Yes, was 100% meant for autocategorization. Great stuff. Love to see all the work you guys are doing. Thanks for the quick response!
1 Like
IME even the longest races (2h) are brutal at the first couple of selection points, after which the pace is much more moderate.
Climber’s Gambit was 33km (?). Granted it had a climb, but I was over FTP for the last 19 minutes.
But overall a fair point on length of race. Proper races over that length, assuming they are not pan flat, are likely to have areas where you are pushing hard for periods of time.
1 Like
Yeah, i forgot about 20-25min time trials. That’s basically the same as the 20-min test. And to be fair, the current system also punishes people who’ve done those because most other 40-min+ races wouldn’t have a max 20 min effort like those do. So a racer in today’s system who is close to going up a category is incentivized to avoid those 20 min TT races.
Personally… I’m in favor of all biking activities counting (including workouts) to get the fullest picture of the strengths of the rider. I was brainstorming why workouts were excluded in this iteration (in case xflint doesn’t give us that insight), and “to failure” tests - which I presume more people do these than do 20 min to failure on a TT - were the best thing I could come up with.
Only one left I’m not sure about is the ramp test. For a 20 min FTP test and a 20 min time trial a biker has to be pretty experienced, and very well paced to actually go to failure, but for a ramp test it is much easier to find a rider’s actual failure point and perhaps folks who took a ramp test would be bumped up in terms of CP more than they would in the other race conditions?
Guess it’d be better to get the lowdown from xflint 
1 Like
Thanks Zwift HQ. Just entered as C for a race on Monday. It’s a flat race. The distribution of As, Bs and Cs makes me think quite a few of the sandbaggers have been bumped up. Will be fascinating to see how much power we need to stay with the front group of C’s for the first 3 mins. My guess is it won’t be quite as insane as usual because the MAP monsters will be pushed up to C…
4 Likes
Another example would be the Bologna TT (I got my best 20 min result there).
2 Likes
Sandbagging is not really that much of an issue in women’s races in the first place, and random one-off women-only events rarely attract enough riders to create a quality race, so I understand the preference to start with mixed races only.
That being said, I hope the team has kept in mind that the ZP women’s categories have different definitions (different A/B boundary and no minimum absolute watts), and that these are being implemented correctly (with or without CP tweaks) alongside the mixed ones and are ready go live (or at least start getting tested) starting in March.
2 Likes
We absolutely know this
3 Likes
Can we access our VO2 data anywhere? I can’t seem to see it on the companion app or in Zwiftpower currently.
yes apperently but the most interesting is that MAP and Tmap accordingly to Intervals can’t be calculated from your efforts in this screen shot. I wonder if Zwift is using some kind of crippled MAP formular?
Welcome to the forum
see: Category Enforcement - How is my category calculated? [February 2022]
1 Like
Hi @Lars_Lange
That was before Zwift fixed the formula, That is now a B category rider.
ah so you’re to blame for my Friday night pain sessions and increase in vo2max efforts lol
4 Likes
If you literally iTT you might be able to flex the calculation a bit as your short term power won’t be high but anyone doing longer races will still have short efforts 2-5 mins climbs (and starts) that will push them to their limit on the power curve which would be taken into account with this new system.
it’s not like before where anything below 20 mins you were free to push whatever what you want.
2 Likes
I’d second (or third or wherever we are up to with this) this. The ftp tests I would have thought are exactly what you are looking for in terms of testing a riders absolute best output.
I hadn’t thought about it actually disadvantaging riders that do ftp tests, but I think this is a valid concern. I’m always surprised by riders that get the ftp increase banner in Zwift after a non TT race. I do a ftp test every few months and I’ve never come close to posting ftp increase numbers in a race since I’m constantly having to surge to stay with (or try to stay with) a group which tires me and so can’t usually actually do a 105+% effort over 20 minutes to improve my ftp.
The other reason I can think of is ERG mode. It is just different riding in ERG mode and not having to change gear and not being able to slack off.
Everyone has different things which motivate them and you’ll see my best results where I have the most fun. I love chasing the next wheel on Zwift, and will push incredibly hard to hold that next wheel, but I don’t love workouts, and so you won’t see my best results in workouts.
7 Likes
@xflintx If you’ll forgive me for jumping ahead, what is the plan for how soon this is likely to be rolled out across all Zwift races in the future? Clearly evaluation of the test events will need to take place, but presumably there is some sort of plan conditional on how those events go and what tweaking takes place.
Is there buy-in from Zwift to roll this out across all races in the very near term (as opposed to buy-in to run the tests but no further decision past that)? I’m sure the last thing that anyone wants is just to have it in the corner like the exisiting ZHQ Anti-Sandbagging races dead-end?
3 Likes
i physically can’t finish a 20min test without my legs tying up at the 14/15min mark but 30 mins of 60/20s would give me an ftp north of 5wkg which is closer to what TP estimates me as. i actually got my A upgrade doing exactly that. ftp tests are less than useless to me
1 Like
This is an interesting point. I forgot about this. It depends on how much of the power curve they are going to take into consideration. If they are making sure you are not out of line anywhere along a number of spots from, say, 1 min - 1 hr power, then this will be not an issue like it used to be.
Take the MAP check as an example. It sounds like it will be something like this:
Basically, you are not allowed to have a bump in your power curve. So if they can check for bumps like the one in the picture all the way up to 1 hr or even longer along this “ideal” power curve you get fitted with, plus a confidence interval, a span above that ideal curve where you are allowed to deviate, then for the first time race length will no longer have to be a race selection criterium.
But if the “bump check” stops at 20 min like before, then yes, race length will still matter as much. That is actually a cruising option, the sort of semi-cruise we have all grown accustomed to.
In current Zwift racing, signing up for a short race is risky if you push hard, because it might get you upgraded to a cat where you will get dropped in every mid-long distance race. And this is because your race average WKG is not what determines your cat but rather the 20 min WKG. So if you know you are basically up for an upgrade but still want to linger in your current cat and also want a good workout and want to win, then you can pick a longer race. Most “equal” opponents in your cat, those who don’t pick long races that often, will not be able to keep your average Watts or W/kg for that long. Which is exactly why cruising routes like Alpe du Zwift or Ven-Top is so easy - semi-steady effort, minimal draft, over a long time that will wear everyone down except you.
It is, however, during the first 20 min that people get upgraded in, I’m guesstimating, 90% of all cases. I’d wager it will be the same during the test races. And this is because of the hard starts that pushes up not only a measure like MAP but also the 20 min FTP/WKG. So anything above 20 min in race length is usually not that dangerous for normal non-cruising people who don’t want to get upgraded prematurely. If you didn’t get upgraded in the first 20 min, then the risk that you get upgraded past that point is usually low and for several reasons. It matters for cruisers, though, since they can often beat the pace of the first 20 min later on in the race if they don’t watch their numbers closely.
I don’t think long distance racers will get undervalued though, since they are not (only) categorized by their race average but also (and perhaps primarily) by e.g. 20 min averages, which will be basically the same as for anyone else. They will be the same since even a long-distance specialist can’t afford to keep Watts too low and stay too far from the front at start.
2 Likes
My understanding is they use the modelled MAP (based on your CP and W’), not from actual efforts, so bumps are not really relevant. They take it from your modelled (smooth) curve.
1 Like