alex_cycles
Veteran
- Location
- Oxfordshire
Ok so I'd much rather talk about something technical like this. So here goes...
They've based CE on two things. MAP and CP. But they've confused the issue horribly by calling them zMAP and zFTP.
The reasoning behind zFTP was that people are more familiar with the concept of FTP than CP.
MAP is your maximal aerobic power, which is estimated from a maximum effort somewhere between 5 and 8 minutes (I think). From my own experience they didn't use 5 minute data but 5.5 minute - but that doesn't mean it's always like that.
CP - critical power - is a more complex calculation and it isn't fully transparent exactly how they're calculating it. But it's usually calculated from ~3 data points, but I think Zwift is using ~5. There's a very good article explaining all about CP here but essentially it a 'best fit curve' and Zwift are using it to estimate how much you can output for a longer period of time - "about 40 minutes" they say. In my case. They estimate 216W, which is pretty fair. I think I did exactly that plus or minus 1 when I recently raced up the Alpe.
Each of these two has thresholds for A,B,C,D categories
and they do not line up perfectly with the old ZP 20-minute w/kg - as evidenced by the fact that I was C in both in October, B in CE for Nov/Dec, C in both again from Jan-March and then became a ZP B in March. At no time have I been B in both at once
The issue with the new CE is that - particularly now they've shown people their numbers - there are things you can do to manipulate it.
e.g. not doing a very hard 6 minute effort if your MAP score is higher than the category you want to be in
Conversely, I have heard some people say that if your zFTP (CP) is too high, sometimes doing a really hard 3 minute effort can bring the other end of the curve down a bit. So it is open to manipulation to a certain extent, but probably not as much as ZP was.
It was as confusing as hell having both though - particularly if you have a foot in each camp. Recently I've been able to do Tiny Races in C or Race London series in C but not any races which use ZP cats in C. But to complicate matters further, there were some races that were set up to use CE AND ZP. So you could do a race that CE let you into but STILL get a DQ. (I managed to avoid those).
Better than both of those is zwiftracing.app which is getting a reasonable amount of traction now. It uses a different power metric for initial 'seeding' - which is a pretty good estimate - based on modified CS (combined score) but also an ELO-based algorithm to adjust people's ranking based on their race performance. Currently, the Chasing series and upcoming DRS will be using this. Essentially it splits the categories, so B is more or less Gold 1, Gold 2 and Gold 3 (over-simplification) so DRS in Gold 1 should be a lot of fun as it'll be like low B/high C. But all the more powerful Bs will be Gold 2 or 3.
Zwift is working towards a results-based system but they're still at really early stages with it.
Any system used by humans will be manipulated. I totally hated ZP cats because they were so heavily biased towards >85kg sprinters and <50kg climbers. People in the middle gets called names or kicked out of category for going over cat limits while still not being in the top 10 finishers.
In my opinion a results-based system should reduce this. It simply makes sense to me that if you regularly win races, you should be racing against better opponents. But not so much better that it isn't fun any more - this is key. So smaller groupings are needed.
It makes no sense to me that anyone should be forced to race better opponents if they've never even got on a podium (and it did happen with those middle weights).
Same. I'm sure it will, but I'm definitely not doing lots of evening races.
Last edited:
