Stop race category cheats - opinions?

Racing
265

I prefer “E cat” rides and instead of having an impossible goal of being 1st, my goal in a group of riders, might be the top 150. It’s just as rewarding (since there’s no actual Zwift tangible rewards for winning 1st). My goal of which tier to finish in might have to wait for a few miles to see how many and who I’m racing against as to what will be a realistic finish level. At the worst, there’s usually a few riders I can “try” to beat.

As the categories are now, I’m a new (lowest) “A” rider, so I have no chance of winning anyway and could be last in the group with no one to realistically compete with. In an “E” cat race, some top level “B” riders may be having a good day and I can race against them.

266

Keep in mind that the suggested group watts/kilo are based on 20 minute efforts. Therefore in a shorter race it’s quite normal to see higher watt/kilo averages. As Zwift grows you have Poole strictly training for short races as it fits there physiology. This doesn’t explain a 5.1watt per kilo, but it is something to keep in mind for short sprint races.

267

so now its middle of november. nothing has really changed since april. still riders with 4.5 wkg are entering C or D races. Still it is frustrating that zwiftpower is now owned by zwift and i think it will disappear in a near time period…this thread shows that er are a lot of points that can be changed by zwift. but nothing will improve. it is really sad…remember its november :frowning:

3 Likes
268

Thinking from a business point of view, it is easy to understand why Zwift does nothing about cheaters or about physics that do not match the speed that cyclists would reach in real life.

It is clear that only people who really have high speeds in real life will complain, and those people are really a minority.

From a business point of view, it is better for Zwift to make other people happy (possibilities of cheating, unrealistic speeds) because they are more numerous and that means more income from fees paid.

I think it’s that simple.
:relieved: :wink:

2 Likes
269

A real Cat C rider doing 2.8 W/kg cannot stay on the wheel of a rider doing 3.5 w/kg on zwift. No way. No how.

2 Likes
270

It’s a real big problem and I think it’s actually getting worse. I say one Category.

271

I’m sure its possible if the person going 3.5 w/kg is much lighter than the other

1 Like
272

It’s certainly possible, real C riders frequently get pulled by sandbagger C’s putting out 3.5+, they end up minutes ahead of the rest of the legit C’s while outputting similar power to the guys minutes behind them

I can put out 2.9 and get pulled along by guys doing 3.5 all day

2 Likes
273

My suggestion is to have at least one series of races that require you to do qualifying races that determine your category before you can enter. If you were riding at a lower wattage to cheat the system in the qualifiers, determined by your output in the official series races, you get DQ-ed from the entire series. The first race or two in the series might have sandbaggers but they should get weeded out pretty quickly.

1 Like
274

I and about five other “real” C riders got yanked around Sand&Sequoias for a lap and half by a sandbagger C rider putting down a consistent 4.2w/kg during the ODZ event a coupla weeks ago. It took 3.2+ to hang on. He dropped out right after the rollers on the descent so he wouldn’t get flagged. Since I spent the better part of the race on the rivet, I cannot remember his Zwift handle to look him up. It did make for an incredible PR on the course…

1 Like
275

The real problem is that it anything like a flat or downhill the watts count but the w/kg is irrelevant. If like me you weigh 62kg you end up in Cat B but you cannot possible compete with C cat riders putting out more than 250 watts let alone B Cat riders, It is the real flaw in the system and has generated much of this discussion about not wanting to upgrade, Of course there are always people who want to cheat and like winning at any cost as well just as there are drug and mechanical cheats in live Vet time trials with a max prize money of £50. These people are even more pathetic than Lance Armstrong as they don’t even have a valid motive other than it makes them feel good at everyone else’s expense.

276

There are two reasons for the sandbagging that are often looked over: 1.) there are too many races on any given day and 2.) the A category is way too big a w/kg range.

I’m an A racer. I got upgraded from B to A in February (I think) when my FTP got updated to 4.01 W/kg. Human performance improvement generally follows a logarithmic curve, which means that going from 2 to 3 W/kg is easier than going from 3 to 4 W/kg, which in turn is easier than going from 4 to 5 W/kg. etc. So a properly designed category system should start with bigger ranges of W/kg in lower levels and those ranges should shrink as it gets bigger. But instead, C is between 2.5 and 3.2 a gap of 0.7, B is 3.2 up to 4.0 a gap of 0.8, and A is 4.0 to 6.0, a gap of 2.0. The categories get bigger when they should be getting smaller. This means that someone at the bottom of the Cs is closer (in terms of training hours) to someone at the top of the Cs (and therefore closer to someone at the bottom of the Bs) than a person at the bottom of the Bs is to the top of the Bs, and again, someone at the bottom of the Bs is closer (in terms of training effort) to someone at the bottom of the As than someone at the bottom of the As is to someone at the top of the As.

This is a real problem, because I went from mostly being able to actually race in a B race (and occasionally podium, though I never won one) to always getting dropped in every A race. This sucks, because I’m doing races for fun and to get a better workout by using my competitive nature to push a little harder than I would on my own. But if I get dropped, I’m just on my own anyways, so what’s the point? I’m very tempted to race Bs again just so I can ride with people again. I was mostly doing between 3.8 and 4.2 W/kg on 20 minute crit races as a B (with three really great races of 20 minutes at 4.3 W/kg where I seemingly peaked, and got upgraded to 4.0 W/kg FTP). Since switching to A, I mostly do around 3.3 to 3.5 W/kg, because once I’m dropped, its just not interesting anymore. This problem would likely go away if I were racing only people between 4.0 and 4.4 W/kg, but I’m racing actual pros (sometimes) who put out 5.5 W/kg for an entire 20 minute race and start the first lap or two at 6 W/kg. I just raced today and got dropped on the second lap of the dolphin after averaging 4.7 W/kg just to stay in the back of the peloton. Again, it sucks. So I can completely understand why lower level A racers drop down to B to sandbag, because I’m tempted to do the same (but haven’t). And then that makes it harder for the B’s, so now you get a trickle down effect where B racers on the lower end are going to want to drop to C’s and C’s down to D’s and it makes it bad for everyone.

So the problem is the categories are poorly thought out and make the cascade of trickle down sand-bagging happen. Why are the categories so bad? Well, this is obviously because there are too many races. In lots of races I sign up for there will be 20-30 B’s signed up but only 10 A’s. That’s why Zwift has made A is such a big range–if they made it a proper range the A races would have 5 people in them and the A+ races would have 5 people in them, and Zwift is trying to make the races bigger. This is because there are just not enough racers in the 4-6 W/kg to fill a race every 30 minutes.

If they offered fewer races it would be slightly less convenient, but they’d be able to fill up the A races better and then could break it into several (probably 3) categories (interestingly, the W/kg limit for A roughly corresponds to Categories 3,2, and 1 in UCI racing). If A was 4-4.5 W/kg, I bet there’d be a lot less sandbagging. I can hang in the back of a peloton doing 4.4 W/kg at the front and I wouldn’t be tempted at all to sandbag. But I will get dropped from a peloton doing 5.5 W/kg every time. (BTW I have not/do not sand-bag, but I really strongly get the urge to do it. I loved racing and got a lot faster doing it until I was upgraded, now I basically avoid it… thank goodness we’re coming out of winter and I can just ride outdoors. Zwift is about training in a more enjoyable way. Racing B’s was enjoyable. Getting consistently dropped from the A’s, which I won’t call racing, is not at all enjoyable.)

3 Likes
277

Well, what about this solution then, @John_Bowers :

Imagine a world where groups of football teams were divided into divisions based on how many games they won during a season instead of, like today, by measuring the players’ Watts? You know what it’s like today in football. Last week Manchester City was on top in Premiere League, but then Cancelo went over W/kg limits in a duel against Chelsea defender Silva and the whole team got a DQ.

So instead of promoting the football teams that went over the W/kg limit to the next division you would move the one or two teams that won the most games during a season. Wouldn’t that work too?

What if Usain Bolt had earned his way into the Jamaican Olympic team by winning track & field events on Jamaica and elsewhere instead of the story we all know so well, with those scientists putting him and other athletes in this sports physiology lab and him coming out the highest on the W/kg tests? It might have worked too.

And maybe it could work in real-life cycling too? What if riders like e.g. Dekker, Champion and Aleotti hadn’t been upgraded to UCI WorldTeams for the 2021 season because they busted out of W/kg limits in races? What if they instead had been upgraded because of their past successes in UCI ProTeam? And similar at the amateur level. What if a measly US cat 4 rider was upgraded to cat 3 not because of his W/kg but because of placing well in previous cat 4 races? I have no idea exactly how a system like that would work in reality, maybe some points system or something, I dunno but… maybe some brainiacs could figure something out.

I know this is a heretic thought, so don’t tell anyone I said this, but what if the same would work for Zwift too? Then we wouldn’t have to define appropriate W/kg bands. They would be what they are, whatever they are. They would probably change from race to race anyway and we wouldn’t have to care. And you wouldn’t have to worry about getting upgraded too early - just don’t win too much and you won’t have to. And if you ended up racing alone, barely in front of the virtual broom wagon, so to speak, then you would get downgraded soon enough due to poor placings and you could once again race with people around you.

Actually, when I think of it, it would be awfully hard to cheat too. If everyone was forced to race only in the category they belong in, then if some guy somehow managed to cheat anyway, by cruising races or something, then he would win a lot of course. But winning would also upgrade him. He couldn’t keep on winning in the same cat over and over. And he would face real competition in the next category. So somehow it seems to me the chances of getting to race with at least some others of roughly equal strenght in races would be fairly good. Regardless of whatever W/kg they happen to be pushing for whatever reason.

Just a thought…

4 Likes
278

Well said @Andreas_Traff . Quite why the obvious logic that is coincidently used by AFAIK 100% of cycling leagures outside of Zwift is not applied here is just plain strange . We have a situation where either they are not prepared to think rationally about it , or just dont see it worthy of consideration. Either way no good will come out of anything in this space until categorisation is based on results not power metrics.

2 Likes
279

@Andreas_Traff You are a genius!

1 Like
280

@Andreas_Traff Couldn’t agree more. Although probably to host as many races as Zwift needs to host to keep users happy the right approach is something like how every other online game does matchmaking based on an ELO score (that is computed by who you beat or lost to, so its essentially a dynamic version of rankings like the UCI).

But yes, I concur. Everyone knows this is the right fix, presumably Zwift knows this is the right fix, so why haven’t they done it? I really wish they’d stop being silent on this.

4 Likes
281

Which of course is avaialble to Zwift because it is stored and calculated in Zwiftpower already !

Dont understand what the problem is , or dont care . Select your choice .

3 Likes
282

Like you say, this ELO system, or one such out of many possible at least, is already in place on ZP. And it has to my knowledge never been used for anything except maybe questionable bragging rights among riders. However, right now the individual rankings are highly unreliable since people are getting ranked under the W/kg rule set with all its flaws. But the current ranking system could definitely be used. Ditch W/kg and the rankings will become more reliable too.

I remember some people have suggested dynamic sizing of cats in a race. Not very UCI-like and very online game-esque, but it’s not a bad idea. Actually, it could well be how Zwift could max out on customer engagement in races. So if you have 200 participants in a race you would just split them into four (or whatever) pens based on their current rankings in that particular race. Easy-peasy.

Or you could be a little more sophisticated and somehow weigh no of participants in a pen against proximity in ranking between participants, since both are important for the fun factor of a race. (Typical machine learning task to get the balance right, should be doable.) I’m thinking it’s not fun if your cat/pen is too small and you don’t always want to race very large fields either. But it’s also not much fun if the spread in rankings in a pen is too wide. Maybe you could even make the no of pens dynamic to make the fine-tuning of fun more accurate, i.e. not always A-D, but you could have 5 or 3 or 7 pens depending on turnout and field.

The upside to ranking, one of them, rather than a points system with fixed cats, is that not only do you not have to worry about W/kg bands at all. You also don’t have to worry about the impact of and reward scheme for races of varying lengths. It’s already taken care of implicitly by the ranking model.

In any case, ELO rankings or points with fixed categories, I don’t care. Anything results-based would be vastly superior (and more fun) than what we have. It would even be hard to go wrong since we’re at rock bottom already.

6 Likes
283

Just race B John.

Zwift Power is broken and utterly pointless. Enforcement, after the fact, is a truly idiotic idea.

You pay to use Zwift. Use it how you enjoy it. Obviously, don’t do this at others expense. If you find yourself winning regularly then you’ll simply have to move up.

Or just participate in the Bs, don’t negatively affect the race, don’t go off the front, don’t sprint the finish, pull over at the end etc. I can’t see how that would negatively affect anyone.

1 Like
284

Will the new race score help address this, or will it penalize honesty?