Well obviously if it is worse for 50%, then it is likely to be better for the other 50%.
The point is, it hasn’t been improved overall.
I do think the lowest category is worse though. People used to say that sandbaggers, who should be cat C were ruining cat D races. Now there are loads of (old) cat C riders in the bottom category and genuine old cat D riders are being forced out of the lowest cat races while cat C riders remain. It makes no sense.
It might be better if the cats were overlapping, or ranking for a race was changed depending on weight and terrain. At the moment, heavier riders are generally out the back as soon as it goes up hill because there are riders with 20% better wkg with the same ranking. It doesn’t work.
I’d be happy racing on Tempus Fugit in the higher cat, but not on many other courses.
Anyway, last Zwift group ride tonight, then I’ll unsubscribe for a while.
Maybe things will be sorted in 6 months.
Unfortunately I don’t think anything will fix most people falling off the back unless seed = floor as was already suggested. Probably need to force people to actually make representative efforts as well.
Categories are sized such that one can almost always tank to 80% of their seed score to move down a category. IE a 560 seed rider should do well in 450-570, but they could also reach a floor of 448 and briefly race down a category. The higher you go, the smaller the category relative to the overall score, so the more people eligible for such a scheme. The entirety of Cat A seeds, 570-700, could drop down to B, for example, all the way up to 710 actually.
That problem is less pronounced in E/D but instead you have people who just haven’t raced in a while or never made a real 5 minute effort. Like, right now I am sure I could push my seed into Range 1 C (270+) if I really wanted to, but I could also tank one race to get into Range 2 E and have a field day. That shouldn’t be an option, in my mind.
I’ve given this some thought, and I don’t think setting the floor equal to the seed (instead of 85% of the seed) really fixes much. It will help the unfortunate riders who enter the ZRS system with an accurate seed score, but the core imbalance between riders with “true” max efforts in the last 90 days and those without would still exist. The whole reason we want a results based rating system (which ZRS is supposed to be) is that it is essentially impossible to get equivalent maximum efforts from all riders.
Case in point: myself. I’ve been Zwifting for a couple months, and prior to this morning, I had never really done an all-out effort that was done fresh and paced to exhaust me at 5 minutes. You don’t do that in a race, and I didn’t even do that in a VO2 Max workout. It only happened for me because I wanted to do a VO2 Max estimation. The difference between this effort and my best 5 minute output in a race was 10%!
Prior to this optimized test, my best 5 minute output was from the finish up Box Hill in stage 4 of the Zwift Games. For that effort, I was fatigued from the first 40+min of the race, extra fatigued from the beginning of the climb (which is longer than 5 minutes), and not paced evenly because I was racing another rider. This was definitely a true effort (no sandbagging), but it wasn’t representative of my 5 minute max while fresh. I think you could do a thousand races and never get into a situation where you’re called to go all out for very nearly 5 minutes while fresh. Maybe riders who do short crits, but not riders who just do zRacing events.
So, after that all out effort, my ZRS seeding score went up by about 50 points. That means the version of me who does a dedicated effort gets seeded a full half-category higher than the version who only races (not sandbagging!).
My take-away is that trying to fix ZRS by tightening the floor is a fool’s errand. If the system is relying on enforcing floor values, then it’s really a power based system and not a results based system, and I think we already know that power based classification isn’t very good. What really needs to change about ZRS is the adaptiveness to results, especially for new riders in low categories.
In my fifth ever race on Zwift, I placed third in a field of almost a hundred, and my ZRS went up four whole points! And my ZRS was still over 20 points below the ceiling for the category I raced in. If that isn’t a bug, they need to consult with people who know how to build a rating system that can accommodate new players who gain skill/fitness quickly.
If nothing else, setting floor to 100% seed score lowers the top end standard of the easiest pen for most races, which is good.
For a lot of Racing Score so far, the five pen system weakest pen has been higher at the top end than the weakest pen for Category Enforcement.
Let’s aim lower, how about just weekly updates what they done and what they hope to achieve next week?
I see months go by and they are not doing anything. No one like DOGE’s Elon, but they seriously need to implement the job report documenting what they do or resign. From an user perspective they are not doing anything,
Setting the floor at 100% only helps if the users has a remotely accurate seed score. I think the lower pens are full of new riders who get into racing without an all-out 5 minute power test.
It would probably be helpful to have a required “Race Pass” workout that directs riders to do an all-out effort of 5 minutes (or whatever duration works best for setting seed scores).
Setting the floor to 100% makes racing fairer as the difference in abilities between the top seed and bottom seed in a pen narrows instead of making it greater which the 85% floor currently does, you can have a difference of 1.5 w/kg between riders in 180-350 which is greater than the old CE system.
If a riders seed score is not accurate it will only be because it under-represented which makes setting the floor to 100% even more important.
For heavier riders Zwift racing is tough.
I averaged 330 on Dolphin crit course yesterday and got my ar*e handed to me by 55kg guys averaging 250.
No way that happens in a flat real life cobbled crit with one short steep climb.
In zwift racing a short 60kg climber gets an amazing draft and CDA on the flat, then gets rewarded for lightness on the climbs.
Heavier riders seem to get less of a draft on the flat, then need insane wattage to stay with the lightweights on the hills.
IMO a crucial part of the problem is the 120lb 5’1 people (males) are actually substantially bigger IRL - and have the bigger person’s power. If all the tiny male racing Zwifters were authentic we’d see realistic performance on the flats or rolling/punchy terrain from them because they’d have nowhere near as much strength.
I’m bored, and on a Friday night and everything, so here we go:
I don’t buy it. I don’t think that is what is really happening. Likely it’s a combo of serveral general factors. I’ll name two of them. And then have a quick look at this particular race.
First, there was a nice study on what makes RL climbers so good, good enough to win mountain stages. I can’t quote a reference for you, but it’s out there. Anyway, the conclusion was that it seems you do have to be small and light, but also that this was not enough. Because of the WKG equivalence (same WKG = same climb time), a bigger rider wasn’t disadvantaged enough, or at all in fact. To be successful, the climber type also had to be relatively explosive and have excellent recovery. In essence, you need to be small and light and be able to change tempo repeatedly to really screw the big guys over. Now, Zwift physics isn’t necessarily RL physics, we’ll come to that, but let’s keep this finding in mind nevertheless.
I think this type is kind of similar to the guys you met in this context. What you don’t see, however, and I mean it both literally on the Zwift screen and in terms of what you pay attention to when things aren’t going great, are the little guys behind you. The guys who couldn’t possibly toy with and torture you if they wanted to. Not too many of them this time, but don’t forget about them.
Second, I really don’t like the PD4.1 OR PD5. They both suck. So unrealistic. Being light does seem to pay off better now uphill compared to PD3. You can really tell the weight differences between riders going from flat into an ascent. But light riders, or practically everyone, who want to avoid getting dropped from the back of (small) groups have to struggle now. If you are badly positioned - and sometimes you have no choice because all the “good spots” are “taken” - it’s like draft is only partial or intermittent. So for a light rider, the race gets unnecessarily punchy on a regular flat these days, regardless of what the front is doing. Riding in a group is harder work than it used to be. And don’t even mention riders with and without steering, pushing people around into bad spots or locking them in. Anyway, the point is I think there are some things about the new racing that isn’t about ZRS at all but about other things, such as PD changes.
So, into the race then. Sorry for snooping, but I had a look at your race, the results and stats. Take a look at eg. the no 7 guy. He’s a representative contrast. He loses the final sprint by a second but is with the front until the end, way ahead of you. From a sort of all else equal perspective, not knowing any exact RL physiological details about any of the riders, a few things stand out. If we disregard the fact that he’s pulling +1 WKG on you (he’d have to since he is so light), and I think we should, then there are these interesting observations:
Look at your HR histogram. Man, you worked HARD. You might say you had to, but really it’s by choice. Respect!
Then look at #7’s histogram. Spot the difference? He is not working very hard at all. Oh, but here comes the proverbial protest. Let’s get it over with straight away: “But… but… HR spans vary wiiiiildly… the histograms they… they are never comparable!” Yeah, whatever. Show me the data! Or just take a look at some other races this guy was in, eg. Zwift Games on the 21st. This guy can clearly work harder (unless there are bad heart days beside the bad leg days). So, if his heart had wiggle room when racing your race, which he did, then being that ideal climber type guy, who can be both small and punchy and recover well, isn’t really all that hard, is it?
Now take a look at your HR span in this race. Compare your average to your max. Compare them knowing you raced on the limit. You didn’t hold back like some sissy/cruiser. But was there wiggle room to handle tempo changes?
Then take a look at the no 7 guy again. Compare his average HR to his max. It’s a much wider span. When he had to, he could step up the pace or respond to a tempo change. When he didn’t have to, he could ease back and recover.
By comparison, that guy really rode this race in easy mode. And it wasn’t primarily draft that made it easy for him. No wonder he could establish himself firmly in the ar*e handing business.
This is what it all looks like from the back seat at least.
If you are on some of the mountain routes in Zwift you might encounter some of the real top climbing riders - from the big real world events - the 7 day events with huge mountains day after day. Then you know what fitness really is. 
Those events are a lot harder than a 45km Zwift “queen” stage.
This is I think what folks called “making racing more exciting”, isn’t it? Having steering or not doesn’t make much difference these days, you still get shoved out to the side no matter what.
The factory tour power units using w/kg is bias towards light riders too.