Category Enforcement Test Events - 21-28 Feb 2022

Sit and do nothing waiting for things to turn out for the best by themselves? I have pretty bad experiences from that approach in general in life and try to avoid it if I can.

You see, from their communication in the recent past, which has been pretty vague (basically: we’ve got stuff cooking, just you wait and see) I can, with a little imagination, see two possibilities, something like:

A) They have taken the first logical step towards better, fairer racing by testing cat enforcement. They do also see the need to move to category definitions based on past race results rather than some Newtonian measures. But it’s a big transition, one where they have to reshape racing across the board, rally the company and ultimately involve the grass roots, the race organizers. This is obviously going to take time. We’re looking at a release earliest around the start of the next winter season.

So while they were at it, implementing a cat enforcement test as the first step, they decided to throw in a new temporary categorization that would ease the pain we’re in a little while they keep working on the real solution.

There were talks with representatives from USAC and FIS around Oct-Nov '21 to learn more about the experiences of the USAC points system and the FIS ranking in practice. They also invited TrueSkill and some other competitors for product demos of online ranking/matchmaking systems.

From what they learned they decided that even though a points system would be simple and intuitive, it posed a set of unforeseen problems. TrueSkill, although a much more expensive solution overall, seemed a better fit for the Zwift online environment and the diverse Zwift racing ecosystem.

They have been in negotiations with Microsoft both regarding a price tag for the license itself plus the adaptation of the model, which requires a joint effort between Zwift and Microsoft specialist consultants. Of course, because of the ongoing negotiations it has not been possible to talk openly in here about the shape of the solution.

But you should know that every drop of sweat you pour into the test races will come to good use as initial seeds in a preliminary calibration of the ranking system. So keep at it and be patient, please. It sounds cheezy and not very convincing, they know, but what can they say more than “Good things come to those who wait”? Just have faith, you’re gonna love this once it goes live!

B) Apart from adding more options to play with for race organizers (cat enforcement: ON/OFF), there are no further plans than this to reshape Zwift racing. Not at this moment. Maybe later. There are options, yes they realize that, but… We’ll see.

What do you think is more likely, A) or B)?

3 Likes

Sounds about right, except the roadblock in scenario A is probably going to be developer resources, not something like 3rd-party licence fees. And that basically kind of brings both scenarios together, because nothing more is going to happen until dev resources are committed to continuing to improve racing. So the real question is, has such a commitment been made for real, or will the devs behind the category enforcement work next be working on the new start screen or pace partners or something like that, or is there now a dedicated team working permanently on racing-related things only.

As of late, the product manager for the clubs functionality has been exemplary in communicating the status of things on the roadmap here on the forums, and the same kind of transparency would probably be helpful here as well.

4 Likes

This is a pretty decent road map isn’t it? Especially compared to what we had before.
https://forums.zwift.com/t/category-enforcement-trial-next-steps-february-2022/578545/29

1 Like

I see nothing about a results-based system, even in the long term. That speaks volumes.

1 Like

… and many drawbacks, not there least of it being its opaqueness, e.g plenty of the data used too calculate your category not being generally visible because it was a workout, incomplete race etc

But let’s not tell the Emperor about his New Clothes (Cats), eh?

1 Like

It’s a good start but definitely not a roadmap. I would like to see something with a scope of at least a year or so, and also including things which are not going to get done (for now or ever).

3 Likes

Yep don’t disagree with any of that.

B is all i want… more options for organisers. why complicate it. the pen enforcement works well, it’s a good addition, especially because you don’t even need to be on zwiftpower, which means that people who are new or otherwise uninterested in even pretending to be in their correct category can’t just join D with a 4wkg ftp they recorded on an elite muin or anything like that and rip everyone’s legs off. i can take or leave the category threshold changes

3 Likes

Maybe they don’t think a results-based system is a good idea?

I also don’t think a results-based system is a good idea, by the way. So the absence of a roadmap that includes this doesn’t concern me, although if you think this is what should be designed then you are welcome be aggrieved that is has not been mentioned. I personally think it has a huge number of drawbacks .

I would certainly like to see a longer term roadmap with an end-goal mentioned. For me this is a new global ranking system, fully integrated in the Zwift client.

If we take the logical next steps from where we are today, and allow race organisers to determine the thresholds for each pen, then that creates a problem. No rider knows where they stand any more, how they match up to other riders. A ranking system fixes this, and also means that race organisers can use these rankings to sort pens.

So my end-goal is:

  • Organiser defined pen-splits
  • A new global ranking system baked in to the client, making every race (and part of a race) meaningful
  • Option to sort pens by ranking

Zwift’s may be different, but yes it would be nice to know it and be able to discuss the direction.

What is yours?

1 Like

He did, even trying to degrade me on the basis of my ZwiftPower stats… even though all I’ve done on Zwift recently is either lead my weekly ride (easy z2 for me) or have it running in the background while coaching group sessions. :man_shrugging:

It’s like he’s never heard of Covid or that maybe, just maybe, I haven’t raced in years because of sandbaggers like him who whine about being forced up, even though they finish in the front of the upgraded pack.

If it’s no fun for him, good. At least the rest of us can have a race without it being ruined by selfish (multiple expletives deleted :wink:)

2 Likes

Is your ranking system based on who you beat in a race? Because that’s what I meant. Not a points-based system (which I did not mention), but one based on who beats you and who you beat.

Aren’t we talking about the same thing?

As I said, this being missing from the long-term plans speaks volumes.

2 Likes

I took results based system to mean based on placing in a race, so yes a points system. If in fact you think a rankings system is better, then we’re aligned.

Whilst not mentioned yet, I wouldn’t say it speaks volumes, as actually nothing has been mentioned yet regarding a longer term roadmap. They did have a job open which had experience with global ranking and progression systems listed in it’s requirements. That’s a hint at least. Would be good to know. Maybe now dialogue is starting to open a bit, and they can see the benefit from great communications with the lead set by Eddy on Clubs, @xflintx or @Mark_Cote can share a little more.

5 Likes

I thought half term was last week?

1 Like

Simmer down a bit folks, Zwift isn’t that important in the scheme of things. There’s a war kicking off!

3 deep breaths, in and out.

8 Likes

:man_facepalming:

4 Likes

Is this sort of ranking system really all that different than what Steve may be trying to refer to though?
I kind of feel like they both fix a similar issue, or can potentially.

Steve’s solution I believe takes more of a human touch than what programming will allow, or just needs one extra bit to make it work. James I think you’re on the money though.

Zwift has our stats, it can easily compare and check to see if we raced within say 15% of our capabilities or not (or, whatever percent seems reasonable, I don’t know I’m too new).

So where’s Steve’s solution fails, could be it’s too easy to take advantage of just losing to lose, to be dropped down… if it’s too simple that is.
But that doesn’t include Zwift comparing stats.

There’s nothing in place according to the statements thus far, that says people near the crit category edges will get bumped.

The ONLY thing that puts us in categories is our wattage capabilities, that’s it.
It’s a fine starting place. But… that’s not a finish line either.

So is the cat enforcement system working for me?
Well I’m glad it’s being looked at. But I would argue it isn’t working for me.
The person that’s 1 digit over the line, and knowing I won’t be competitive for a long time.

Meanwhile there’s an entire potential that folks in cats below me, are winning left and right, but they can also just watch their watts, and still “cheat” the system if you will.
It’s probably a lot more difficult, but this doesn’t all-out fix the issue at hand.

For clarification; I can literally bump myself back to Cat C, right now, by going into settings and changing my weight by like 4 whole pounds.
Because I’m so far away from the VO2/MAP, at my going rate I wouldn’t hit Cat B under that for a long… long time.

This system still majorly relies on people putting in their weights accurately.
There MUST be something else in place, that doesn’t take watts (and w/kg) into account.

Until Zwift forces heart rate tracking, accurate weights (which it probably never will for both)
We’re stuck with a system that is still relying on two different approaches to your output.

It needs to be able to compare outside of our own results, see that we’re winning or severely losing to much, and be able to bump people around the edges.

Again, what we have now is probably just fine for a starting place, but this cannot be the end.

TLDR:
Someone can still easily watch their watts, fake their weights, and fit themselves nicely at the top of a cat. In its current form, this Cat Enforcement, will allow someone to do this, and win all day every day.
Meanwhile myself just playing fair 1 single digit over the line to be cat bumped up, nowhere near VO2 capabilities to fit into this cat, can lose, all day every day forever, and nothing will change.
Person in cat adjusting their weights, and staying within the definitions of VO2 can continue to fake their weight, and win, and still get stronger.

Again, this same system, is entirely at fault still, with one of the biggest issues Zwift has to handle.
Rider weight.

1 Like

By that comment, I’m guessing toys are being thrown out of the pram but such are the joys of the Ignore function of the Forum… :smirk::sunglasses:

2 Likes

Your issue is that you are at the very bottom of a category. We get it, you don’t like it.

The solution is to have flexible boundaries set by the race organiser, as they are in TFC Mad Mondays, or FRR Tour Series. This problem already has a solution, it just needs to be implemented and enforced in the core client, rather than managed through hidden race series and private links.

5 Likes

Unless I was being thick, which is very possible, I looked at weight doping by temporarily changing my weight to 95Kg iirc on Monday… I still had to enter B as a minimum.

I guess for me it still wouldn’t work; as it’s pulling historic data.

So if you alter your weight, it’s not changing the historic data, just changing current results.

So yes, weight doping can still occur… it’s just going to take 90 days for it to be taken in account. (Probably not an issue for the people I see on ZP that weigh under 100lbs and are full grown muscular adults in their profile pictures)

So there is that; but one can argue that “cheat” can still exist for anyone thinking about it now, or taking advantage of it already.

@James no you’re missing my point.
It’s that there’s nothing in place to be shifted around.

Say the event creators move things around, but I’m still forced into a cat that’s above my capabilities, and I do 30 more events of theirs, and still come dead last…

I’m not going to register for any more events, because I know the next 30 nothing will have changed.

I’ll say it one more time for everyone in the back.
This system is fine as a start place, this is what you’re also saying James.
However, it is not, and can not be used as the sole primary ruling. There has to be some sort of ranking system in the background to show imbalances.

Meanwhile I have literally no clue what TFC or FRR is because of how convoluted the current system is at finding events and again, why the ABCD cat system is an issue.

How do I, still a newbie to all of this, find things that fit me, guess?
Asking seriously.