Strava isn’t part of the core Zwift functionality?
Doesn’t that violate your rule of “if its not in companion it doesn’t exist”
2 Likes
Date in June in brackets.
But your comment prompted me to delve into my 90 day race stats…
27 race results that were from pen D (so removing the few pen A races and DQs)
Average field size 17.3 racers (average dropped by a few hill climb and TT events, where typically <5 in D)
Average finish position 5.25 (two DQs were pen D on courses that I tend to do better on, but I was technically “sniping” in Tiny Races)
And yet oldskool Zwiftpower Race Ranking puts me 4th and the new Zwift Racing Score puts me 3rd in D. 
Simply trying to demonstrate how flawed the oldskool Zwiftpower Race Rank and new Zwift Racing Score are.
2 Likes
So if we don’t have these volumes, no matter how great your racing pen or ranking assignment system is, it won’t really matter? There just won’t be enough butter to spread across the toast. Fewer race events on the calendar is the solution?
Many of the ZHQ races though do hit higher numbers, and if not enough for 16 pens, why not 6, 8, or 10 of them at least?
It took me a while, but I’ve moved to this line of thought as well. Anymore, I don’t really care if other people are height or weight doping or have a miscalibrated trainer, as long as, in the moment, the race is competitive.
If they’re a sandbagger blowing up the race by flying off the front or pulling a breakaway that the other riders couldn’t sustain on their own, or are a cruiser hiding their true ability while everyone else is at the limit and then sprinting from a long way out because they still have reserves that no one else can match, that’s disrupting the race and making it less fun.
As long as I don’t know you’re cheating (or you’re using a very inaccurate trainer) outside of those types of behavioral cheating situations, if you pushed me to the limit and forced me to respond and we both had fun, it’s just a C (or maybe if I get my lungs back again, a B) race on the internet. I got what I needed to out of it.
If you tell me about it afterward, I’d probably get pissed, but during the race, as long as you’re not blowing it up, my mind isn’t on who MIGHT be cheating. I just don’t have that bandwidth while I’m racing. If I’m spending time thinking about such things, I am not spending time thinking about strategy or pushing myself beyond the ability to have idle thought (and am probably not racing as well as I should).
On other platforms where I’m racing well-designed bots, same thing is true. If they’re pushing me to my limit and are realistic enough that I can’t tell if at least some of them are real humans, it doesn’t matter to me and how hard I’m racing and the strategic choices I’m making. Either way, I’m working harder and having more fun than I would be doing a workout staring at some numbers on a screen.
So what I want out of a ranking system if something that clusters people moderately closely by ability, and doesn’t allow outliers in the pens with them. Based on the flaws pointed out here the new Zwift Racing Points scheme doesn’t do that, so I’m still hoping that Zwift implements something directly on the site that does.
Heck, just allowing external links for category determination and pen enforcement, so an event organizer could choose to use the ZR.app ranking as category criteria and all I had to do was sign up on Zwift (or, probably ZwiftHacks), without having to jump through other hoops would make me ecstatic.
8 Likes
The problem is we have to setup pens in advance and are limited to 5 per event.
It would be great if zwift had some sort of matching making that splits pens on the fly but that’s not possible.
Some teams like tfc do have 8 pens but it means creating two events which for a lot of organisers is just an unwanted pain.
I believe ZHQ have a plan to offer more pens per event in future which should make things easier.
zmonthly events certainly don’t help the community events which is why a lot of them are struggling for numbers over the low season of zwift.
2 Likes
I’ve read most of the posts, but couldn’t get throught them all so apologies if my points or similar ones have been made already…
For me I think this new system won’t work as it is, you’re better off leaving it with cat enforcement based on power until this can be tweaked or made better, for a true fair way of racing there needs to be a fundamental change in the Zwift programming to enable race pens to be automatically populated.
IMO the pens need to be based on both power and race ranking, I’m currently at 666 points in the new system so I’ll be racing mostly B cat, however my max 20 minute power is around the 3.2w/kg level which is traditionally C so I probably won’t have too much chance of doing well which will put me off racing and I’m sure a lot of other people will get demotivated to race if they’re constantly getting dropped or the only way of doing well in a race is to stop racing until your rank is lowered as your old races drop off the system after 90 days.
I’m sure many will disagree, but I think the pens should be put in power based categories like they currently are then sorted by race ranking giving a C+ if your score is over 500 or C- for under 500 for example. However if there are not enough racers, say under 10 or 15 in each of those cats then all C’s start together… I’m sure it can’t be that hard to program this to happen on the fly. You could even consider having a staggered start line with the strongest C starting at the back or having a few seconds delay so you have to blob up the stronger riders and chase a small gap to the lower ranked riders.
I’m not saying this is perfect, there won’t ever be a perfect, but it seems to be a fairer way to me. I’m ignoring the William spanner in the works as he’s a known sandbagger who uses Sauce to monitor his draft the whole race and keep his watts to an absolute minimum hence him being able to race and win C without getting upgraded by keeping at 200% draft the whole time using his Sauce overlay which isn’t something available to all, and sprinting off in the last few hundred metres. Everyone else is doing the work and dragging him around the course. His ‘D’ ranking of 700 plus is just a blip and 99.9% of racers are not doing what he is.
The draft algorithm definitely needs work too which will make the races better. My last race nobody was putting out more than 3 w/kg on the front and we were doing 47kph or more on a flat section, surely that can’t be right.
I’m glad Zwift are looking to improve things though, it’s a great platform and one I’ve thoroughly enjoyed over the last 3 years or so.
1 Like
This logic breaks down quickly in A cat. Races just become noncompetitive, uninteresting, and lack fun. ZRL was a perfect example of this, no one could beat the guy on a ski erg machine or the guy on a handcycle because they’d juiced the translation of power so high. People tried though… by adjusting their own power up.
Maybe we need a different set of rules and rankings for lower cats.
A cat with no performance verification or “ground in reality” checks for riders could just be a dynamic pace partner event that punches over climbs.
I wonder how many people commenting even have a zwift subscription active 
As far as I can see all of them.
DM me if you suspect a bot.
[quote=“Matt [+R], post:341, topic:608136, full:true, username:Matt_Wheeler”]
I wonder how many people commenting even have a zwift subscription active
[/quote]
I would think that some of the commentors are truly wanting a better racing system whether they are current subscribers or not. I would like to see Zwift improve racing and have been waiting and waiting and waiting…,
After 7 years as a continuous subscriber and 66 years old I can wait no longer. My subscription ends in four days and IndieVelo is my new home.
1 Like
Is there a requirement that all abilities have to be invited/included in race setups? I thought I’d look at one of the Rhino events. e.g. ht tps://zwiftpower.com/events.php?zid=3740434
Couldn’t you consider just having these as B/C cat levels only, but you’d have 4-5 pens to play with? Give up on A and D in other words.
Many of us would prefer the ZR app versus what is and what is coming from Zwift.
I know the Club Ladder racing series is using the ZR app to establish ranking and categories for it’s races, so why couldn’t more race organizers use this system for their races?
Just want to show you guys point inflation is already happing. My recent D races
Even without the A races, the point inflation is happening.
These under under the old ZP system 600-0 lower the better.
Recent
Racing D tiny race 237 points
Racing D+ TFC Mad Monday 229 points
Yeah, a few months ago, winning the zwift monthly race in a field of 100 was only 300 points.
Back in Jan I won a Stage Race Tick Tock with 67 riders it was only with 355 points.
I won’t be surprise give us a few months we will see D races going under 225 points.
There are organisers and events using it.
ZR support a few ways for organisers to work, the main drawback is Zwift’s lack of ability to allow sign-in to a specific pen, or allow organisers of races to move entrants to the correct pen.
It is quite a challenge to get people to sign up to a pen A-D when they already have the mindset “I’m a B” and they can’t or won’t read the description asking them to join a different pen.
1 Like
[quote=“Keir, post:338, topic:608136, full:true, username:Keir_Wilkinson”].
I’m not saying this is perfect, there won’t ever be a perfect, but it seems to be a fairer way to me. I’m ignoring the William spanner in the works as he’s a known sandbagger who uses Sauce to monitor his draft the whole race and keep his watts to an absolute minimum hence him being able to race and win C without getting upgraded by keeping at 200% draft the whole time using his Sauce overlay which isn’t something available to all, and sprinting off in the last few hundred metres.
[/quote]
Just because I out sprint you, I don’t think bash or name calling is called for.
1 Like
I don’t use Sauce and probably never will, but I absolutely try to draft to minimize my power output. I want everyone else to close the gaps and drag me around the course until it’s time to do an effort. Not a sprint if I can avoid it - preferably 5 minutes, but that’s me and others should do what they are best at. If someone else will get on the front or close a gap, let them. “Never interrupt your enemy when he is making a mistake.” That’s just racing 101.
Your point about the game physics is on point for the Pack Dynamics 4.1 thread. When that comes out my approach might change, to whatever I extent I’m forced to.
Hey @xflintx,
With several days and several hundred comments behind us, we’re not likely to get much further on our own - so it’s probably time we hear from you and the Zwift team again.
To that point, though I’m sure I missed a number of things, I believe the main points of interest we’d like to hear specifics on are:
Are you/Zwift still planning on starting in the same manner?
If so, what do you feel are the advantages of the originally planned course of action - as opposed to something along ZR.app’s philosophy?
If you have, or are looking to make immediate changes, in what direction(s) (perhaps tracking concurrent models) are you leaning towards?
Are you open to tracking more events beyond the ZRacing monthly races to get a larger sample size?
Has/will anyone at Zwift reached out to Tim’s team for advice or help?
If not, some thoughts on why this isn’t necessary
Any other new information you’d like to share
Apologies for putting some pressure on, but assuming the intention is to have everything ready for Season 1 of the ZRL, time is of the essence.
Thanks again for engaging with us and looking forward to this system being a huge success!
2 Likes
yes you can do that.
Herd do something similar with their weekend event targeted at just d riders with 4 separate pens for riders that are sub 2.5
What would be interesting is to just run test races base on the Zwift Racing Score. The debate is rather hypothetical without actual races to base off of.