Page 1 of 4

Better ranking system than A/B/C/D and W/kg?

Posted: Sun Mar 03, 2019 11:24 pm
by profile.php?z=938283
I've been pondering how bike racing (and Zwift racing in particular) handle ranking and categorization. I realize that grouping into broad classes has made a lot of sense in the past. In real life, having Cat1 racers race with Cat5 racers can be dangerous!

But Zwift is different. Wouldn't it make more sense to do something like the ELO or Glicko rating systems, used in Chess and other zero-sum games? https://en.wikipedia.org/wiki/Elo_rating_system or https://en.wikipedia.org/wiki/Glicko_rating_system

The math is a bit complicated, but the way it works is intuitive and fair: You get points when you win, you lose points when you lose. If you beat a stronger player you win more points than if you beat a weaker player. If you lose to a stronger player you lose fewer points than if you lost to a weaker player. Over time the math makes it easy to compare competitors.

When a person completes a race on Zwift, you could calculate the pairwise scoring against every other rider, then integrate the results to get the new rating.

Many benefits from this:

1. Rating would be based on performance history.
2. The math has probabilities associated with it. If some magically performs far above/below their rating, the math can help in the detective work.
3. The ratings can be recalculated trivially based on age / gender / nationality / whatever buckets. For instance, older riders could know how they stand in the "female 60-65" group. There is really no way to know this if we only have A/B/C/D.

Sidebar: Is the ZwiftPower data public? I would love to take a crack at what I just described, and see what the result looks like. I'm a software engineer by trade and I have the skills to make it happen, but I don't have the data :-). (I'm sure I could scrape the data with a crawler, but I don't want to violate any terms of use :-))

Cheers,
-Mark Rebuck
http://www.markrebuck.com/

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Sun Mar 03, 2019 11:41 pm
by profile.php?z=938283
(I realize we have rankings.php , but that's just a straight implementation of the real-world USAC rankings. Zwift races tend to have larger fields, with a larger variance of rider strengths. USAC rankings feel like the wrong tool for the job in Zwift races.)

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Mon Mar 04, 2019 4:38 pm
by profile.php?z=411944
There are upsides and downsides to any system.

There is a major downside to this idea in that it would be very difficult for a casual D or C level cyclist (who has zero aspiration of becoming an A) to be successful.

Many (if not most) beginners on Zwift are NOT beginners to cycling. There is a large influx of people signing up for Zwift at any given time who are already competitive category racers in real life.

On your ranking system, with no history, they would constantly be winning the lower category races. Sure, they would quickly move up to the higher categories, but next week another Cat 1 4.5 w/kg superstar is going to sign up for their first Zwift race. This guy will blow the Cat C/D racer (who will never exceed 3 w/kg in their life) into the weeds.

The lower category racers will get discouraged and it will become a dog eat dog free-for all for the big guys.

The current system allows lower category riders to have a place to race and be competitive. It's not perfect, but it is more inclusive to all rider abilities which *in my personal opinion* makes more sense.

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Mon Mar 04, 2019 7:50 pm
by profile.php?z=7280
This suggestion has some merit but seems like a complicated arrangement.

I think at the end of the day people want to race with people of similar abilities and have the race be interesting in the sense that they can be competitive in their category.

There seem to be a few problems:
- sandbaggers that stay in too low of a category for too long
- cheaters (weight/height doping, etc)
- w/kg is not an effective measure of category for most zwift races since it's w/Cda that matter more unless the race has a 10-20+ minute climb

Forced upgrades would take care of sandbaggers - if you've placed top-3 in say 5 races (or you win by a significant margin) you should be forced to upgrade.

Cheaters is a tough one but Sticky and others seem to be doing a good job busting people.

The w/kg conundrum is somewhat more complicated. Smaller riders have to produce 15-20% more power just to keep the same speed on flat/rolling terrain. This can get you bumped into a higher category where you get thrashed.

Perhaps we should just fall back to a reputation based system. That is if enough people are suspect of your results you either get pushed to a higher category or temporarily banned. Even though I got bumped to be an "A", I'm nowhere at the level of the rest of the A racers nor do I want to be at this time of year. I was quite happy racing in the B's and I don't think anyone would call me a sandbagger (i.e. it wasn't like I was winning every race). Later as the year progresses I'll be more interested in going A level hard. But right now I doubt anyone that I was racing against back in January or early February would accuse me of not belonging in the B's.

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Tue Mar 05, 2019 9:06 am
by profile.php?z=164586
dgiampaolo wrote:
Mon Mar 04, 2019 7:50 pm
if you've placed top-3 in say 5 races (or you win by a significant margin) you should be forced to upgrade.
This is a tricky debate though isn't it... Back in 2018 i won 5 C races in a row... . yes 5... on the bounce...

In 2019 my power is no lower, but i've not won a race in 2019.... I've not changed...

People want to race in a Cat where they get the most enjoyable racing... For me that's racing in C, as i'm at the reasonably pointy end for the majority of the race, until we hit hills as at 89kg i'm history ! If i jump in the Bs i'm left lonely and bored with the Cat, propping up the wooden spoon positions.

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Tue Mar 05, 2019 3:53 pm
by profile.php?z=7280
Well if you won 5 races in a row that's pretty dominant... so maybe the alternative is you're forced to take a break and let others have some fun OR you get pushed into the B's.

That seems like a reasonable suggestion to me: people that don't want to push themselves at the next level just have to take a break (say 2 weeks). Or if they do want to push at the next level they take the upgrade and get beat up until they get stronger.

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Tue Mar 05, 2019 4:12 pm
by profile.php?z=209977
dgiampaolo wrote:
Tue Mar 05, 2019 3:53 pm
Well if you won 5 races in a row that's pretty dominant...
When i race in the morning at 7am and the "field" is only 3 riders, it is no problem to be dominat. And when i do this every morning a week, then i have to upgrade?
I follow you, when you win the KISS race every week.

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Tue Mar 05, 2019 6:31 pm
by profile.php?z=7280
good point - a race with a low number of entrants probably shouldn't count against you since obviously getting 1st of 3 usually isn't as big of an accomplishment as when there are 20 or 30 people in the race.

USACycling deals with this by assigning points based on the number of entrants and the duration/difficulty of the race (i.e. a road race win against 50 guys counts more than a crit win with a field of 10 guys).

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Wed Mar 06, 2019 6:35 pm
by profile.php?z=884482
I think we just need a B+ and C+ catagory like they did with the A’s

Re: Better ranking system than A/B/C/D and W/kg?

Posted: Fri Mar 08, 2019 3:54 pm
by profile.php?z=107292
Cjostberg wrote:
Mon Mar 04, 2019 4:38 pm
There are upsides and downsides to any system.

There is a major downside to this idea in that it would be very difficult for a casual D or C level cyclist (who has zero aspiration of becoming an A) to be successful.

Many (if not most) beginners on Zwift are NOT beginners to cycling. There is a large influx of people signing up for Zwift at any given time who are already competitive category racers in real life.

On your ranking system, with no history, they would constantly be winning the lower category races. Sure, they would quickly move up to the higher categories, but next week another Cat 1 4.5 w/kg superstar is going to sign up for their first Zwift race. This guy will blow the Cat C/D racer (who will never exceed 3 w/kg in their life) into the weeds.

The lower category racers will get discouraged and it will become a dog eat dog free-for all for the big guys.

The current system allows lower category riders to have a place to race and be competitive. It's not perfect, but it is more inclusive to all rider abilities which *in my personal opinion* makes more sense.
Imo, categories based on wkg are really unfortunate, and I say this as someone who is (when in shape) on the upper edge of the B-cat. Just because, coincidentally, I have an FTP very close to 4, I am able to win (B) races, or at least compete for the top spots, even though I have at best average 1m and 15sec power. Someone who has an FTP of 4.1, but an otherwise similar power profile, is doomed to get dropped in the first couple of kms of the A race, forever. Obviously, I have little incentive to improve my FTP. And the 4.1 rider has all the incentive to sandbag the B cat. Let's be honest, nobody at the top B cat really looks forward being promoted to the A cat (especially not those riders who are close to their potential). And that means the cat system is flawed. With a (periodic) system based on race results, these edge riders would fall back and forth between A and B. Getting promoted would become a goal, not something to fear.

To avoid the dog-eat-dog situation, you could combine such a system with wkg limits, initially. Or alternatively, come up with a system where the D and C cat are based on WKG, but transition between B, A and A+ is based on race results (where top (5%?) riders in each cat are promoted at the end of a given time period - say, 1 month, based on their (USAC) race ranking)? Eg,

Riders are obliged to race either in their assigned, or above their assigned categories. Categories are based on the USAC ranking system. Category membership is evaluated at the end of each month, looking at the prevailing USAC ranking within each category. The top 5% riders are promoted, the bottom 5% are demoted. To avoid the chaos that would prevail in the lower categories as stronger riders (who don't have the decency to sort themselves into the cat they belong) make their way to the top, lower categories still have a wkg limit which when exceeded excludes you from the results.


I understand the challenges in organizing this, but this definitely deserves a serious debate!