-pl AND phc/plc > f AND phc >m
-
Relegation: pl < t AND 100-pl > ph AND plc/phc > f AND plc >m
m = Minimum races over/under the threshold
f = Is the relation between how often the racer places within t in one end vs within t in the other end.
If a racer has an equal number of wins vs losses, we should not move her. f could be e.g. 2
NOTE: Both formula and limits are meant as an example. Here are many variations.
The first part checks that the racer is at the extreme of the category
The second and third parts check that the racer has a trend in one direction vs the other
The fourth part checks that the racers have enough races in the category
When the rider is promoted or relegated the stored numbers are nulled and they have to do a minimum of m races in above or belov the t percentile to eventually go back to the category they came from.
I’m sure there are a lot of holes in this system, so don’t be too hard on me 
There are many unsolved problems:
Should the total number of racers in the race matter?
There should be a quality of field. And points should be in relation to field size.
So a field of 5 riders will give you less that 20 and more than 20 is the same as 20. Random numbers here.
Hopefully it doesn’t look dodgy. I weigh myself every day, and have withings hooked up via the fitbit hack (which apparently some people are having trouble with too). Originally I didn’t know withings integration was broken (so all my TdZ rides show the same weight), but now I’ve got the Fitbit hack set up so technically my weight “should” be going up and down every day a bit if the fitbit hack is now not broken.
Just looked, I’m currently 900g heavier than my game weight. Changed it. 
1 Like