Time to shelve w/kg as the category determinant

Years. It got worse with the huge influx during the past year with the Covid Shelter in Place business.

Many people have been pounding this drum consistently with ABSOLUTELY NO RESPONSE FROM ZWIFT HQ. None whatsoever. Very, very frustrating because we’re pretty sure this would be a fairly simple change but if it’s not, how about 'splaining it to the unwashed masses…

1 Like

This discussion started with a great post. In my opinion, this is physics. Watts/Kg determine speed of climb. On a hilly terrain with high vertical, riders with similar watts/kg will climb at a similar rate. On flat terrain, in it primarily Watts. Consequently, heavier riders do better on the flats (due to their higher power, for the same watts/kg) but suffer on climbs. Lighter riders suffer on the flats, but excel on climbs. There is no such thing as “riders in a similar category being able to get to the finish in about the same time”. It heavily depends on terrain.

Watts/kg is a strong stimulus to both, improve power and lose weight. At least, for recreational riders who usually have room for both. When I know that my watts/kg play a major role, it motivates me stronger than anything else.

That is an accurate analysis of the simple impact of power and weight in overcoming inertia.

The trouble is that is only the simple impact , it neither goes into all the other variables that impact this or how each interact with each other in different ways depending on a multiplex of parameters . This is not a nice neat set of linear straight lines , and all the other variables impact through that to determine relative speed .

Have zwift nailed this one shut , definately not , despite the view that things should vary differently on flat vesus climb parcours in zwift the simple evidence is that the same riders tend to dominate irrespective of what it is , something we agree should not be the case . Is it reasonable to expect them too have got this right … no . Is it reasonable to expect improvements to the algorythms , yes of course but that can only come over time with the accumulation of data and serious analytics. Now the contentious part , is there confidence this is happening fast enough , even Is any change happening . If it is its not very quickly , not very openly , not giving us confidence its being taken seriously . So we ride on until either zwift or someone else does do that . Zwift own us nothing and we own them nothing either . We pay a monthly fee we are free to do if it delivers something for us , if it doesnt we wont . We post in this forum we want to see change and the direction it should go. They can read and either ignore or deliver. If they ignore then they cant be surprised that they loose market share if someone else doesnt and comes up with a better offering .

I don’t think this post is about the game physics, it is about how racing categories are determined.

1 Like

While I’ve not been in a race where I was aware of this situation at the time, there have been lots of reports , and after the fact review of race results, where very light , usually adolescent, riders have unrealistic game speeds even on the flats.
The combination of light weight and short height seems to give an advantage over the larger rider with more watts.
I don’t do crit races but I assume these riders also accelerate quicker.
The physics is not linear.

Another area for improvement.

1 Like


It seems there’s a significant misunderstanding on expected results.

It’s unfortunate that everybody cannot win races. However, in the real world it is very rare to win a bike race outright.

I run a real life cycling club with hundreds of riders. Many strong riders. I can count on one hand the number of riders I know who have won a legit and competitive bike race IRL. It’s very rare.

Now, I am in no way supporting Zwift’s rudimentary W/kg system, or the absolutely terrible category based racing. They are both fundamentally flawed.

However, for those more experienced with real world bike racing, you’ll understand that certain riders generally win, others generally don’t.

It’s for a multitude of reasons. A majority of your potential success is based on two metrics. Your power curve, and your skill at racing.

I can use myself as a clear example. I often win bike races IRL. It is not because I am fitter than the other riders. Most of the time I am far from the fittest. I do however, race incredibly tactically and have very high anaerobic power.

To be clear, I didn’t train it, I was born with it. I certainly train it effectively before entering a race. This means that if I make the finish group, I will virtually always podium, or win. I totally suck at most of actual life, so don’t feel like you lost the genetic lottery :grinning:

If you race a criterium and in that race there is a born sprinter, who also races perfectly, unless you can match them over the finish both on positioning, timing and power. You will lose. This is more often how it plays out in real life, and on Zwift. Oh, and it’s exactly what we see play out in professional racing.

If you race Crit City and do not race efficiently and have a very powerful 15sec sprint finish, you will lose a vast majority of the time. As you would IRL.

Now, if you had a fantastic threshold ability and took that same crit racer on in a mountain top finish. It is highly likely you would win. BUT, the other riders with high threshold ability will also enter mountain top races. So, you’ll have more competition. Exactly like it is in real world racing.

The facts are, it is very rare to win a mass start bike race IRL. Very few people ever succeed at it. Of those that do, most are generically gifted and or talented racers.

It’s no different in E-racing. The genetically gifted do a majority of the winning. (And anyone on Zpower :joy:)

The solution. Have realistic expectations. If you’re not a sprinter, expect to never win a competitive crit race. If you’re not a climber, do not expect to win up the Alpe.

I’m not saying, don’t try and win, not at all. Race strategically, be very tactical with your course selection and use perfect timing. You may win, you may not.

Just don’t expect to win, or even regularly podium in highly competitive races. That is an unrealistic expectation, regardless of what category you race.

Obviously, with a vastly improved racing system Zwift could eliminate or greatly reduce the volume of cheating. They could ensure a fair and equal chance for all racers to win.

Take a long hard look at your power curve, learn where on that curve you excel in comparison to other riders. Enter events that feature decisive segments or climbs at your ideal distances or times. You’ll significantly increase your chances of success.

In short, hopefully Zwift actually listen to us and at least attempt to fix the rampant cheating. Until then, race smart, race hard and have realistic expectations.


@TheBandit I’m pretty sure that here in Zwift you are beaten most of times just by 155-160cm 50kg climbers in Crit City races… :joy: :joy:

Also the real Fabian Cancellara was destroyed by that kind of riders in his flat TT event, or Edoardo Affini (only Filippo Ganna beat him in the recent Giro d’Italia TT) also destroyed by climbers pushing 320W…

So, hope you don’t feel annoyed by Zwfit’s reality. :wink: :grinning_face_with_smiling_eyes:

1 Like

Tour de Zwift isn’t a marquee event, it’s a bunch of joke rides. Only reason people do them is because you get a kit, and the name implies that it’s a marquee event.

Exactly so and that’s the long and the short of the problem. If Zwift can’t take the event named after them seriously enough to establish a decent race format, then why in the world are we spending all this time on the keyboard trying to get them to see reason?




1 Like

Tour de Zwift absolutely is a key/marquee event for Zwift. All the events they advertise - such as the Giant Crit Crushers - are promoted to try and get people - especially newbies - to try different activities.

Some ride for kits AND others ride to try something different. That’s why I did my first TdZ stage, and riding that made me realise racing could be more fun than I ever thought. I also enjoyed chatting with people and experiencing the social side of riding online in big events.

I had a fun experience in the TdZ last year; I didn’t find it as much fun this year because changes to the chat function meant we didn’t see chat from as many riders in the activity.

Zwift should be making sure these events as good as they can be, the best of the best, but I definitely feel there are ways they could improve them.


TdZ was not a race this year, did not use W/kg categories.


Simple solution. The thresholds for upgrading category should be based on a sliding scale.

For example: upgrading from C to B.
A 100kg rider gets upgraded when they average 3.2w/kg.
A 60kg rider get upgraded when they average 3.5w/kg. And everything in the middle be in between.

It would even out the discrepancy and allow zwift to keep the W/Kg system. Category allocation would have to be automatic though, or at the very least tell you which category to enter. W/Kg shouldn’t be discarded, it is relavent, pros use it.

I am a recent victim of this problem. I entered a cat C race. Stayed in the main bunch for 4/5 laps but then got dropped on the last hill. I came 12th out of 20 but got DQ’d on zwift power with a WKG label. I was drafting the entire time except for after I got dropped and plodded to the finish.

So the current system doesn’t allow me to put down enough power to stay with the main pelaton without disqualifying me. Just for the record I’m 183cm tall and weigh 68kg. Tall and skinny but not an abnormal stature for a recreational endurance athlete. I would say this problem affects enough people for something to be done Fast.

As for weight doping. That’s a different discussion. But maybe all those serious enough to be a Cat A should have there weight verified with a video as part of their profile. Competitors could then see it if they want and flag it to zwift to review if the video isn’t clear or looks suspicious. Video should be less than a month old. But you wouldn’t want to do that for cat C or D… People who just do it for a bit of fun, it would put them off.

Check this out: Autocategorization Test Events & FAQ?

Lots of good information.

Also looking at your last FTP test I would assume your FTP is close to 3.5w/kg that is a solid B


I can’t understand this system.

My FTP is 252, I’m 72kg (3.5w/kg) I’m categorized on “B”, but doing these numbers I barely ride one or two laps with the leaders. After that, I’m in smithereens and I end the race alone.

The leaders put 4w/kg and above, which should have been Cat “A”. But they’re in “B”.

Hi @Alex_Santin

If you are referring to the Crit City Race you just did, then I would assume the results are not final yet. Those people over 4w/kg should be DQ.

The problem is people enter the wrong category. But Zwift And WTRL are working on a system to limit that. Try this series starting next week. Zwift Classics - WTRL

That’s not a bad idea.

I don’t race on Zwift but if I did I’d be shoved up with A on watts per kilogram because I’m very light, although to stay with them I’d need FTP around 40-50w more.

As much as I’d like this, that’s just not going to happen.

Whenever it is flat or downhill I get obliterated. Far more in Zwift than I do in real life.

In real life we have less of this problem because us officials (former in my case) know most of the riders and step in when they sign on for a race and move them to whatever category is appropriate.

Hi @Gerrie_Delport_ODZ, that was a great example.

I know the results can change. My point is, I completely lost interest in that race because I got dropped by competitors that shouldn’t be there.

Some days ago I raced the same course in cat “C”, and it was awesome. In “B”, no way.

1 Like

Wow, so much written about categorisation… the solution is very simple, it has to be based on race results. If you win or place highly regularly enough, you get promoted. If you don’t win (or place highly enough) over a period of time, you drop. That’s all. The precise details of what it means to place highly, and what historical period is considered, can certainly be debated. But any system that does not follow this general approach is broken by design. It’s not rocket surgery. Winners need to be bumped up into more challenging races, for their own benefit and for the benefit of the others in the same race. And losers need to find a lower level.

Trying to engineer some analysis of performance to match up people, that isn’t based on their actual performance in the one metric that matters (did they beat or lose to the other guy) is just fundamentally wrong-headed.



Yep, I completely agree with you. It seems fair enough for everyone and simple to implement.

1 Like