Zwift officially takes over Zwift Power

Zwift Insider had a great series on the topic of racing and improving the overall experience:

3 Likes

This is the correct path. Execution from an engineering perspective at scale is not trivial though and it requires a commitment to the path rather than something half hearted.

2 Likes

Now, now.

Let’s give Zwift time to port it all over to C++ before we criticize. :face_vomiting:

1 Like

Totally agree that an ELO ranking system similar to chess is the best solution, and matching players based on their ratings.

Also agreed on the artifacts of history leading us to where we are today. There are so many different ways to set arbitrary categories and I’m sure after much internal debate, Zwift went with wkg because it’s one of the simplest metrics to draw assumptions of relative skill.

However… going back to ELO, imagine if we had this working in Zwift today with the population calibrated to their respective ELO racing categories. In this world, what I envision is now instead of a battle of where Zwift draws the category ranking lines, people will have opinions on the key player skills and rulesets of the game that allows some players to do well and others poorly.

Instead of people complaining about where the lines are drawn, we will see opinion posts nitpicking at the details of the rulesets like:

  • Nerf the CdA advantage on shorter heights, I keep losing to short guys on the 1min flat sprints even though our wkg’s are the same and I’m doing +400 more raw watts!!” or…
  • We need more race courses that finish on a 5 minute rolling hill – think of us puncheurs!!

And then I start dissecting the rulesets of Zwift in my mind… and realize there is still much depth to be desired for Zwift to even be considered a viable, competitive game. And deep down, I don’t think there’s a good way to add complexity to a game where the main inputs is how hard one pedals over the course of 1 hour.

1 Like

It’s the first step in a long staircase. We’re not sure yet if that staircase is leading up or down, though. For sure, the concept of an independent app to do what Zwift was too lazy to do in their own app to their own event is not quite brilliant. Just read the how-to of a ZwiftPower account creation and link to Zwift, and tell me it’s a user-friendly process… Certainly Zwift needs to include event and results management within its platform. Can they do it, that is the real question.

Can Zwift acquire a company that does usable UI?

Good one.

1 Like

I agree. I basically said the same thing. We need zwift to do something. This is the first step. I did mention that they may screw it up as well. Would we rather they do something with a chance to fail, or do absolutely nothing?

1 Like

I personally see it as a good thing. Could they screw it up? Sure. Is it screwed up now, pretty much.

You do a race on Zwift, you choose a category, Zwift has all the information from every ride you’ve EVER done, yet they still let you choose any category you want. Say your FTP is > 4w/kg and the C category is < 3.1 w/kg. No worries! Zwift let’s you sign up for C, and wow, you got a win in Zwift, pat yourself on the back.

But but but, the true results are on Zwiftpower, where you would be disqualified had you actually signed up, but you didn’t as you wanted the win. I see it over and over. It wouldn’t be a big deal but those of us in the right category get absolutely blown apart by those levels above us. How fun is it to be off the back 3 laps into an 8 lap race only to find out you “won” on Zwiftpower? That’s crappy.

What concerns me is that Zwift hasn’t seemed to care at all about any of this. I’ve always felt Zwiftpower allowed Zwift to let them be the bad cop while they coddle subscribers that want to stroke their ego blowing apart lesser categories. Now that they’ve taken it over are they going to take a more fundamental look at correct categories and sandbagging? That’s yet to be seen.

6 Likes

WELL, at least we can all see a silver lining now that it wasn’t Garmin that recently purchased our beloved ZP.

5 Likes

Oh, so salty :stuck_out_tongue::laughing:

2 Likes

Is anyone else able to load the Zwift Power website? It appears to be down?

Yep, same for me

1 Like

back up now

Not sure I agree with that. If everybody only did Zwift racing, they all did enough racing to provide a statistically significant sample, and the goal was to have an accurate ranking of who the best Zwift racers were, then I’d agree. But seems to me that the goal here is to have a more level playing field in races which makes for more enjoyable racing. And that plenty of people don’t race on Zwift enough (either because they mainly train not race, or because they race in the real world) to make the ranking accurate enough to achieve that goal.

E.g. I’d never raced on Zwift before covid because I prefer racing in the real world and using indoor cycling for training. So I launched into the Zwift racing scene with an FTP of ~4W/kg in March when lockdown kicked in, and under the current system that meant I was entering A/B races straight away and after half a dozen or so events ZwiftPower was telling me I was now a mandatory A and I’d be DQed if I entered B. Now I can ride outside again I reckon I’ll only race on Zwift pretty sporadically. Under an ELO system I think there could be a lot of people like me who are new to Zwift and/or race infrequently enough that their ranking doesn’t reflect how strong they are and they could crush and distort lower category races. Under a properly implemented W/kg cat system, then it doesn’t matter how new I am or how many races I’ve done - a single FTP test, race or hard ride would be enough for the system to put me in roughly the right category.

The key being the bit in bold of course! In my case I’m honest enough and have enough real life riding buddies who race Zwift and would give me hell if I tried to sandbag/cheat, that I signed up for ZwiftPower straight away, put in accurate height and weight data, and entered A/B races from the start. Obviously there are plenty of people who deliberately or through ignorance take advantage of the current system. I see the acquisition of ZP as an opportunity to enforce the W/kg system properly. I think the W/kg system would work fine if the following were applied:

  • Minimum category enforcement at time of entry, based on previous training/racing data
  • Real time DQs for obviously dodgy numbers
  • PM or smart trainer (maybe HR as well) required for higher race categories
  • Post-race DQ reflected in the results table for numbers that are sufficiently suspicious or have been flagged (e.g. somebody pointing out that the 1.8m guy they ride with in real life appears to have shrunk to 1.5m in Zwift)
  • Maybe add an A+ category for >4.5W/kg in races with a big enough field, though as somebody who is currently 4.3W/kg I admit to some inherent bias here!

Of course you’ll still get some situations which seem unfair to some people. People with good hour power who can’t sprint and get beaten in crits. People who just scrape into the bottom of the next category and find they can’t win any more. But that will be the case under any system unless you start having different rankings for different events (TT, crit, road race) at which point I think the complexity starts to become counterproductive. Maybe there’s a case for races with big fields simply having categories split by 0.5W/kg instead of 1.

1 Like