Good news is, that’s for the mathematicians at Zwift to figure out, not us
BUT, point is, it does matter; so it should I think be calculated into seed score. It should bias taller people a few points down; if we’re aiming for be-all-end-all seeding.
As far as ZRL proceedings go I just stopped caring. After seeing a B get to rock a 5 minute power higher than all but a single A+ that night and getting extremely pissed off and being told that was “reasonable” for a Cat B I said fine… I’ll just stop caring; and I have lol.
The only good thing to come of ZRL is the quantity of racers, not the quality of the event by any sense of the word, but the physical number attractions it gathers.
Unfortunately for us in the Americas time zones, we don’t get the popularity of racing events, even come winter times… so it turns into beggars can’t be choosers.
Height 100% should matter for seed scoring. The best riders would be able to produce the least amount of power while still winning. And to produce less power, the shorter you are the easier it will be.
The thing, many riders will be able to win but likely they would be ZRS seed scored floored base on CP30.
If they have a very short height profile, CP30 will not be as high as they won’t need to generate as much power to get up to speed.
Everything you said, makes sense. And?
If you want to run a cat that is for the top 5% of zwifters, you make it scored 950-1000. No mysteries. If you want top 3% then 970-1000.
If you want to run a race with 5 cats, each cat containing 20% of riders from the top 30% of riders, and with even numbers, it’s simple to figure out 700-760, 760-820, etc…
prohibitively difficult, i mean. for example, when i showed the graph of TDF winner heights increasing, you noted that it tracks close enough with the increase in global average height over the past few years… but the data wasn’t measuring that. it just made intuitive sense to you to assume that because people are getting taller, successful athletes at the TDF should also get taller
… and disregarding every aerodynamic and biomechanical variable, of which there are vast and numerous possibilities, as inconsequential in the process. i agree with your assumption that short people are advantaged on zwift, but actually proving it to be statistically significant isn’t easy
personally i would just make everyone 175cm tall in zwift because i hate the mechanic. that’s an ill thought out solution in and of itself though
I’ve been pondering this exact problem because I think it strikes right to the heart of why its quite a tough challenge to make Zwift racing categories work. Not only is the fitness disparity curve not linear, the the population curve is also not linear - and they are non-linear in ways that are quite counter-productive if you want pens that are BOTH competitive AND well-populated.
Let’s take a hypothetical where we absolutely nailed getting racing score perfectly correct - it is spot on for representing each rider. If we have done this, there’s a very good chance the population curve will end up looking something close to a normal distribution bell curve maybe slightly weighted to the left with more population at the lower score range.
Meanwhile, if we take a performance capability curve with the x-axis being score and the y-axis being performance capability, we’d see that looks nothing like our population curve. Instead of something that looks close to a bell curve, we get a curve that asymptotes toward the horizontal axis as score goes lower and toward the vertical axis as score goes higher (i.e. much lower performance difference between 300 and 500 where the curve is nearly horizontal and parallel to the x-axis and a much wider almost chasm like difference between say 600 and 850 where the curve has gone almost vertical parallel to the y-axis).
Basically from 200-600, you have the population to support very granular pen score ranges but you don’t necessarily have the need to do so from a competition perspective. You could probably even go 200-wide on pens at that score level and still have a fairly competitive group top to bottom. I’m not saying you should - if population size gives you the option to go narrow on a category band, you probably should.
Meanwhile, say from 600-850, you very much have a need to be more granular with pen score ranges to keep things competitive top to bottom but you likely don’t have the population to support 100-wide score pens there. I don’t know how you solve this problem perfectly to get a well populated competitive pen at that score range - you can’t make the population performance distribution be different than it is.
You are basically left with a choice at higher score levels - optimize more for population or for competition. I guess I would just say that, while I understand if we are optimizing for population at high score levels, we have to make the pens wider there, I’d much rather have us optimize for competition and make them narrower. 350-wide score pens at the top-end feel bad. I’d honestly prefer a race with 5-10 people close to my ability than 20-30 people where it feels like high school vs the pros.
We input our height, and it’s calculated along in the CdA calculation. We all know it makes a difference, and Insider has the tests to prove it, so we know it does make a difference.
So there’s absolutely no reason for it not to be a subcategory of the calculation.
If it only affects the seed by 5% then… they should be able to make it affect the seed by 5%.
But height shouldn’t continue to be ignored; because outdoors where it “doesn’t matter as much” because everyone’s aero tucking abilities varies greatly…
In Zwift it’s static, and can’t be altered; short people will always be faster than taller people.
So there’s just frankly no reason for it not to be added to seed.
If we want fair racing; just because you’re extremely short means of course it matters less.
But it also means that someone with your equal power abilities can never go as fast as you in Zwift because of the code says they can’t… and if they do; they would have to increase power, and thus get an upgrade… while you wouldn’t.
Ignoring height in the seed calculation is unfair, plain and simple.
I’m agreeing with you, but when you feel strongly about something you have to go to the appropriate lengths… i hate the mechanic, but as a 5’4 rider the only solution i can be bothered to propose is this:
how would you approach it? like i said… people don’t think about height enough. me included
No harm, I wasn’t entirely following your train of thought / how you meant it.
It’s kind of the same reason I like to argue that frames should be grouped by type, and wheels grouped by type; but all else should otherwise be equal across the board.
So in that case, I’m mostly agreeing; mostly / not entirely because I feel like it’s taking something away; but what to put back in place, I can’t say.
Granted to be fair, everyone visually on Zwift looks the same (height) wise; the weight is nice to see.
I personally wish more body sizes and heights and frame sizes were in-game just for visual differentiation, but I know that adds a lot of extra work for something not all that important.
I get what you mean though; either calculate for it, or drop it; both are kind of “the same thing”
But how relevant is that so-called normal distribution when you can plug in the data for two people of vastly different abilities and the formula spits out the same seed score? A C-cat sprinter with a seed score of 550 is not the same as a TTer at 550 who never sprints. Until the seed scoring is fixed Zwift could probably run 3 pens: (0-150)(150-700)(700+) and the racing would be just as fair.
What’s not true? You said the same thing with more words.
This is a fact that holds true regardless of the terrain.
Increased gradient reduces the advantage significantly but it never disappears.
In this example the lighter rider is in the B category, the heavier rider is in the C category.
The other group that gets an advantage is those whose power is below the floor, but who have a weight that puts their w/kg higher than anyone else is allowed to have. Because yes, higher w/kg is faster too.
But taller people are always disadvantaged compared to shorter people, and that isn’t accounted for in the ZRS seed score. Which was the point I was trying to make.
I am not saying the same thing you did. Pay attention. Only people with more watts get advantage relative to people with lower watts, at the same wkg.
Heavier riders don’t get a special advantage whatsoever, never did, never will. Especially with the changes to pack dynamics, heavier riders are actually screwed even in flat races if they can’t match the wkg of lighter riders. The funky drafting in the game makes it very difficult to stay in the draft.
People with more watts get advantage relative to people with lower watts, at the same w/kg is exactly why w/kg categories give an advantage to heavier riders.
I feel like you’re disagreeing with a point I didn’t make.
You are still mind-numbingly assuming that being heavy somehow gives people more watts. Let me go eat some donuts to get this advantage that has been eluding me then. You are completely overlooking denizens of B riders with 3.2-3.5wkg and struggle to stay with the riders who can push 3.99999999wkg.
Going to hammer this into your skull: Being heavy, when everything else constant, DOES NOT GIVE ADVANTAGE.
No, but I also don’t think it’s that extreme either.
If it is, then I would suspect one of them is less likely to do races on Zwift at least, so at what point does it no longer matter, trying to chase a problem that might not entirely exist beyond theory.
At some point, “best fit” is just that.
Imperfect, but… the returns for effort become so small it begins to not matter.
It almost becomes the argument of “oh you are light weight, maybe avoid races on flat routes”
Which then just turns into the “how about I just don’t race because majority of races on zwift are quite flat”
Seeding can’t chase perfection forever is all I’m getting at; and if in your example, one or even both of those riders don’t race, then the chase to success is meaningless.
(and I guess just to be pedantic; there’s nothing stopping the “TT” rider to sprint, and then reset their seed when inspired to sprint for the first time)
And then to the whole point of this thread; even if the seeding is “off”, the whole objective is to get them placed correctly based on results.
But that just still involves doing more races. Thus… keeping people motivated to join races then becomes the biggest challenge.
Seeding matters; but in the grand scheme of things, it will stop mattering “eventually”
That “eventually” would come much sooner if the seed score wasn’t also the score floor.
It makes sense that there is a mechanism to prevent people throwing races to get themselves into an easier field, but I don’t think they’ve chosen the right one given how they’ve set the seed score.
“Security through obscurity” aka the stupidest way to manage anything because someone who really cares is going to figure it out no matter how well you try to hide things
I still give ZHQ the benefit of the doubt though as far as competency goes. Which leaves maliciousness as the alternate explanation.
Mainly responding because you seem to be getting angry about it, but you’re wrong.
Take an untrained 50kg and an untrained 90kg person and the 90kg person will put out more power.
There are likely a couple of reasons for that… you will naturally have more muscle mass if you are carrying more weight, and a pedal stroke is primarily weighted towards the ‘push’ phase where additional weight means additional force.
