It would be great if a person could see their “predicted” power values that Zwift’s model believes they are capable of to help understand the category assignment.
That would be great, and it would be a cool feature to gamify to see if Zwift can help you hit those predicted values on your curve. How far off your theoretical max are you on 1min/2min/5min etc.
For me Zwift has my zFTP at 315, but I’ve never held a value that high for a significant time. I think the 315 comes from my max 12min effort where I somehow managed to hit 324W for 12 mins, but my best 20mins (during a 20min FTP test) was 311W, so nowhere close to what you’d need to get a 315W FTP result - and I felt like I paced those 20 mins as best I could. I know that technically zFTP is not an actual FTP, but it is a power that Zwift assumes you can hold for a good length of time, so for me 315W seems a bit high.
That would really be helpful to dial the sandbagging in to the 10th decimal.
Check Zwiftpower under the power tab. It doesn’t tell you the whole story because it only takes races into account. Zwift use all riding on the platform.
You guys are awesome!
I did exactly as you suggested - a ride with a couple of hard 6 min efforts and I am now back down to Cat C
I hope to move up to B one day but I’d like to actually deserve to get there
Thanks for all the input.
PS can anyone tell me how to view my power curve as shown above? I can’t seem to find it - thanks.
Are you signed up at www.zwiftpower.com ? Under “profile” click “power” for pure Watts or W/Kg.
Although the power curve there will only show race and group ride efforts I think, certainly efforts from free ride are excluded.
Big improvement in your ~6min?
I am on ZP - thanks for pointing me at the graph
I didn’t do it as part of an event so it doesn’t show on ZP but it has obviously worked on Zwift itself.
Alexa, show me a broken system…
Totally agree - crazy.
I think prior to today’s effort, you were at an zFTP about 3.44 wkg, and per the CE FAQ you only needed to change the curve enough to get to a zFTP of <3.36 wkg.
So you’re definitely right on the border and may be moving back to B pretty soon.
Power curve held at ZwiftHQ can only be as good as the input from you in Zwift, feed it garbage, be ready for rubbish zFTP that might give you an unwanted promotion.
Enforcing something like mandatory periodical ramp tests to race would help.
The biggest issue for me is probably how zFTP is predicting 40min power, while Zwiftpower uses 20min actual, with both using the 95% value to decide pen thresholds. It’s making the top end of B, C and D higher under Category Enforcement (which is really bad for D, given the massive range of ability there).
For example a Zwiftpower 95% 20min average of three being 2.6W/Kg would be a C, but under CE they would be safely under the zFTP promotion threshold of 2.625W/Kg for 40mins and so still be a D.
In my opinion zFTP thresholds should be lower than 2.5;3.2;4W/Kg.
Given Ds and possibly even Cs race a lot on “pancake” routes, there should be a pure Watts promotion threshold for at least these two pens.
I don’t agree with minimum pure Watts for a pen promotion, at least how they are now (150; 200; 250)
I’m not sure how much that would help because you’ll still have a lot of folks who want to stay at the top end of a cat, so they would just not put as much effort in as they could anyhow. Plus FTP tests are super painful, so it’s a hard thing to force on everyone, when some people just want to race hard to have fun
The more I think about it, the more I think the solution is flexible cat boundaries with cat enforcement (or rank enforcement if we do get ranking). This way there isn’t a perpetual “top of category ‘x’” that people can sit in all season. With only 4 static categories the deltas between low and high end of the cat are big enough that some people just won’t progress to the next one in a season, and might be continually at the bottom of one if their fitness has peaked as well. Having flexible cat boundaries would give people the ability to be top of their cat on some days, bottom on others within the same week of racing.
But if riders with Zwift history sandbag the ramp test, ZwiftHQ can point to previous data and say that was a sandbag ramp test, you have a history of hitting over [insert heartrate]… Sorry, you need to do another ramp test before you can race.
Ramp tests really aren’t anything like as fatigue inducing as a proper FTP test, for someone with an FTP of ~300W (which means they will be probably attempting the 400W ramp interval), it takes little more than 20mins to reach point of exhaustion before a cool down… Of which ~5mins will be at or above threshold and the first 5mins is freeride, it’s easier than two Tiny Races!
Example from 2 years ago…
If you want a accurate power profile then I will suggest doing the 4DP test it will generate all the needed data points.
The original categories was based on a FTP test and that worked well but racers that didn’t want to be upgraded made ZP implement more rules but then those rules became the norm and sandbagging became a problem again.
Now CE is trying to level the field again until racers find a way to sandbag CE.
So suggesting doing a FTP test is going back to the original base of ZP
There’s all sorts of tests that could be used, I imagine something more generic like a ramp test will cost Zwift less than paying Wahoo to integrate their 4DP.
But what it comes down to is an efficient racer is going to try and use the least amount of power to try and win a race, basing a pen allocation system on this data isn’t likely to produce a true representative power curve for a rider and so give a reliable zFTP or zMAP prediction.
Sure, Zwift uses all Zwift activities for Category Enforcement, but there’s nothing stopping racers from doing their training on other turbo platforms and just using Zwift for “sandbag racing.”
One year zoon all this power based pen allocation will be history, just don’t hold your breath in the meantime!
One can hope.
But there will always be the need for initial power based seeding to get people in the correct starting pen if they have no racing data.