Table of Contents
Zwift Racing Score (ZRS) went live as the de facto race categorization method on October 7, 2024. But that doesn’t mean Zwift has finished its work with ZRS! In fact, they’re working on lots of ZRS improvements in parallel, and it’s become rather difficult to keep track of what changes are being made, what’s in the works, and what Zwift is still considering for further changes down the road.
Related: All About Zwift Racing Score >
So I decided to create this page as an up-to-date tracker of Zwift Racing Score development. First we’ll list work Zwift is currently doing with ZRS, and below that you’ll see potential work Zwift is considering. Lastly, I’ll list the changes Zwift has completed.
Current ZRS Projects
Score History View
Zwift keeps a record of each rider’s racing score changes, but that history isn’t visible anywhere… yet. Riders have been asking for a way to see this history since the early days of ZRS testing, and Zwift informed us in January 2025 that they are actively working to create such a chart.
Potential ZRS Projects
The following changes are being discussed by Zwift staff, but are not currently being worked on:
- Improvements to the volatility of score movement at lower ranges (est Dec 2024). Scores in the lower ranges aren’t moving as quickly as Zwift would like to see, so they’re looking at ways to change that.
- Looking further back than 90 days for the seed score (est Jan 2025). This is an important change in regards to seeding riders accurately and thus reducing cases of underseeding, which negatively impacts race experience in the lower categories.
- Arrow indicators when min effort mechanism kicks in (est Feb 2025)
- A historical view of how your score has changed over time. Zwift said in this post “We are planning to introduce a view for viewing your score history. We’re already tracking full history in our backend, it’s just a matter of implementing a view of this.”
- Dynamic pens: dynamically grouping riders based on signups to automatically provide for the most competitive and race groups.
Completed ZRS Projects
The following changes have all been completed since ZRS originally launched under Zwift Labs in June 2024.
- Re-introducing Zwift Racing Score Decay, Rewarding Personal Bests with Instant Score Updates, 30-Day Max Score for Category Enforcement, Score deltas in race results (September 25, 2025) See forum post >
- Event Category Range Revisions (January 2025): Zwift modified their ZRacing category ranges, creating three different events. A new set of “Advanced Category” races for riders with a Zwift Racing Score of 650+ was created, and races for riders at 700 and below now feature tighter, alternating category ranges.
- Tiny Race Scoring Improvements (January 7, 2025): Because Tiny Races are short races in quick succession, there was an issue where scores weren’t updating fast enough race over race, causing inaccurate results until the score finalizer ran 12 hours later. Now you will see a more accurate score ahead of the final processing. See forum post >
- Score Finalization Fix (January 7, 2025): Zwift says, “An issue was fixed with scores calculated from races held between Dec 30 – Jan 6. We identified an issue where scores weren’t finalizing as expected during this period and have fixed the issue (and our error alerting). During this window, you may have seen scores that didn’t correctly consider past race results and therefore were off by a few points. All results from this window are now fixed.” See forum post >
- New Seed Formula (December 4, 2024): The previous seed formula was based on your 90-day power bests for 30 seconds and 10 minutes, but Zwift changed it to be based on riders’ 5-minute power, both in pure watts and w/kg. Seed v2 launched December 4, 2024 (see forum post), then v3 launched December 6 (see forum post).
- Race Requirement Lowered To 1 Activity (December 2, 2024): Zwift added an activity requirement in late September so riders had to have at least three 10-minute activities recorded in the last 90 days before racing a scored event. In December, Zwift reduced this requirement to just one 10-minute activity because they “received feedback that the barriers to race are too high.”
- Requiring a Minimum Effort (Anti-Tanking) (December 2, 2024): With a small minority of racers purposely “tanking” their scores by entering races and essentially soft-pedaling to get a bad result and lower their ZRS, Zwift implemented changes to detect and stop this sort of intentional score manipulation. The first release of Zwift’s anti-tanking measures went live in December, and you can spot it in the race results… if the result has no arrow that means anti-tanking kicked in and their result was not scored.
- Decay Turned Off (October 29, 2024): “The original intention with decay was to account for decays in fitness after going an extended period of time without riding. After evaluating the functionality and reviewing feedback, we have decided to disable decay for now since it doesn’t seem to be providing the intended value. We may turn this back on later after making some improvements to the mechanism.” People will see small increases in their scores once it is turned off.”
- Podium Bonuses: The top 3 finishers in each scored race are now receiving a podium bonus meant to help move racers up a category faster if they are regularly winning.
- Finalized Subgroup Score Ranges: Zwift has defined the score ranges they will use for Zwift-owned races and recommend for community organizers. (For more information, see this forum topic.)
- Faster Score Calculation: Scores are calculated and adjusted as each racer crosses the finish line, instead of being finalized after 12 hours to account for DNF’s. This will eliminate issues where individual scores would drop when checking the day after a race.
- 3×10-Minute Activity Requirement: Riders must now have at least three 10+ minute activities recorded in the last 90 days in order to race a scored event. This helps ensure an accurate seed score, to improve initial subgroup placement.
- ZwiftPower Score Range Filters: ZwiftPower.com will be updated so users can continue tracking General Classification (GC) standings after event series like the Zwift World Series move to score-based subgroups.
- Bug Fix for Stuck Low Scores: Zwift says they’ve “fixed an issue where scores that drop too low get stuck with low values even after winning several races.”
- New indicator icons in race results to show if you hit your floor value or had a major jump based on setting a new 30s to 600s best (Aug 22, 2024)
- Score volatility fix: “We will also be fixing an issue that prevented score from moving as quickly as they should when a new 90-day power best/floor is established”
- Fixing web results so up/down arrows display correctly
- Slowing the rate at which your ZRS decays (July 8, 2024)
- Showing your updated score when you return to the game homescreen after a race
- Changing ZRS-based bands in Zwift-owned events to use different score ranges: these score ranges have been modified several times as ZRS has developed over the past months.
Forum Posts from ZHQ
Zwift staff shares information about ZRS updates via the Zwift forum. Here’s a log of these update posts:
- September 25, 2025
- June 27, 2024
- July 8, 2024
- August 22, 2024
- September 19, 2024
- September 25, 2024
- October 29, 2024
- November 18, 2024
- December 3, 2024
- December 6, 2024
- January 7, 2025
Questions or Comments?
Share below!
Classifying 650+ as “Elite” has to be joke, right? Good one, Zwift!
He literally said “numbers entirely made up by me”.
I also made up the “elite” title. No idea what Zwift will actually call these…
Sorry, my bad!
Elite by definition is a select group more advanced than the rest of the population – there really aren’t that many zwifters at 600+.
Mine is 820 and there’s a vast difference between me and people with high 900s. I’m not elite and never will be
Thanks Eric for again a great summary of ongoing and past activities on the ZRS. The transition for some has been long and painful and even if I remain positive about the end result. I myself am disappointed from my personal perspective of the transition process. As a 90th percentile Zracing and heavy Cat D rider I initially felt I had been overseeded. It’s not fun anymore when you can stay in the lead or chasing peloton of a 50 rider group one day and you end up in the tail end riding by yourself the next. I was counting on ZRS adjustments as I continued to race but score adjustments seem to take quite a long time either up or down. A lot of old close competitors have not been racing exactly because of the scenario I’ve outlined above. Those who do seem to be experiencing similar impacts. The race I did this morning is a good example (Going Coastal) at 8:13 Eastern. I stayed with the peloton exactly 6 minutes before being dropped. Most riders were old Cat C riders. The old Cat D riders finished at the tail end of the group 4-6 minutes behind in a 19 km flat race. A few went DNF and I can understand them even if I have never given up on a race yet but it does get harder to push yourself to the limit when you are not competing with anyone near you.. I came in 33rd out of 38 who finished (initially 44). My ZRS score went down from 195 to 192 and I beat people with scores well over 250 who are probably in the same position I was initially as I was seeded somewhere around 255. I can understand the fact that Zwift may have moved the line and that I may find myself in the same situation as when I started Cat D (at the bottom) for a while. I could live with that but right now things just seem still unclear and in flux. I may know definitely if I end up in the sub 180 group and finish close to 90% percentile. At this rate it might take another 5-6 races… and not the tiny one’s
One thing I would love to see is in the ZRacing series if they slightly altered the bands on different days:
i.e. on Monday / Tuesday the racing bands are like the “snazzy race event” bands Eric indicated above, but on Wednesday / Thursday the bands are changed to the following:
And Friday / Sat / Sun slightly different bands.
I think that would be a great way to have competitive races as more people tend to race when they’re towards the top of the band. The issue becomes what to do with that top band – I’m not sure what to do with that 600-650 in the above.
Please add the option to opt out of a racing score for those of us who don’t race.
If you don’t race what’s the point of opting out? In such a case it’s just a number, which shouldn’t bother you.
Maybe the point is to not give the impression that you are a terrible racer (with a mediocre score), but rather that you are not into racing at all? 😉
I’m a group moderator who approves quite a number of new members. I see some ZRS listed as “–” in Zwift Companion when I check. Seems to cover it.
Looking forward to the minimum effort piece. I occasionally ride races at z2/z3. Not because I’m trying to tank my score, but because I’m trying to create accurate course notes including distance and elevation changes from the start pen onward (not just where they plop you down for a free ride) for big races that I actually care about.
Prior to ZRS, I could just join an event on the course and then do an easy route recon of it to get that data. Since ZRS (or, more accurately, since I started caring about my ZR.app vELO score), I’ve stopped doing that and only been doing route recons when there was a race on the course that fit into my training schedule for a hard day because I feel the need to go all out on them to not hurt my ZRS (and vELO). Being able to cruise off the back to get that data again would be really nice. Honestly, if they would just say that events where your average HR was less than 75% of your max (made up number that they obviously wouldn’t publish) don’t count toward your ZRS that would make me happy.
I totally get that I’m an edge case.
I was a D rider (now back into the bottom of C) with an initial seed of just under 400. I was racing B riders. Farcical. My score is now 327, still far too high for my abilities and I’ve lost all inclination to race anymore. No point when I’m dropped on flat races in the first few minutes, and dropped in seconds as soon as the road goes above 2%. I could race in old C cat and hang on to the pack for most of the race if it was flat, only getting dropped as the pack picked up the pace at the end. That felt right, but ZRS is still after all this time and feedback still a shambles
The old cats mostly worked – a few gamed the system, a few complained that they were being promoted because they were lightweight, but the majority were ok with how it worked. All Zwift needed to do was introduce a promotion element based on results and not power, and equally a relegation based on results. Instead they have, and are still, trying to re-invent the wheel.
Lots of scientific terms in the write up to mask the lack of progress and lack of admission that they have this so wrong. Along with the greedy price rise, Zwift are alienating their user base and will lose a lot of riders to other platforms. Whether enough leave that makes them take notice, only they will know. Not good..
The changes notes above should remedy the poor experience you’ve had.
Zwift should do a better job showing all riders their actual power profile ala Coggan levels. A lot of riders in the lower Cats have massive holes in their profiles and no amount of tweaking will solve an A sprint with a C/D FTP. The old system almost rewarded the casual racer profile with a big sprint in sub 30 minute races. Reality is FTP matters far more than everyone wants.
With all due respect, can you try harder?
They should use compound score for seeding. Just for you to know, If you go to zwiftpower and pick any race result, you will find that compound score for the average power (entire race) is the data the best correlation with the race result.
CS works well for those that race a lot and seems best overall but goes crazy when used to classify casual racers. FRR found this out the hard way and was too much work to manually move riders around. Ladder challenge races as a qualifier would help sort people into their actual groups and being mixed cat creates a fun environment for everyone racing their peer.
Yeah, Zwift has tested compound score for seeding, and found it’s not ideal for their entire audience. It’s not an easy nut to crack…
Great to know about that. But why it didn’t work? It’s the same that w/kg, but with an aditional depth from the absolute number produced by the very same effforts.
Where did you find this?
Adding podium points to an account stuck at 11 ain’t a fix.
What about age related, master’s divisions ?
Bless you, sir.
Hard to believe after 10 years that this doesn’t exist as an _option_. Exists in real life for many reasons and yet is treated with absolute derision when brought up around zwift.
Herd Racing League had a four-week age group series but hasn’t repeated it.
Interesting Paul. Do you know if it was well accepted Four weeks doesn’t seem like much time for something to catch on and get traction. I don’t recall ever seeing it as an offer. As you probably know I am a big Herd fan. I know, I fly a lot on Saturday, but its a great workout and I am riding with people who I am competitive with once a week. There is always a Masters race out front. And I am usually the oldest guy in the group.
As I said in my post, if they would wake up to the fact that they have a very large senior population, maybe they would pay attention. Seniors use Zwift because it is safe and convenient. We can’t afford to get hurt. If you ride, you are going to go down sooner or later. The only time I ride IRL is when I am training for an event.
We would end up with “age” doping.
Seventy-five year olds with ftp’s of 400+. 😉
Cheaters gonna cheat.
400w is my 15 second power.
Couldn’t agree more.
Anti-tank definitely won’t work if people are intending to tank. How hard is it to:
Swap into a MTB.
Wait before the finish line.
Start slow out of the pen and let everyone go first.
Pedal hard on the flats and soft on the climb.
To come in last isn’t that hard.
As for improving the algo? What did they do before? Didn’t they run test? This thing been in creation since 2023! So far it’s like the beta version of Zwiftracing.app and they are straight up copying it. I have no issue for them copying it but why does it take a year again? Zwiftracing app is tune by 1 dude, Tim, how the hell is Zwift so slow to improve ZRS with a freaking team?
This is like one of those 15 managers and 1 engineer thing?
I agree with this. I am all for preventing people sandbagging but ultimately there was always be a way to cheat the system atleast partially. That said it looks like it will be much harder to do with ZRS than it is with the w/kg system as you will only be able to adjust your score down somewhat.
What the Zwift would detect as tanking is my wife on my account. Missing of a good Family plan is the real reasson.
Interesting…. I’ll open by saying I’ve been a fan of ZRS since the beginning. The possibility of regulating races better, oh yeah! Looking at things you’re reporting there’s a lot of moving parts. The first is the new seeding. Seeding and how people rank is critical to a system like this. That said, the current system has a number of riders who are Cat Ds, some who race, and many who don’t with any regularity who all have a ZRS of 11. Matter of fact it’s kind of a dead zone. If you have doubts, look at riders who finish in Cat E for Herd Beginner Races this weekend. Hopefully the fix Zwift is implementing addresses the lower Cat D riders too.
There’s a fallacy in the ranking you’re suggesting Zwift is going to use for Event Category Ranges. The range of 0-250 and then going by 100s is the same fallacy Zwift set up with Cat D being from 0-2.625w/kg and then having the other categories go up by much less than 2.625 w/kg. Please… this is distinguishing between Cat A, B, and C without recognizing that there’s difference in Cat D riders. Do a finer fidelity! Watching upper cat Ds finish a race in 1 hour (with or ahead of low Cs) and having lower Cat Ds finish a race in 1.5 -2 hours is no fun for the lower Cat D and does it really promote lower Cat Ds to challenge themselves except for the few?
I’m saying this as one who started racing as a medium Cat D racer and now has a ZRS about 250. Without good means for newer Cat Ds or to enable lower or medium Cat D riders to grow, you’re limiting the growth of racing on Zwift.
Organisers can actually determine their own racing categorisation for example eSRTs Ultimate Women’s Mini Races actually look to be more beginner friendly and it seems all categories A-E are accurate we use the following ZRS cat scores :
New Racing score categorisation:
500-1000 (A)
350-500 (B)
200-350 (C)
100-200 (D)
0-100 (E)
It gives new racers motivation to want to race and therefore improve their performance
It’s done but so limited. How many event per week vs the number that are set? HBR runs 6 events per week. Another organization runs a ladies only event focused lower and Zwift is guiding toward A, B, C, D because they have 5 pens. Tighten things up or promote greater division to promote growth.
500 to 1000, are you joking? Why would anyone join that bracket? That’s everyone from strong C to A+ in a mass start event.
It would mean then we will need at least 7 zwift Racing Score Categories and we did have quite a few women join that bracket the lower brackets were more accurate and if it encourages new riders to race then I am more than happy
That would be great. Each organizer came time the bands to their needs. I also wonder if Zwift could have some auto band thing where 125 riders join, there are 5 bands so it equally dividers the field into each group. Indievelo has this and it works extremely well.
Big advantage is you never know if youre the fastest, slowest, or middle of the pack. It’s great.
My last 2 races have been ZRS and I’m in the 350/450 and it’s a C cat race too, but I’m getting smashed and at the finish the top 10 are 3.7w/kg for 40 mins! It’s just starting to get a little disheartening now!
All Categories are experiencing the same as there are and will always be strong Cs who should be Bs that are in your category.
Whatever anti-tanking measures are supposedly built-in to the game they’re clearly not working very consistently.. I can see a guy with a youtube channel who does the tiny races under Cat D zone 2. Last weekend he got three first places and one second ,which moved his score up to over 400. During the week he then did numerous community festival races, even winning one of them but obviously tanking at least three others he took part in. And his score as of this morning was 351 so it had dropped 50 points after just three poor race results. He then did the tiny races again today ,placing 3rd,5th,10th and 12th in big fields. Yet his score is still 351.
This points to a surely arbitrary and unreliable metric they are using to adjust racing scores.My own score only ever varies by a few points after each race,despite some widely different placings in the results.
The same thing happened again today and last week in Zone 2 B tiny races. Same guy consistently hammers the field then soft pedals during the week in community races to make sure that he can do it again in the next set of B level tiny races.
It would be good if they take races like ladder races into account. As you have a 3 rank spread (ZR.app) and finishing in front of teammates is not beating them.
Better is to just ignore these kind of races.
One of the reasons zwift chose to base ZRS on the OpenSkill model is that it can be applied to team events. The functionality isn’t there yet but it’s something they want to implement
Great source of information as always Eric.
I’m an injuered former top end B rider and I’m currently at 475 ZRS which on this mornings tiny races performance seems to be spot on.
I was in the ‘c’ group and was blown away at how strong the competition was and looking at the zone 1 results the best of my group would have been right up there in the group above.
Hopefully the changes will come good in the long run but I don’t think they are too far wrong just a little slow on moving people up/down
i gained 5 points per race and I am at 360 right now. I am definitely a mid B rider. I race twice a month. How long does it take till I am in the right category?
Zwift is really losing me as a subscriber with this scoring. A C category rider getting dropped quickly and finishing near the back by yourself isn’t that enjoyable and doesn’t help with training. The winners were putting out a full 1 w/kg more. It is like a USA Cycling Cat 3 racer being in a Cat 1 race. Not that fun.
Exactly. People simply want to race in groups of similar abilities. Racing score is not delivering this in any of the recent middle-category races I’ve done (the equivalent of what used to be C), where the race has been destroyed by A riders doing huge numbers at the front. And at end, their score goes up by a few points. I was all in favour of improving the old A-D system but this is worse, sadly.
Only 10 subgroups in this hypothetical future? They also need to add age categories for each one. They should also allow for division based on numbers of hours per week you train. And number of hours you work. And if you live over 4000 feet elevation. Eventually everyone will be in their own category and everyone can be a winner, which is the only thing that seems like it will actually make serious Zwift racers happy.
Well said. People never complain this much with IRL racing. Yes, there’s always talk of sandbaggers, but people think that Zwift categories should be perfectly tailored to their exact fitness and only include others who happen to have their exact fitness. The reality is that even if that was possible, there would still be winners and losers and the losers would still complain that those who beat them were sandbagging. Your solution of everyone winning is the only way to make everyone happy!
As a rider at the lower end of “A” category, I’d really like to see the equivalent of low A / high A category.
I enjoy a challenging a race but with no upper limit above 690 it turns into the Wild West.
I see many regular riders lowering there race score to sub 690 deliberately and although it’s not something I personally would do, I can see why some riders do.
Until they increase the number of Pens to 10 (smaller cat bands) ZRS will be no better than the old ABCD system
I mean… you just doubled your power numbers in terms of w/kg. What would you expect your score to do?
I love the anti-tanking feature, my only concern with it is how is it going to decide whether I am tanking (and thus I shouldn’t lose score) or I am having bad day/detrained (and thus I should lose score).
If a person returns after an injury/illness and is stuck with a high score because the system thinks they’re tanking, that could lead to a hell of a lot of frustration.
I would like to think it will depend on the individual and they will look,for example, at what your race winning history is and your current score and whether it is near to a threshold between two categories. The people who are abusing this it at the moment and manipulating their scores are those who want to carry on winning in their current category (which they are doing with monotonous regularity ) but at a certain point they have to get that score back down again,so they will ride a number of tanked races in quick succession. These are the people the system needs to catch ,not the rider who is usually mid- table anyway (never wins or even podiums) but tanks the occasional race for reasons best known to themselves.
Essential changes like a new seed formula that definitely removes 30secs and ideally replaces 10mins with an interval 4-6mins in duration are happening too slowly.
The balance between pure Watts and W/Kg in the curent seed score formula is far too biased towards pure Watts.
It’s placing far too many ex- Category Enforcement Cs under the old four pen system into the weakest of five pens under the new system for events like Tiny Races with modified pen ranges such as the current 0-260 for pen E.
The bonus points for a podium finish are too low and I’d argue there should be bonus points for the top ten finishers at least for large field races like zone 2 Tiny Races.
No reset after the seed score formula is changed is an odd decision.
Unusually, the December Zmonthly starts on 25th November and there’s no sign of any changes ahead of it starting.
And everyone knew exactly how and why Zwift had ballsed up the seed score. I wonder what exactly they’re testing now…..
Could have just used a 50/50 conpound of zmap in w/kg and watts and it would have been good enough I’m sure!
On tanking, i think this article and Tim Hansons quote greatly misjudge the incidence, importance and impact under ZRS.
I’m sure what Tim says is correct…..in the context of a CE world and a small ZRapp community.
However, just as Zwift totally missed the importance of building their algorithm off mass start races only, tanking is clearly a bigger issue when the pen ranges are large, fixed and used in the vast majority of races.
I’m yet to be convinced that the ZRapp solution is anywhere near up to what is needed, and tanking/low effort racing is happening at scale already. If there’s one thing we DO know about the zwift racing community its that lots of people will take actions to be at the top of a category rather than the bottom.
Useful summary, and glad to hear they realise it doesn’t work.
the main issue – the mist realised this – is that FTP determines whether you can stick with a group. Shorter power may determine you winning from the groups but it can’t replace FTP ie >20 min as the primary data point.
i did the 350-520 community race today. I pushed > 20% over ftp for 10 mins and was dropped. Winner was 4.3 wkg, that’s A cat in what used to be a C race (it’s called C in the title!! Same as above. Bottom 3, no reason to race once you’re in a min group. Seeded at 494 and should be about 340. I don’t want to race with As… or mid Bs to be honest. If they don’t fix it, I think it might be subscription over after 8 years for me. You can get the virtual workout for much cheaper elsewhere.
Zwift needs to fix this quickly, we have already lost 20% of our team to other platforms: rouvy, mywhoosh etc, I now race less as a consequence of the changes as no fun riding on your own, so my strategy (i’m an old high cat D, both in age and capability 😉 is to only race hard once every couple of weeks to keep me in the 0-180 range as this is competitive for me (if i can manage to stay in the 1st group i can finish in the top 10) it was a double whammy with the price rise and zrs which i fear will continue to have repercussions!
Mad system, they have to take FTP into account – 30 sec has no bearing on a race result if you can’t stick with the group for 5 mins. I’m a C and having to race As. No fun. Might be time to retire my subscription if they can’t get group races back to something where you actually race with people of a similar ability again. Sandbagging was far less of an issue than this.
I completely agree. That and racing 10 year olds that can pull 4.0 wkg.
As a B rider with a good sprint, I have been put in the A category where the group pace exceeds my FTP. Having a good sprint can be fun, but it does nothing if I get dropped within 10 minutes at “regular” pace. The system is seriously flawed and Zwift can expect dropping participation numbers.
I don’t mind participating and not winning, that is part of the game. I also like growing as a rider. But when participating in my only allowed category, I do neither. I have no chance to stay with the group, leading to me not being interested in racing. Which is a shame because racing is what keeps me on the platform. Just virtual trainings can be found on other, way cheaper, platforms.
If the system would work, then you would be moved to B after two races…
Thanks for feedback Eric. In my view a good ranking should achieve two things, and two things only : 1/ When I compete with someone with a similar ranking, the race should be balanced : maybe I will beat him on hilly routes, but he should beat me on the flat, or vice-versa: I should not beat him always, and 2/ When I have a great race, beating some people I usually don’t, with a good tactic and/or bigger watt than usual my ZRS should go UP; and when I have a bad race because I took some wrong decisions or did not push enough, my ZRS should go DOWN.
In my experience, current ZRS does not (yet?) comply with those two requirements, contrary to vELO score in ZRacingApp. Going up or down is depending mainly (if not only) on your finishing rank in absolut term (first 2/3 gain points, last 1/3 loose points) which is stupid : if my rank was the lowest in the penn and I manage to finish before the last, my score should go up, not down. And we all have examples of riders with lower score that are beating us consistently because they are tweeking the system.
I hope ZRS will get there, but I agree with other comments : why does it takes so long, when a very good system already exists (ZRApp) ?
So much this!
When ZRS was brought in I was a top end cat D (pace group), but heavy with high raw watts. It was entirely reasonable that I ended up in the new cat D 180-350 points. 1st race in that I was pretty much lowest ZRS of all the entrants (mostly old cat C and a few Bs). Bust my hump trying to stick with riders with FTPs 1w/kg higher than mine to finish lower mid pack but above where ZRS would have estimated me and … I lost points. But given where I started I was never going to lose enough points to end up in a new category.
If score was calculated against expected result for the event then a) it motivates racing even if you’re at the back of a pen allocation and b) it conceivably removes the need for pen allocation anyway which removes the upper/lower end of pen boundary issue which also removes the incentive to tank scores. It does mean that those that were at the top of the lower categories don’t get virtual trophies in their virtual cabinets though.
Does anyone know if there is anything built in for super small fields? I raced in a 350-520 race last night (Fresh outta 24) with only 4 in the pen. One missed the start, so we started with 3.
Races like this shouldn’t really carry the same weight as a race with a normal sized field but my score went up significantly (15 points) winning this one. I don’t want to start sandbagging but winning a race with 3 people shouldn’t make a big change in racing score.
I’m sorry but it’s not working, it’s reason for being created is to make racing more evenly matched in a pen and therefore more fun if you’ve ridden more than a few races ( I did 35 in the last 7 weeks) you have experienced huge disparity in the ability of riders in the pen
Possibly not in top tier races – being amongst the lesser riders I can’t say.
Sitting on the starting line knowing you’re in for a beating is not fun
Battling to keep up and cresting a climb with the leading group or having a go in a bunch sprint and finishing twentieth is fun riding in alone minutes behind is not.
Patently very good riders popping up in lower level races is fun for them and them alone
ZRS as it stands isn’t working
If I tell you your riding against a riders whose scores are 289 250 340 and 450 does it give you any clue which are the strong riders?
If I tell you they’re 2 cat B a cat C And a cat A could you guess which was which? As a standalone number ZRS really doesn’t tell us anything – except high numbers for high end riders
Thanks for the update. As many others have stated, ZRS is a debacle. Now in light of this article, I have to wonder if the zwift camp baseline results will be used in the adjustment of ZRS….sprinting, attacking, endurance.. seem like key race metrics..
Eric, is it possible to increase the html banners on the sides of your posts? The pop up ads are blocking the post content on a standard laptop screen and the close out X isn’t always obvious. It’s quite annoying (while also acknowledging you likely need the ads)
The groups have to be random. Best 25% in points are group A
next 25% are group B
so with a racing score of 500 you can be put in different groups depending on how much stronger and weeker riders enter
Anyone else noticed that starting at a higher category and finish somewhere at he end of the top half getting you 0 points?
are there mass starts possible in the system? Maybe that will change the scores quicker
I’m wondering what the anti-tanking measures look like. Would I get caught up in that if I do some of the Tour de Watopia stages full gas, but others at a Z2 pace (because that’s what my schedule called for that day)??
I think Zwift Plays braking finally found it’s purpose to help people come in last. You can ride Z4-Z5 all race but if you accidentally tap that brakes a few time, you are coming in last.
Race Score is stupid broken. It is heavily bias to light riders.
Look at the stats: 2 riders with the same W/Kg can have huge difference in ZRS score.
Rider A: 176 watts at 57.7kg, 3.05 w/kg = Race Score 22
Rider B: 265 watts at 88.2KG 3.05 w/kg = 132 race score
If we increase Riders A power to 187 watts 3.24 w/kg he still only have a race score of 67.
Here is the kicker. This rider would actually pull away from rider B on a long climb. But he won’t have an issue staying on the wheel of rider B’s wheel on the flat. So essentially rider B doubles rider A scores, have no ability to pull away but in certain spots rider A can pull away from rider B.
Slow down my man! First of all, where do you get those race scores?
My FTP is about 235 watts at 61kg. A respectable 3.85w/kg. My race score is 523.
Secondly, as a light rider I can tell you that climbing is not all that matters.
Maybe the scores you mention should not be that far apart, but a heavier rider with the same w/kg is obviously better than a lighter one. At least on Zwift since we’re not racing 5h, 7000kcal rides.
I’m pretty good at climbing, but trust me when I say I’m often pushing FTP just to stay in a group on the flat and most of the times I get dropped are downhills. I reach the top with others of the same w/kg, but since they are 20kg heavier I somehow need to keep pushing downhill in an attempt to follow.
Is this system perfect? Probably not. But the previous one was also not great. It leveled out the field where a light rider has an advantage, uphill, because the cats are based on w/kg. Yet in literally every other circumstance the lighter rider has a disadvantage.
Part 2 of how broken ZRS is:
If you ride around on erg mode that caps your sprint power. You can win nearly every race.
Check out these ZRS scores if you don’t have a sprint with a flat power.
A rider with 65Kg, at 245 watts 3.784 w/kg would have a ZRS score of 225.
B rider with 65Kg, at 259 watts 3.984 w/kg would have a ZRS score of 268.
C rider with 65Kg, at 270 watts 4.153 w/kg would have a ZRS score of 303.
If you turn on erg mode and limit your power or just turn trainer difficulty to 0 and keep a constant cadence. You can ride around like a boss in D pen.
Just know if you have the slightest increase in 30s power, it will have a huge impact to your ZRS score. I.E adding a 1 w/kg on 30s power over your 10 min power, will likely add 50% score gain.
Here is an example, a rider @ 65kg with a 8.27 w/kg 538 watts 30s power, plus a 10 min power of 3w/kg 195 watts has a ZRS score of 315.
If we compare this rider to the rider that doesn’t have a sprint. The rider C at 4.53 w/kg have a lower score than this rider. ZRS 303 vs 315.
I see quite some potential in the Dynamic Pens idea, I really hope they will look at the development of this. I believe it is how the system is supposed to work in the end as it would solve many of these issues.
Also for initial seeding, is it not possible to let newcomer/returning races just race an x amount of races but make them invisible and undraftable to the other racers and not include them in the public results (but for their own benefit show where they would have finished). I think it would lower the bar to start racing because to them it would be just like participating in the race, and it get’s you race-quality seeds to start with.
When combined with the Dynamic pens idea this would work even better as the system can “try-out” the first few seeds.
Just a few ideas I had… 🙂
Great ideas!
Feels like something changed indeed. I was grinding my way to cat B, and truth be told I was having a lot of fun in C.
Today I kind off easily won a C race and suddenly gained 20 points. Into B I go!
I do a B race, got dropped pretty quickly and settle in with the C’s. 5th/8 B riders. BAM, lost 35 points. A bit excessive if you ask me.
I’m planning on trying B for a while though, to keep things fair 🙂
I think one thing missing is locking you into a category upgrade for some amount of time (i.e. you can’t drop back down a category the very next race or three). Whether that’s a week, a month, a couple months, no idea.
anywhere where you can see development in your racing score?
got it:
I’ve given up on any races using the racing score – as a mid C grade rider I’m sick of 20% of the field being B (and even the odd A) riders. I’m happy to smash myself riding against other C riders but when you see 6 or 7 riders in a field of 30 who are B it kills the enjoyment.
Race score has ruined racing for me.
i would race in C and would come inside the top half, and race with people for the entire race.
now I race, I’m in the 350-520 category and I’m dropped immediately. And I mean immediately.
There are normally 3 or 4 of us at the back that often never see another racer for the rest of the race. This is despite me racing above threshold for the whole race.
i will no longer do races, and it’s just shit.
racing was fun on zwift. It is no longer
Same. I have to look for obscure races with like 5 people in them.
I think I’m an outlier here, but I’d prefer to race all together with more riders. Too many categories and you just end up racing a handful of people. Pretty boring.
Hi ZI – any idea when the ZRS system will be finalized?
It is starting to get a bit demoralizing… A recent race I have been given an underlined down arrow score of 530, indicting I have reached my ceiling floor based on my recent 90 day power numbers. I have an ok sprint but my 20 min w/kg is 3.2 at a push more like 3.1.
With a score of 530, this often has me racing against people who can put of in excess of 4.5 w/kg for 20 mins. This is simply not fun anymore to just get dropped within ten minutes in most races or realistic for me to improve my fitness by a considerable amount to be able to compete in the bunch..
I am hopeful the new system will improve but currently I am left feeling quite dejected.
Thanks again ZI in CC and.
Regards
Christian
Changed today. I was moved up 20 points, came in 77th and lost to someone who more than doubled my 20 min power. Don’t worry. It’s still just as demoralizing.
I actually mispoke. I was dropped down maybe 90 points. The C category just got even more competitive. If you’re a 3.2 there are even more B’s in your category now.
Did something change? I raced your mini races series on Saturday (thanks) in the 400-540 range. I was 508 racer score to begin. I won one in a sprint, 8th in another and blew up in the other two on the hills to finish 30+. At the end it said I had increased my RS to 516, then down to 512 after last race. I checked today, I’m 592! Haven’t been on Zwift since. What’s going on?
Probably this: https://forums.zwift.com/t/zwift-racing-score-seed-v2-release-december-2024/640588
Ahh, thanks.
woah.. my racing score bumped up from 348 to 550.. i get that zwift changed something, but that was quite a jump?
My racing score went from 248-527 overnight, and I’ve only done 5 events, none of which was a race. I’m a 57 yo female with two replaced knees, and I started riding in July. If I have to wait 90 days to get out of category B, I won’t be able to race all winter. I was in a D level group ride last night suffering and was mid pack. I bought an expensive trainer to be able to use Zwift. I’m not sure I’m going to keep my subscription now. Does Zwift have any customer support?
Betsy, can you email me your info: [email protected]
What are the best racing series or weekly races that use ZRS? I primarily race in WTRL and FRR where ZRS is not used. For trying to make ranking simpler and more competitive, I feel everything has just become more convoluted. Thanks!
Zwift’s ZRacing is probably the most popular: https://zwiftinsider.com/zracing
Tiny Races are use ZRS and are quite popular: https://zwiftinsider.com/tiny/
As for whether things are more convoluted now or not… that depends on how far you want to dig into things. It’s quite simple to race on Zwift in ZRS-ranked events. Just click and join. Now, if you want to understand where you score comes from… that’s more convoluted…