Big news for the Zwift racing community this week: Zwift Racing Score has entered the building! How does it work, and what does it mean for Zwift racing? Let’s dive in!
First, a Bit of History
It’s long been known that Zwift’s power-based categorization scheme, while easy to implement, is far from perfect. Grouping riders based on their power numbers will always enable sandbagging on some level since riders with numbers near the top of their categories can keep on winning without being bumped into a higher category.
We all knew that a results-based categorization system was needed. But how exactly should that work? While ranking models already existed in both the video game and bike racing worlds, there was no “off the shelf” formula Zwift could grab that worked quite how they’d like.
One year ago yesterday, Zwift announced the public launch of their Racing Score metric. But the racing community wasn’t impressed with the first iteration, since it was based on ZwiftPower’s ranking metric, which many racers see as flawed.
So Zwift went back to the drawing board, bringing community Zwifter and race ranking guru Tim Hanson of ZwiftRacing.app in as a consultant. The new Zwift Racing Score is nothing like the original, and that’s good news! Zwift heard from the community and made big changes. Next week, Zwifters will begin experimenting with the new Zwift Racing Score. Exciting times!
[adinserter block=”4″]
How It Works
Zwift Racing Score is on a scale of 0-1000 (higher is better), and categories are broken up by assigning Racing Score ranges to each category. For example, July’s 5-category ZRacing events use the following score ranges to populate categories:
- 725-1000
- 575-725
- 400-575
- 225-400
- 0-225
Zwift isn’t revealing the precise formula for computing each rider’s Zwift Racing Score. But here’s what we do know: everyone begins with a “seed” (initial) score computed based on your 30-second power, 10-minute power, and body weight. Zwift says:
“We calculate a new racer’s predicted score (seed score) using a compound formula based on CP30s and CP600s power outputs. We then normalize the power by weight and make further adjustments based on our analysis to ensure the score is fair across various weights and performance levels. After normalization, we apply a specific formula to transform these values into a final seed score between 0-1000.”
The primary way your score will increase or decrease is based on your finishing position in races and the quality of fields you race against. Again, Zwift isn’t sharing specifics. But the way they describe it working is exactly what we would expect:
“A higher finish against stronger competitors in races will increase your score more significantly. Conversely, lower placements or less strategic engagement in races will result in smaller gains or even reductions in your score. This method ensures that each racer’s score truly reflects their performance and growth in capability over time.”
There’s some level of “decay” built into the system, so your score may decrease if you don’t race. That said, your Racing Score won’t fall below your most recent calculated seed score.
To learn more, see Zwift’s support page on Racing Score >
Racing Score UI
Zwift Racing Score will show up on both the Zwift Companion app and zwift.com website in three different ways:
- Viewing your current Racing Score
- Signing up for scored races
- Viewing scored race results
IMPORTANT: you must be a member of the Zwift Labs club in order to see your current score. Otherwise, you will only be able to see your score in scored race results. Join the Zwift Labs club >
Here are some screenshots showing how Zwift’s Racing Score will be displayed in Companion and online:
Viewing Current Score
Race Signup
Race Results
Upcoming Events
Zwift is setting up particular events from Zwift-owned racing series and some community-organized races to begin testing Zwift Racing Score. I call these “scored races” below, which means two things: the races use Zwift Racing Score for categorization, and the race results will affect your Racing Score.
- Zwift zRacing Series (starting July 1): 4 time slots per day with be scored races: 0810, 1110, 1710, and 2210 UTC. These time slots will use 5 categories, except the 1710 time slot which will use 10 categories. The 10-category events are actually two separate events – a “low end” and a “high end” – each with 5 categories.
- Zwift Crit Racing Club (starting July 1): 3 time slots per day will be scored races: 0940, 1740, and 2240 UTC. These time slots will use 5 categories.
- Zwift Hill Climb Racing (starting July 5): 1 time slot for each week’s race will be scored. This time slot will use 5 categories.
- Zwift Women’s Racing Series (starting July 2): all events for this series will be scored, and will split riders into 5 categories.
- HERD Beginner Racing (starting July 6): all events for this series will be scored, and will split riders into 5 categories.
- SZR Summer High-Low Series: all events for this series will be scored, and will split riders into 10 categories.
- Zwift Insider Tiny Races (starting July 6): all events will be scored, and will split riders into 5 categories.
For now, all scored races will live under the Zwift Labs club, to make it clear that Zwift Racing Score is still in its testing phase. Join the Zwift Labs club to easily access these events in Zwift Companion and share feedback via club chat.
See upcoming Zwift Labs scored races at zwift.com/events/tag/zwiftlabs
Feedback Requested
Zwift says, “Through Zwift Labs, we’re activating Racing Score on a select set of Zwift-owned and community-led races. Our goal is to engage with a diverse group of seasoned racers and thoroughly test various racing scenarios during this initial phase of implementation. But it’s about much more than just racing, it’s crucial that we hear about all your individual experiences, as this will shape future development.”
If you participate in a scored race, be sure to share feedback in the Zwift Labs club chat afterward!
Concluding Thoughts
While it was a long time coming, I’m stoked that Zwift is finally rolling out Racing Score. And I’m crossing my fingers, hoping that everyone’s experience with scored races is a good one.
Because if Racing Score works as intended, it’s going to be the biggest change to Zwift Racing since category enforcement rolled out in February 2022. Results-based categorization means the end of sandbagging, which is huge. It also opens up the possibility of custom categories, meaning race organizers can modify the number of categories and their breakpoints, mixing up the race packs so riders aren’t always near the top or bottom of a category.
My guess is Zwift plans to test Racing Score for July and August, so it’s ready to go when Zwift season begins ramping up in September. I can’t wait to hop in and give it a try myself!
Questions or Comments?
Got comments or questions about Zwift Racing Score? Share below, or in the Zwift Labs club chat!
This is encouraging! I was promoted to Category C from D and I’ve been dropped due to be being at the lower watt level. I’m hoping that this instills parity and I feel like I did in Category D which was actually racing rather than ride with a group for 5 to 8 minutes and then get dropped and be too wiped out to ride with the chase group.
Cool!
I don’t see my racing score in CA yet though I can see the test events and which categories I can (and can’t!) enter in them.
Any word on when racing score will appear in CA and whether there’s somewhere else we can see it for a group of people? Would provide very quick feedback if we could see all in, say, our club/team who we usually race with regularly and know how we stack up against.
It’s there for me. I think you might have to join Zwift Labs, the link is in the article. Ride on
The score should get initialized when you go to your profile in CA after joining the Zwift Labs group. Once initialized, you should see it reflected in events than span a range that includes your score.
While I am a member of the Zwift Labs Club, I cannot see any scores in the Companion app and zwift.com profile
I can see my score in the Companion app, and on Zwiftpower, though not in my zwift.com profile.
Correction – it’s also in my zwift.com profile (as shown in Eric’s example image – duh)
It also shows in game on the homescreen, top-right!
Just saw that in the game homescreen, too. Wasn’t part of the club so I didn’t knew what that number meant, but guess I found out now 😄
You might have to update Companion? They released a new version recently. Racing score should be right below pace group.
What I’m still missing is age groups, now I’m racing as a 69 year old D against prime time youngsters who pick their races to stay in D. They can and will do the same to stay under 225
“What I’m still missing is age groups” – Good, I’m 72 and I’m not interested in any “participation awards” because my age. That garbage has ruined amateur racing.
We don’t need that here or male Baby Boomers putting their age in their user ID either
I’m going to keep my ID as 72, old, fat, and slow. I’m coming back from a serious IRL bike accident and confident I’ll just be old and fat soon enough.
Just saw my racing score and I’m a two … out of 1,000. If this were golf I’d be Tiger Woods … but it’s not.
You sound a bit annoyed Steve, get rid of the garabage and enjoy zwifting!
Cheerz and ride on!
It’s not about age, it’s about excuses
If you were younger it would have been something else
I zip past 20 year olds on Zwift every time I’m on it.
I do/zip the same Steve but I would like to see results like they are in IRL and hope everyone is honest about gender, age and weight!
I’m honest so I don’t need any excuse
Awesome! Excited about this. Been loving zwift racing.app, so curious to see where zwift takes this
Does this have any impact on group rides? I usually attend B-bot rides with Jacques and try to stay in the middle of the group. Hopefully, this is not considered sand-bagging?
No impact on group rides. Zwift Racing Score only applies to scored races.
What about TTs
Half way up Cat C – perfect 👌
Do we know whether this is going to be used for Time Trial events as well ?
I believe this could be used for iTT events.
It’s being used for iTT in Women’s Racing series this Tuesday.
Music to my ears 🎶
This is great news and congratulations to Zwift for teaming up with Tim Hanson. Tim has always been extremely responsive to my questions about Zwiftracing.app in the past and showing great concern about making a fairer system that incorporates the IMPORTANT component of race finish performance.
My biggest recommendation is to come up with a system, from the start, that clearly states a policy for penalizing riders who bail out on races to avoid finishing in a bad position with the intent to avoid the loss of points. I KNOW that some riders are using this loophole to avoid losing zwiftracing.app points.
For sure, someone could have a “mechanical”, but not over and over again. My suggestion is to get a free pass the first time. For the second “quit”, a partial penalty, and for future bails, increasing penalties up to the point to the loss of points that would have resulted from a last place finish.
Riders will be frustrated if this is not addressed.
Quick question…. Will you be able to rate riders for point races, or is this only going to be for scratch races?
Great news and looking forward to seeing your plan go in motion soon.
The counterpoint to a partial penalty is that sandbaggers could abuse that to get knocked back down to a lower level. Its a potential loophole that would need to be considered.
True and good point. But at least riders will not be able to go below their “base” score. Plus, if there are sandbaggers who are racing to win against lesser competition, they probably will still get points for winning, which will move them back up.
What I have been seeing is that some riders do not want to take the hit in points that get them moved down below other riders they know. I definitely see your point though.
It is a balance and I hope that some type of compromise is considered.
Hey Brent !
Can you explain how that is a problem ? I honestly do not see an issue with ppl racing higher as opposed to losing races on purpose for example all week between ZRL races to stay in a lower cat.
The point of the scoring is to place riders into the group of racers you should be racing. If someone wants to sandbag that way make them earn the loss of points by making them ride the entire event to earn the points loss. A DNF is as good as a DNS in my book, they didn’t earn a placing.
The other counterpoint is it increases the scores of all the riders who finished the race against the competition they raced against to the finish.
ZwiftRacing App has a feature where if your power numbers for a race are clearly below what you’re capable of, it won’t lower your score too much. Hopefully they’re keeping that in mind as an option.
I would see myself as someone familiar with racing on Zwift but not so much good tactical rider yet. For me Zwift in general is a great place to ride a bike just like that, without thinking mich about route, traffic and other stuff. Racing is another level of engaging, making the whole experience even better. But it’s still just fun. I understand that people want more fair rating system to be actually competitive (in a good sense). But going further and penalize some folks because they quitted for really unknown reason is a bit too far for me. At the end it’s not elite racing with where each rider is somehow controlled.
Why penalize someone’s score here. It doesn’t make much sense although I know Tim listens to everyone. That’s easiest way to sandbag a points system without much effort.
It’s really frustrating to race a whole race against someone only to have them quit shortly before the finish so they don’t lose vELO points on zwiftracing.app. Especially if I’ve gone all out to beat them because I know they’re ranked higher than me and suddenly they quit so I can’t beat them and all that effort (other than a great workout) becomes pointless.
I’ve seen it happen where I was in the middle of the field of a race only to be the last finisher after the race because everyone else after me quit. And I’m a mid C – it’s not like I’m a top end racer going for high points.
There needs to be some sort of way of discouraging people from doing that. Maybe make them wear a cone of shame in their next few races that displays how many times out of the last 5 races they DNF’d. I don’t care if it’s a point penalty (which, as you say, could help sandbaggers), but there needs to be something.
EXACTLY!!!!!!! I’m glad I am not the only one who feels this way and sees this. Thanks Paul.
If they completed 50% of the race distance you will get credit for beating them in vELO. But they will not take a loss.
Penalize was maybe the wrong word. I feel that riders who fairly consistently bail on races when they aren’t going to finish well (to avoid losing any race points) and DNF should be allotted the points (negative points) as if they finished below the last place finisher. Maybe a rider gets a mulligan every quarter in the event that the DNF was legit.
At the very least, the DNF riders should count as finishing in last place for the rest of us who raced against them. Even if they don’t lose points (to prevent them from being able to sandbag – even though the point floor of their seed rating sort of does that), those of us who beat them should be able to benefit from our effort. Not the same feeling as actually crossing the line in front of them, but that would give us something.
Maybe this system has this baked in, but it’s the sort of thing that they need to tell us about for it to be effective (both to prevent the DNFs and to make the rest of us feel better about them).
I can see your point. But I guess this approach is an improvement because keeping your race score up will mean you continue to race against riders at your level or higher if you’re dropping out. There is no incentive to sandbag I think..
Will races be able to change the categories, or better yet calculate categories dynamically? Otherwise old fogies like me who are unlikely to be able to significantly improve our fitness, will remain stuck in mid-cat no man’s land. Not good enough to win or even podium, but too strong to be relegated. I am currently a true high D – I win D races regularly, but even if I blow my brains out in a C race, I can’t generate C numbers. Now it looks like I’ll be dead middle cat D (321). Oh well. It’s just a video game.
Yes, race organizers can set up custom category windows.
I think Curtis makes a strong point. It would be huge if race organizers are given control over how they organize the range within and number of pens.
And looking above under “Race Sign Up” the categories are different from those listed under “How It Works”.
Would also be great if an autocat feature, (where you just sign up for a race and it automatically divides everyone into a pen a few minutes before the race) is considered for the future.
Regardless, this is a big step forward and looking forward to seeing how it works out! Fingers crossed!!
Autocategorization based on who is signed up (and allowing organizer to choose it by pen size or number of pens or pen point spread) would be really cool for one off events. Other events, I want to ride with a team that I know who to expect in advance, but, on a random Thursday night when I’ve got some free time and want to jump into something race, that sounds awesome.
Completely agree. Be nice for organizers to have the choice between both options.
Thoes that are stage racers(FRR,CT, etc) , climbers, and TT focus this seems to hurts them as most zwift races favor sprinters. I ignore the z monthly series mostly with much of races on zwift. One of thoes people that went B+ to C with the new system. That will mess up stage races As the As fall off that focus on standard zwift racing and stage racers in lower cats pass them. Making a mess with scoring. With stage races the final GC needs to be accounted for as well in this ranking system.
Any race that has more than 1 finisher at the front will go to the fastest sprinter of the sprint.
Not sure I understand what you’re getting at here. As a GC rider, of course you’ll be worse than a sprinter on a flat race. Is the concern that you will be under-ranked when you do a stage race? Or that sprinters will be over ranked?
Ultimately, this new scoring system isn’t about your overall score, and getting as high a score as possible, but about placing racers near other riders at their level. As long as the system is consistent, you should be riding GC races against similarly able riders, so the racing should remain solid.
My question is if it is going to take into account the type of race? Because tiny races are diferent from crit races, or ZRL, or stage races like the FRR, Chasing or HISP tours.
There are people pretty strong in long races (and several consecutive days) but not so good in 30-60 minutes races. They will probably have lower 30 s an 10 minutes power numbers.
That’s just life though isn’t it? You don’t get categorised by race type IRL either. If you’re an A grade road racer, you race A grade on the road. Maybe you get slapped about in a crit, but you destroy everyone in the iTT next week. You’re still A grade.(for example).
I do like how indieVelo rates you based on 4 different types of categories (climbing, sprinting, endurance, and time trialing), determines what ratio of each of those is present in each race, and rates (and places you) according to the custom blend for that race. Seems like that’s a bit more complicated than what Zwift is doing here, but this seems like a step in the right direction.
I do see the potential issue though if I only race the style of races I’m strongest at, I’ll get promoted based on that style and get absolutely stomped when I step away from that style.
I guess the only solution (for me) is to do a variety of races (even those that I don’t particularly enjoy) so that my score is based on a balanced profile.
Those that become specialized will have to deal with the consequences if they try a race that’s different from that specialization.
Did wonder what the number next to my Level was yesterday when on the Zwift App’s Homescreen- IE the actual App on PC rather than the companion app
Looks like it’s displayed alongside a little flag icon
I’ve been doing the ZRacing events for maybe 2 years in Cat D. It took me 3 months to get my first points and I’ve been slowly climbing the Zwiftpower charts since then. Even with a current percentile of 93 or 94 I’m still at least 8-12 minutes from 1st position (monthly series) and rarely in the first 200 riders. With a race result of 232 I will now be with the 225-400 group. Feels like starting down the ladder again without really reaching the top of the last one…
Bringing Tim in is probably the single best choice zwift have made in a number of years. I worked with him in the early stages of zr.app, and it has shown exactly what is possible. It’s been a frustrating few years trying to get zwift moving with this, but can’t look a gift horse in the mouth… it looks like this is finally heading the right direction.
I hope they are able to block the noise that will come from the vocal minority that will see their hobby (winning 5 races a week) ruined
I’m really excited that this is finally here. Like you, I wish it had come several years ago, but I’m just trying to be happy that we finally have it. Stoked to race again.
Why not use the ZwiftPower Racing Score?
You can read the comments at https://zwiftinsider.com/racing-score-launch/ (among other places).
Will these new scores and categories be reflected in Zwiftpower? Zwiftpower does a great job of implementing basic “integrity” checks on riders (HR, weight and height consistency, dual power …)
Yup, they’re visible now under riders’ profiles on ZwiftPower!
Wow, they finally made some changes to the Zwiftpower code after it was acquired by Zwift. Should we have a little celebration?
I’ve been looking at Race Results scores of close finishers (to myself). Some numbers seem reasonable compared to my own score but others look outrageous. How much data from old ZRacing races have they used to establish RR or are the current numbers just seed positions? Something seems off.. I would expect scoring would align with current results in the GC of the 4 Sea Breeze events, but I see a lot of variation in some areas (looking mainly at Cat D).
Current scores are just the seed scores – no race results factored in.
Will the scores be affected only by performance in races that use the system? Or will results in all races (i.e. including those that use the traditional A-D categories) count once the system is live?
Zwift says your score will only be affected when you race in scored races. That’s how it’s working for now, at least. It could certainly work differently in the future if Zwift decides to make it so…
Eric. There are a bunch of Zwift Lab races in the Companion App, but these aren’t mentioned in your article under the list of upcoming races that will affect your Zwift Score even though they are using the 0-1000 categorisation. Can you please confirm whether racing these events will affect Zwift Score?
I don’t see any events under Zwift Labs that aren’t also listed above. What are the names of the events you’re seeing?
Thanks Eric, it must be more poor understanding. There are events with the familiar ZRacing, Viva La France badge and racing cyclists motif, then there is an all orange motif with just Zwift Labs. The later also has Stage 1 RGV. This must be part of the ZRacing series with Zwift Score enabled and the others are the ones without ZS enabled? I thought they were two different events/series and was confused about the latter. But maybe it is just the different appearance that got me and these are the ones you list above under ZRacing… Anyway, will try it tonight and will find out one way or another.
I would like to test this in one, or more, WMZ hosted events. How do I explore this possibility?
Reach out to Zwift’s events team.
I’ve quickly looked through the zwift article, as well as this one, but can’t quite figure the scale? 1 is top score of the grade or 1000 is the top score? As in, if I improve my score goes up or down? Apols if I’ve missed an obvious clarification
1000 is top, 1 is bottom. If you improve, your score will go up. This is the opposite of the ZwiftPower race score where lower=better.
Thanks Ikuyo. All clear now 😀
Zwift bringing Tim on board was a decision they needed to make long ago. Tim listened to me about potential problems with his initial iteration with his zwift racing app scoring dealing with sandbaggers possibly. He listened to many other members of the community as well. I think his iterative problem solving will be a huge help to zwift and future zwift engineering endeavors along with ability to listen to the community I hope.
Unsurprisingly, I see a log scale which will even out the racer capabilities. If you can’t sprint, but you can dig deep and place higher than expected on a consistent basis you will move up. Sprinters like me who don’t race only flat routes will not win consistently and have higher variance, but if those sprinters choose to race only flat routes then they will move up till there is a balance in outcomes. The importance is this is a harder system to truly sandbag. It will take about 2-3 months for the results to fully bleed most of the variance and outliers out. If you ride about 10-15 races in that time, your score will reflect your fitness and your race ability to the types of races and racers you are racing in.
So even if I came dead last in every single race (because I’m put in “B” for Women’s series), I can’t go down a category due to seed value?
It’s not exactly that simple… since category windows can vary from race to race. But yes, your seed value is effectively a floor you won’t drop below. Although it too can change over time…
I just looked at the Zwift Racing Score of every rider who outperformed me in the past three Zwift Tiny Races editions. They all have scores about 100 points lower than mine. I’m starting to wonder if it’s really working as it should. Time will tell…
Also, if the zRacing event categories are used for Zwift Tiny Races, I’d be in B instead of my current position in D. I’d get dropped after four minutes if the group maintains a B-pace pack speeds, not to mention any attacks.
I’ve been looking at similar data. Seed numbers that have been generated seem to be dependant as stated on 30 sec and 10 minute power values and not on prior wins. If you compare these power values with your close opponents you should see why you scored 100 points higher. Imperfect initial ranking I agree as I am in the opposite situation. Close opponents scored 100 points higher…
Certainly, I share the same opinion. I have reviewed their power data on Zwiftpower and Zwiftracing, and it appears that their numbers are comparable to mine, or slightly higher by 1 W/kg in sprints. However, despite me having a 100-point lead on Zwift Racing Score, Zwiftracing indicates that all of us have very similar points. It is evident that one application is performing better than the other.
Hi Eric….
One thought that comes to mind is the possibility of sandbagging through deliberate losses, I perceive this as about the only downside to this system (which I think is great BTW). Let’s say you do two or 3 races during the week and deliberately place poorly to reduce points, this then allows you to enter your choice races at a lower cat than you possibly should be. And even if you win and your rank rises, you can reduce it during the week again.
Perhaps some thought to defeat this scenario could be given some consideration. Other than that I have raced a series (Dirt) which classified by ZRapp rank and the racing was the closest I have experienced on Zwift, truly the way to go.
I think it will be much less likely that sandbaggers will purposely lose races to lower their ranking. The sole reason for sandbagging is to win. Looking at the guys who are winnign or podium the Sea Breeze in C cat, their win/podium rate is like 80-90% of their races.
Right now, they’re not riding to power potential and gaming the power based system to win races, this zwift score will improve on the current system
Being one of the mentioned riders, I feel I still had to put on a lot of effort to win those races in C. I would love to dominate each race and be upgraded to B, but in reality I was never able to breakaway in the prime time races. Now in B I am in the pack until the first climb and then its over. I never felt like sandbagging in C, just one of the strongest C riders.
Torsten, now you’re facing the As hiding out in B cat. Were you 80-90% win/podium? If someone in your local races had that win rate, they’d be bumped up categories, not sure why it shouldn’t be the same on Zwift.
I agree, that riders should be forced to use/show their full potential. Sadly most of the races are mainly flat and there is absolutely no benefit to be pushing at the front. Maybe the new categories will improve this situation, when a victory also increases the racing score. Hope they will not give up on it this time…
Hopefully it will pull out some of the A’s that have been hiding in B category. I have higher hopes for this than the current system which is much easier to manipulate.
I have found that the draft dynamics that have improved over the past few years have made the ‘flat’ races more unpredictable and have made it possible for riders to attack and stay away if the 2nd group doesn’t cooperate.
I see how that could theoretically happen, Troy… but like Peter, I just don’t think racers will be willing to invest the time to sandbag in this way. Way too much work for too little potential reward, I think.
Not really sure if I like the idea of shifting requirements. I do a lot of races with friends since we are all kinda in the same cat, but what if we decide to do a rtace next Saturday and the organizer suddenly changes the requirements and we do not get to be in the same group. That would suck massively. I am all for a new system but keep the boundries clear.
I would say that if you want to ride with friends, you can always do a meetup. But if some of your friends are out of category, then they should be racing in their appropriate category to make sure that the competition is fair for the entire race field.
I’m really looking forward to this as sandbaggers will have to choose between winning (which is the whole purpose of sandbagging) or losing purposely to keep their score low to stay in lower categories.
Basically now in C cat, and I’m assuming B and D have the same problem, the racers on the podium finish podium in >95% of their races.
Easy example here (name withheld).
Peter we’re just not good enough. But in all seriousness, this is the single thing on Zwift, which bothers me somehow, although I hadn’t any race, in which it would make a difference for me. I know it’s a game and people have always cheated in games but c’mon. If somebody wins constantly in lower categories then it’s obvious he doesn’t belong there. Zwift doesn’t offer complex strategies for races, which could be potentially a game changer. It’s all about watts and drafting.
Technical note: the email said the seed is calculated from CP30 and CP600. I assume that CP means critical power. So your seed is based on a mathematical model of your performance that requires some maximum inputs.
Normally, I think people say that to properly feed the critical power model (the 2-parameter Monod and Scherrer one, at least), you should give it a max effort of about 5 mins, and one of about 12 mins. It needn’t be 5 minutes exactly, or 30s exactly, or 10 mins exactly. It can be, but Zwift will do the math behind the scenes.
Anyway, my most recent presence on Zwift has been solely interval training. My critical power curve may have a 5-min maximum fresh effort – the efforts should be with minimal fatigue) from a 20-min FTP test I did. It won’t have an accurate longer-duration max effort. This is just something to be aware of, maybe.
Noting that the Tiny Races are using the same racing score brackets as Zwift uses for their Z-Racing series — was ZInsider required to use those score ranges, or you could have used any numeric ranges you wanted?
I was asked to use them for the time being, just to simplify event creation. In the future I’ll be able to customize them for sure.
I used to see you as a fan page that was free to do its own thing, but now it seems like there’s a lot of promotional content on your site. It almost feels like you’ve become a PR channel for Zwift.
I’ll wait to see how it shakes out over time, but it’s going to be rough for some people in the beginning. I’m in Cat D at the moment with a zFTP of 2.2wkg, I had a quick look at a race using this new system and the category I’d be in is a mixture of (presumably lightweight) B and C cats. Hilly courses are off limits going forward it looks like.
I feel the same way, moving two cats up, while my friends who are stronger than me stay in their current cats.Our categories used to be filled with 50-70+ riders each, but now we’re lucky to have 40. And it’s not due to the summer break. These ongoing changes have already upset a lot of people.
I would like there to be automatic grouping. Imagine signing up for a race without having to pick any category. Depending on the number of sign-ups the number of groups could be decided. That way you end up in a group of sufficient size, while also finding that you are quite evenly matched to your competitors. The matching improves with the number of sign-ups, because more groups could be created.
You would also find that you’re sometime near the top of your custom category and sometimes near the bottom, because the lines are drawn differently every time.
[Second try at posting. Hopefully I am not considered spam this time]
LOL. They clearly struggle with showing distances and time gaps between race groups, a feature that the Tour de France has had since around 1913. Or set the pack dynamics properly. Or provide a profile of the climb to keep you informed about your location and what lies ahead. Do not ask them for these advanced AI wizardry.
I saw my score go up already after a race – is the system already in full use, or did my seed score go up? (I made new 5m w/kg record for myself as well)
They’ve said that only races using the score for setting pens will count for results based changes, at least for now.
It would just be your seed score changing, those definitely aren’t set in stone. Mine just went up from doing a climb effort in a free ride.
My guess is that the sandbaggers use non popular races do downgrade their score and shine on the popular ones.
Bus still better as of right now 🙂
There’s overlap between the upper number of a lower category and the lower number of the category above. Is this intentional or accidental? I can see this leading to some confusion. If I have a score of 400, to which category do I belong?
The game will tell you which category you can join – it recommends one. My guess is the categories actually say something like <=400, and >400, but Zwift shows it this way to keep it simple. Ride Score has at least 2 decimal points after it, so chances are nobody is going to be right on a whole number anyway!
An interesting point: You can’t race up category. For instance, my racing score is 532 and I can only get into the 450-550 cat. I’m blocked out of any higher cat races.
And my usual gripe with the special Zwift races: The US West Coast (PDT) starts are at stupid (for me) times: 0110, 0410, 1010 (SIX hour gap, WTF?) and 1510. How ’bout three hour spacing all day, Zwift? I’d love a 0610 or 0710 start. I have to get up at 0330 for a 0410 start and it’s too late in the morning for me at 1010…
Just did one of the Zwift Labs Vive la France races and I can see one big flaw in the method for calculating seed values. By looking at only the CP30 and CP600 for seed values, sprinters/puncheurs get over-seeded while racers with high FTP but low sprint power get under-seeded. I’m a mid/upper C with better-than-average short power but not-so-great endurance power. But the ZRS seeding formula put me in a race with 80% upper B and a couple A racers (same race you were in @Eric Schlange ). Held on to the pack for as long as I could, but my “sitting in” pace isn’t as high as the leaders’ “pedal all day” pace. And this was a flat course! As soon as we get a race that has a climb I’m dead in the water. And because my ZRS apparently can never drop below my seed score I’ll be forever racing against 4+. I know that organizers will eventually be able to customize the boundaries, but many (most?) races will probably just be run with the defaults.
I think the solution here is to factor in 20-minute power as well. Otherwise every high-FTP racer that never sprints is going to be under-seeded and will be able to do an easy tempo ride that half the field won’t be able to hold onto. And while their ZRS may rise quickly – they’ll still be able to decay back to that artificially low seed value.
Not the same race, but I had a very similar experience. Agree that the high FTP people, who don’t have huge sprints, are under-seeded. If I’m completely fresh I can push out a very good 30s sprint, but no chance of getting near that after doing all I could just to hang on.
The problem with any of these models is that they assume maximum efforts have been made over the relevant intervals, then don’t do much to encourage those maximum efforts and effectively punish those who have them.
Still, hanging on to the lead pack in the 575-725 pen wasn’t enough to push me out of C grade in the old system. That system is clearly broken and at least they’re trying to fix it.
The old system is definitely broken and I applaud Zwift for finally stepping up to try and fix it. This is a step in the right direction, but hopefully they’re willing to tweak the things that have unintended consequences.
Ummm, I’m not so sure about that. I’m a 4.5w/kg bottom of Cat A rider with no measurable 15 second sprint (7.4) and mediocre 30 second (6.2ish) to 3 minute power. My 5 and 20 minute is what made me a ZP Cat A. The lack of 30 second power and lack of Cat A results gave me a Racing Score of 532, thus what I’m calling Cat 3. I raced the Vive Le France event this afternoon at 1510 Pacific with 45 other racers. Most of those racers were B’s and C’s with much less 20 minute than me but way more 15 sec to 3 minute. I was on the rivet for the entire event trying to reduce the number of sprinters but still ended up 10th. The upside was being in the mix until the end which is never the case in a Cat A race for me. Seemed like the faster guys from B&C had plenty of power to hang and then put on a decent sprint. I call it a win over the old categorization system with one race done. We’ll see…
So you are classified as a Cat A rider under the old system with a Racing Score of 532. I am a Cat D rider under the recent Insider Tiny Race classification with a Racing Score of 545. Oh man, please do not tell me we are going to meet in the same race. That would be crazy. My 20 minute power on Zwiftpower is 2.42 W/kg. How can I even compete with someone who has 4.5 W/kg? Who invented this nonsense?
Wow, I have no idea how you ended up with an RS545. You must have some kind of incredible 30 second power or something. Yes, you’ll be in races I’m in with Racing Score cats. Like I said in my post, even though I have A cat 20 minute, I couldn’t drop all of the B and C racers in my event yesterday and still ended up 10th. That said, the field started with 45 and we had 18 competing in the final sprint so obviously there were a number of riders without enough steady power to hang. Stay with it Michael and you’ll get the endurance speed you need to compete.
This is precisely what’s going on. The powerful FTP riders are dictating the speed within their Z3 zone, while those of us with strong sprinting abilities and 1-3 minute power but lower FTP levels are pushing the boundaries of Z4. Perhaps in shorter races, I can keep up with the group until the final sprint, but then I’ll be left behind and my 12 W/kg sprint won’t be of much use.
That happened to me in the Zwift Crit series yesterday. I only have a 2.98 FTP according to Zwift, but because I’m heavy that’s like 312 W. As a result, I was put in with a bunch of (mainly) Bs (final 20 had 3 C, 16B, and 1 A). I was on my limit the whole way and dropped like a rock after the final prime. Didn’t go in expecting to win, but to feel how hard it was to stay at the back of the pack and know that, against the same opponents, on any course with any bit of hills at all I’d be spat out immediately, was a bit discouraging. Still a solid workout though.
exactly! i’m in the same boat, low 20min (i’m old and lighter) so i get shafted as my sitting in pace is 2.2-2.4 and high C’s are away!
This is the screenshot of riders who enrolled in the C category so far (6 July 2024 race, Zone 1). Zwiftracing scores from 1719 to 870. Serious? 30 second power from 10 W/kg to 5 W/kg. Serious? Five minute power from 5 W/kg to 3 W/kg. Serious? Anyone is expecting a standard race?
The Zwiftinsider Tiny Race winners, B-cat Rene Olsen (29 June), C-cat John Larkin (8 June), and D-cat Danni Cojocaru (22 June) will all be competing in the same tiny race on 6 July. It’s not a joke, but it sure sounds like one, @EricSchlange.
Race scores says I finished 20/40 but Zwiftpower says 13/21. Will someone explain why the results are different please.
Some People get eliminated(DQ’d)? By Zwiftpower if they have no HR monitor
I thought that they both have the same requirements, so no HRM would be dq both places.
Just tried a scored race. It was awful. I’m low cat C but ended up in a group with Bs. Got dropped on the lead in and saw 1 other rider in 40 mins.
In no rush for this to come in and if It does then my race days are over . I don’t think that the calculation using short period time figures like 30 and 600 are always a fair reflection given most races are 40+ minutes.
The good news is – your score dropped as a result of this race. So next time, you may race against weaker riders. That’s the beauty of results based scoring! It should land you in the proper place.
Is this true as my score has never dropped after a single race, in fact it has went up from racing in chase races using Watts per kg where the higher Cats dont catch the big bunch of Cs and Ds.
How do I get a racing score. Haven’t raced in agars and now can only enter E cat. Which is of course empty
Ages
Everyone has a starting racing score. Read the post above for details!
I hope this initial seed value is updated regularly (monthly?). Otherwise, having that as a floor is going to suck if you get injured or sick or have some other loss of fitness but the system won’t let you drop below a certain value even if you keep losing all of your races. I get needing the floor to deter sandbaggers, but, if my seed value was based on where I was in January before health issues caused a 10% drop in my power values all across the spectrum (that I still haven’t gotten back to where I was) and I was still forced to race with that as the floor I couldn’t go lower than, I’d just go full time to one of the other apps that I use occasionally.
If you race in more than one race in a week is it supposed to take the fastest one for the series as in ZwiftPower I only see the first (slower race due to internet drop out mid race).
I did think the scoring group was about right for me as a Mid-B. Got dropped on the Aquaduct by a few but still was in it as in a regular B usually not even close.
Not sure if there is enough data to know anything yet?
I’m excited for this new system. I expect it’ll take time for people to get used to it and get slotted into the right/competitive categories, but it seems like it’s going to be better than the current system in the long run.
I am signed up for the zrace tomorrow with racing score. I’m in the 400-575 score group (my score is 498), and I noticed that one of the other racers has a score of 592. I’m surprised that is allowed, given that category enforcement is enabled… does anyone know what’s up with that?
Category Enforcement only checks for the category at the moment of signing up. So they will have been within the range when they signed up and subsequently had a score increase. The same thing can happen with the w/kg categories too.
being a high D usually (low C currently but as an OK climber, got upgraded on the Tour de France │ Cote des Forche │ Climb Portal Challenge ) I now find myself in the 225-400 zone where i will/get smoked every race! (currently 266). That’s the background now my thoughts are that in the races where there are high/low i will be in the 200-300 group which should be competitive. I am assuming things will level out as I have friends who are a similar standard, (we have raced together for years) and they are below 200! Also my strengths are that I can do a good 30sec dig and also my 5-10min is good too, but here’s the rub, my 20mins w/kg is a very average 2.5, so it seems i will get penalised in so many ways: age, performing ok in the key measurements and losing out to the C’s who can push 3w/kg. Being an optimist I hope things will pan out ok, if not I will have to race less and chose wisely! either that do more group rides (they are not races are they?😉) keep up the good work Eric! all the best John (spokey-blokes)
Mark Cavendish’s racing score is 591…
Maybe the craziest thing I have heard yet about the new racing score…if true!
It’s true. You can look him up on Zwift Companion. Verified Pro rider from the Isle of Man.
You’re right, lol! Just crazy!!!
I am totally not a fan of the new racing score categorization. It is awful and unfair. I’m forced to ride in a group where I’m close to the lowest end of the category with those whose racing scores are 100 points higher. I was in the C watt/kg category but now In the D group. Before, I could finish in the top 1/2 of the group. Now, I’m consistently one of the last finishers. The changes supposedly discourage sandbagging but the opposite is true since I feel the need to slack off so fall into a lower group and remain competitive. The encouraging and invigorating feeling of virtual cycling is quickly becoming a thing of the past for me unless something changes. Maybe add more categories or account for age.
Being nearly 100kg with a FTP of 2.4wkg and Cat D how long do I need to race 70kg Cat Bs with a 3.4wkgs FTP on a hilly route?