when I got bumped to A I was finishing in the bottom 3 most races and still do in a variety of different courses, what it has done though is given me the motivation to try and improve on certain areas I’m weak at such as my sprinting, my first A race (which wasn’t only me or 1 other person involved) was 10 laps crit city, I was dead last and over 2 minutes behind the winner, next race seaside sprint 8th from 9 in A cat and beaten by B riders due to it being a mass start and I put this down to being a lightweight(ish) rider at just under 71kg, I still don’t see myself as an A rider but I don’t see any point moaning about it (anymore than i did at the time as it happened 2 races into the wtrl series so I could no longer race with friends) but at the same time I’ve embraced being in the A cat and also started improving further
coming over the line at the ■■■■ end of a race can be disheartening but it’s all part of racing
I suppose. I mean, that’s exactly how I feel most of the time. I’m not a very good sprinter either. That just means I will lose the 500m sprint at the end of the race. I can sprint up a hill, but I can’t do that very many times before I get dropped (by all the A riders and the B riders heavy enough to stay in B). Most people would just drop back to the next peloton and draft, which would eventually put me in B as it’s less work. Seems like that’s what most people do (is that sandbagging???). But I can hold more power, so I ride alone, which pushes me into A and the cycle continues.
What bothers me (as you also experienced) is all those lower cat riders finishing ahead of me. Now, you’ll have to excuse me as I need to go buy more fruitcake; my weight is still too low.
Yes it’s not just weight, though that’s a factor. It’s simply that 20 min power isn’t all that relevant to race performance. Differences of 0.5W/kg and upwards are probably decisive, but up to 0.2 certainly isn’t. So a hard threshold means some will be perennial winners and others will be upgraded before they are really challenging for the front.
Agree, the main issue with ZP groupings are they are not time related. i.e. If you can do a 20min race of 3.2w/kg, you will not be able to do a 60min race of same power. For example, at my absolute fittest i can do a 3.6wlkg for 21min race. However my best for 40mins is 3.3wkg and 3.1w/kg for 60mins race. I have high FTP of 310W so i like doing flat courses where i can sprint at end. My physiological cost for 20,40 or 60mins race is low-high Z4 HR with Z5 in sprint so i am not holding back and often holding on. So i choose B grp for short races (end towards bottom) and C grp for long races (end towards the front). Problem is that the stupid ZP groupings are too rigid and dont account for race time in categorising. There should be a time modified groupings not just range based on 20min power but based on the expected race time for that group.
That might be a better system but quite hard for some to estimate time so everything right now is based on best 20 minutes whether you are entering a 20, 40, 60 mins race it should be done based on your best 20 min power.
Obviously that’s how it works in theory except there are number of racers that never do a 20 minutes effort in races so artificially keep their cat lower than it should be…ZP cat system is long overdue a rework to make it fairer.
I am aware that ZP uses 95% of 20mins as its guidelines but as i stated the race categorisation do not take into account the length of a race. Whether the race is 20mins or 60 mins the grps are based on the same range. This is not realistic for the simple reason that some heavy powerful riders have great 20mins power but blow up whereas smaller lighter riders wont have the same 20mins power but can maintain longer endurance. If you had the same system in athletics (for example) you would be basing all races on an individuals athletes 100m time whether the race was 100, 200, 400, 1500m etc. I.e. Athletics has a time / distance based component so what they can do over 100m is irrelevant for the marathon!!.
Why would it be so difficult? MOst races finish in about the same time. Rough example being. Cat c race of 20mile on 100m elevation (i.e. flat) see majority avg 25.5-26.5mph. Cat B on same course approx 28mph and cat A approx 30mph. So you have distance and avg pace so you can set the time component and vary the w/kg to fit this. Just a mathematical algorithm and given the number of zwift races been done to date, i am sure the data has been collected for 10min, 20min, 30min,++++ races. So you should be able to vary the categorisation based on something more realistic.
this is what the WTRL Autocat chase races are using to the best of our knowledge. They put you in the race pen based on how fast they think you will complete the course, not your w/kg.
I understand your comment. But lets just state that using w/kg is already a huge simplification and is basically just weight classes. w/kg is only a real predictor of performance on climbs above 5%.
Way before Zwiftpower and starting pens We used w/kg over 20 min to make it easy to get people grouped.
Maybe i missed something, but my understanding was that Autocat still only bases its data on your 1, 5 and 20min times. Then it auto pens you based on 6 potential racing speeds. OK, this is a small improvement but does not address the main issue that i mention above that these short based power times lose relevance for longer races. Unless the autocat is done as a 60min for 60 min race, 40min for 40min race etc then you are not really addressing the primary issue just putting a bandaid on it.
Appreciate you have to make some generalisations around groupings but the current system penalises riders who mix the type of racing they can do. Sometimes i only have time for short 10miles and sometimes i can go upto 30miles. But if i want to remain “unfiltered” in ZP then i have to deliberately start underperfoming for 3 months in 10miles because of the lack of relevance that this race has to a longer race.
Below is taken from Zwift site about the WTRL Autocat!
"WTRL and Zwift are working together with a multitude of Zwifter’s ideas and feedback, to test and trial several potential new Zwift Racing Category systems. This is no easy task! Prior to the Classics and during the Zwift Classics, we will be employing a test version of an auto-categorization system with the following factors in mind:
6 Rider Classes based on average racing speeds.
20min, 5min and 1min power ratings.
Race Results and course difficulty based TrueSkill style system that influences category upgrade/downgrade as well as providing a ranking system.
We will be testing 1 and 2 from above initially and we hope to see some of the most competitive racing classes available today with a reduction (to ZERO) in the number of incorrect category or sandbagging riders influencing races unsportingly. It is early days though!"
early days though while 95% of riders appear to be in corrects cats there is still some odd stuff going on like the clear cat b put in the d pen yesterday and soloed away for the win.
Plus of course AutoCat is for wtrl events only which is just a small fraction of races available.
But unless the fundamental point of Autocat addresses the limitations of using short time power assessments for all races then all that Autocat is doing is polish a t**d. Isnt this just the very definition of marketing over development?
somewhat true bit even taking 1/5 min power as well as 20 will be far more accurate than what we have at the moment. Agreed though it would be nice to throw in 60 mins as well
Mass-start racing is about trying to get to the final sprint with as little fatigue in your legs as possible (ideally while trying to maximize the fatigue of your competitors), so the fundamental flaw is using power metrics instead of race results as a basis for categorization in the first place. So yes, very much turd polising no matter what.