One can only hope.
Or races on another platform for 2 months.
It is better and the pen enforcement is definitely an improvement…but, decreasing cat by preforming a stronger 3-5 min interval is indicative of wackiness in the system.
Totally unrelated is a separate point that we want flagrant violators ,intentional or not, addressed.
If Zwift can identify them and give them the “missed calling message” then they can also be pulled from the race
Not just the results, but real time out of the race.
It’s getting better,glacially,
1 Like
This is definitely an oversight and only applies to those right on the limits of the category.
This can probably be fixed by taking ZFTP from 5min upwards and let MAP work on 5min and lower.
No one ca intentionally or unintentionally enter a lower category than that they are assigned.
With no data it is impossible to set the correct category so we have to live with the 1% that race once in the wrong category.
You can’t tell someone that they can enter X category and then pull them out of the race, if you have a set criteria then you have to obey that rule. Improving the sorting is the only way to achieve this.
Fixing anything after the race is just a bandaid. It didn’t improve the race.
1 Like
Well… you could actually. With the disclaimer that “we’re only pulling you on just this one race, and you’ll be in the right category next time.”
I don’t have a problem with a DQ for significantly blowing past the category limit, because at some point fairness tips in favor of the other riders, but it needs to be a significant breach, not just a great performance by a rider getting fitter, and we need to be really sure that they get an upgrade. A WKG based on ZP doesn’t meet those requirements because they might remain under the CE limits.
@Paul_Southworth DQ are like Street Cred, more you have the high level you are , i dont think that going to work.
That’s due to pen enforcement, not due to the CE categorisation system though.
1 Like
A system has been created that requires people to undertake maximal effort to place them in the ‘correct’ category, in an environment where the makers freely admit, there are very few users who undertake a maximal effort.
I have no doubt you will defend that logic, but for most other people that logic is fairly flawed.
1 Like
So how would you do it without looking at power?
It’s especially galling because MAP is key to winning in many (most?) Zwift races. If anything higher MAP should factor in upgrading rather than downgrading.
@Bath_Salts throw away your HRM and you can live full time in DQ city.
1 Like
Since discovering it to be the case, it’s baffled me for a while why zMAP is a prediction, rather than actual data for best 6min effort in the last 60 days.
Determining pen allocation by a 40min+ prediction (why has zFTP only ever been sold as a 40min+ prediction rather than a power number for a specific time duration?), then increasing the W/Kg threshold above the old Zwiftpower 20mins levels has created very unpleasant racing for those at the bottom of pens B to D inclusive.
Effectively wasting pen E for those without sufficient data in the past 60 days, rather than making these riders get data elsewhere (such as doing a ramp test) hasn’t helped matters. Splitting ability five ways would surely be a huge improvement on splitting ability between four pens?
Compound Score (best 5mins pure Watts x best 5mins W/Kg) has been found to be an excellent way of determining rider ability, https://www.zwiftracing.app/ has been using a slightly modified version successfully for a while now.
It’s a shame that it hasn’t become the default way to allocate riders to all five pens so far in Zwift, but it is one of several custom options available to race organisers, we just need them to use it and make riders aware they are using it!
@James_Zwift I did wonder if you would consider using Compound Score for a month trial in at least one of the monthly ZwiftHQ races you create, such as the future equivalent to May’s Crit Races?
1 Like
The flagrant violators I was referring to are typically in A cat like the guy who was holding 600 watts for the whole race while drinking coffee.
I thought it was this thread but it was another one.
I think it’s important to emphasize that blatantly incorrect setups can still Zwift - free ride, group ride and work out.
A cross trainer wanting cardio may only care that this week’s power is the same as last week’s even if it is not accurate.
Accuracy only matters for racing others and…can of worms…the KOMs.
I agree with you that should not happen. That 600w coffee guy is probably not you biggest worry because he is so far out of the race that no one will follow him so he is not really messing with the race. But there are those that is just a bit to strong also drinking coffee that we don’t even know about.
A category is the hard one and I really care what is happening there because all those with suspect setups end in A.
I would like to see that A is only open to validated trainers and Power meters. I don’t think it is possible to ask every A riders to provide proof of Power but there need to be something that can be done.
What just popped in my head is if ESport implement a system that all trainers need to transmit an identification code (probably encrypted) this will be on a model bses that way Zwift will always know what model trainer is used. But this will be a huge effort for the manufacturers and cost them money.
1 Like
Results, obviously. I think this point may have been made before though.
Where would one start if they did not race.
Even Zwiftracing.app use power to get people in the correct groups.
This is a flaw that, while only affecting the categorization of people right on the limit, is seen in everyone.
I did a race recently where I had the highest 20 minute power that I’d had in 3 months (I’m well down from where I was 4 months ago because COVID kicked my ass and I haven’t recovered). My zFTP actually went down about ~5W (~305W to 300W) because, in that race, my 3-5 minute power was much higher than my 20 minute and was what drove my 20 minute average so high (~320 W).
So yeah, the fact that I can set a new 20 minute power high and have my zFTP go down is pretty clear evidence that the categorization system is completely and totally broken.
Sure Pen Enforcement works fine, but how they determine which pen you should ride in is borked beyond repair.
At this point we need to discuss zFTP, zMAP, and categorization calculations as completely separate from enforcement of which pen you’re in.
1 Like
Right, use compound score to set your initial ranking for the first 3-5 races and then use your results for anything after that.
There are 2 schools of thought with regards to race ranking:
-
You only move up if you beat people who have a ranking that is higher than yours, and let your results expire after a certain amount of time so that you can move down (to some base level) if you get injured/sick/whatever.
-
Your results don’t expire, but, your ranking increases if you beat people with a higher ranking AND your ranking decreases if you lose to people with a lower ranking.
Option number 2 would lead to a whole lot more people quitting when they get dropped from the front group/see they have no chance of winning unless you institute some sort of “not finishing but going past 50% of the race distance or X km will give you a result behind everyone who did finished and ranked based on distance achieved among non-finishers” rule.
Not sure which I’d pick. I think option 2 (with the rage quit penalty rule above) would lead to more accurate race predictions, but option 1 would make me more likely to do races on days when I’m just not feeling it but know I’ll be more likely to get a good workout in if I do a race rather than a free ride/workout.
So they need to use some kind of maximal effort to place them in the ‘correct’ category.
People don’t have to rage quit just roll in to the finish in 15th position. Many people are happy to work hard during the race only to get 15th position.
It can be good team tactics to have people never win but make the first half as hard as possible.
They’d only need the maximal effort for the initial placement (unless you want your race ranking to expire after a certain amount of time and want a floor that it can’t fall beneath - my option 1).
Using compound score would be more race-relevant than FTP (as calculated by ZP) or zFTP, but those could be used if you really have to. It would only matter for the first 3-5 races anyway. For most of us, we have hundreds of previous races that they could just retroactively use, so it wouldn’t apply to us at all. They could just run all of our races (or our races in X time interval depending on which ranking model they want to us) and have us pretty accurately placed.
If the race ranking goes down when you are beaten by people not as highly ranked as you, that team tactic of going hard early to blow up the pack then staggering across the line will result in your rank going down. Maybe that’s something you want.