# Totally unrealistic

**URL:** <https://forums.zwift.com/t/totally-unrealistic/611267>\
**Category:** Racing\
**Created:** [August 29, 2023, 10:47am UTC](https://forums.zwift.com/t/totally-unrealistic/611267 "2023-08-29T10:47:59Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![Andre\_Hufschmid](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/andre_hufschmid/32/392904_2.png) [@Andre\_Hufschmid](https://forums.zwift.com/u/Andre_Hufschmid)\
**Post date:** [August 29, 2023, 10:47am UTC](https://forums.zwift.com/t/totally-unrealistic/611267/1 "2023-08-29T10:47:59Z")

</div>

I’ve been amused by the Stage Results lately. It’s quite interesting to observe individuals weighing around 40-45KG winning races against those who weigh 70KG on a flat route. Surprisingly, they aren’t generating higher wattage. To clarify higher wattage includes the ratio w/kg.This situation highlights the significant issue with drafting, showcasing a lack of realism. Consider a 70KG rider producing 5W/KG, yielding 350W. On the other hand, a 40KG rider would need to exert substantially more effort to maintain the pace, but this hasn’t been the case in these situations. Contrary to real-life races like the Tour, Giro, Vuelta, or any professional bike race, where a 40KG rider could never outpace sprinters due to the limitations imposed by physics. It’s a matter of curiosity as to when Zwift will finally address this discrepancy.

---

<div class="post-metadata">

**Author:** ![Steve\_Hammatt](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/steve_hammatt/32/210763_2.png) [@Steve\_Hammatt](https://forums.zwift.com/u/Steve_Hammatt)\
**Post date:** [August 29, 2023, 11:16am UTC](https://forums.zwift.com/t/totally-unrealistic/611267/2 "2023-08-29T11:16:49Z")

</div>

On which races exactly?

---

<div class="post-metadata">

**Author:** ![Andre\_Hufschmid](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/andre_hufschmid/32/392904_2.png) [@Andre\_Hufschmid](https://forums.zwift.com/u/Andre_Hufschmid)\
**Post date:** [August 29, 2023, 12:21pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/3 "2023-08-29T12:21:30Z")

</div>

Look at the outcomes of some Tour de Hisp races as an example (not implying any criticism towards the race, which is fun). Or pay attention when you race, ride or whatever you will notice even though you are considerably heavier than your opponents it does not matter if going up a climb, are on the flat or even downhill the lighter riders effortlessly keep pace.It’s evident that this Pack Dynamics lacks realism and comes across as outright implausible.

I have also tested it multiple times. A heavier rider can put down 500W and a lighter rider will easily follow doing much less W/KG on a flat. Its just plain bs.

---

<div class="post-metadata">

**Author:** ![Rich](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/rich/32/431690_2.png) [@Rich](https://forums.zwift.com/u/Rich)\
**Post date:** [August 29, 2023, 3:05pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/4 "2023-08-29T15:05:25Z")

</div>

speed/power is seemingly based off wkg rather than raw watts like it is IRL races, its been like this for along time

---

<div class="post-metadata">

**Author:** ![S\_A\_ccc](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/s_a_ccc/32/529686_2.png) [@S\_A\_ccc](https://forums.zwift.com/u/S_A_ccc)\
**Post date:** [August 29, 2023, 5:35pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/5 "2023-08-29T17:35:07Z")

</div>

zwift keep making the draft stronger with every new version of the pack dynamics they release. it seems to be what people want though. not that i’m complaining either

---

<div class="post-metadata">

**Author:** ![pmbdk](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/pmbdk/32/493694_2.png) [@pmbdk](https://forums.zwift.com/u/pmbdk)\
**Post date:** [August 30, 2023, 11:28am UTC](https://forums.zwift.com/t/totally-unrealistic/611267/6 "2023-08-30T11:28:27Z")

</div>

> [@Rich](#):
>
> speed/power is seemingly based off wkg rather than raw watts like it is IRL races, its been like this for along time

I honestly don’t think this is the case. Speed is based on terrain, rider’s weight, height (for drag coefficient), draft and (obviously) power. w/kg is probably the worst of all metrics to base anything on.

---

<div class="post-metadata">

**Author:** ![Rich](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/rich/32/431690_2.png) [@Rich](https://forums.zwift.com/u/Rich)\
**Post date:** [August 30, 2023, 11:34am UTC](https://forums.zwift.com/t/totally-unrealistic/611267/7 "2023-08-30T11:34:54Z")

</div>

yet you have lightweight riders putting out less raw watts but more wkg than heavier riders and winning

---

<div class="post-metadata">

**Author:** ![pmbdk](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/pmbdk/32/493694_2.png) [@pmbdk](https://forums.zwift.com/u/pmbdk)\
**Post date:** [August 30, 2023, 11:37am UTC](https://forums.zwift.com/t/totally-unrealistic/611267/8 "2023-08-30T11:37:29Z")

</div>

> [@S\_A\_ccc](#):
>
> zwift keep making the draft stronger with every new version of the pack dynamics they release. it seems to be what people want though. **not that i’m complaining either**

😄  
I guess you will never get the perfect draft feeling in a game. Especially when it is relatively difficult to position yourself in Zwift. When 10 cm (towards/away from/to the side) matters a lot in real life and you can’t judge the distance to the person in front of you better than maybe 1 meter, that it just how it is. We should just acknowledge the fact that this is a game, nothin more, nothing less.

---

<div class="post-metadata">

**Author:** ![Ciclista](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/ciclista/32/493603_2.png) [@Ciclista](https://forums.zwift.com/u/Ciclista)\
**Post date:** [August 30, 2023, 3:20pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/9 "2023-08-30T15:20:25Z")

</div>

Ok, if w/kg is so flawed then categorize by raw watts only and forget about w/kg. We’ll see how it works for all those big raw watt riders that are currently in the lower categories when they have to face the A cat riders.

---

<div class="post-metadata">

**Author:** ![pmbdk](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/pmbdk/32/493694_2.png) [@pmbdk](https://forums.zwift.com/u/pmbdk)\
**Post date:** [August 30, 2023, 3:54pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/10 "2023-08-30T15:54:22Z")

</div>

> [@Ciclista](#):
>
> Ok, if w/kg is so flawed then categorize by raw watts only and forget about w/kg.

Nah, just don’t categorize at all…

---

<div class="post-metadata">

**Author:** ![gloscherrybomb](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/gloscherrybomb/32/493046_2.png) [@gloscherrybomb](https://forums.zwift.com/u/gloscherrybomb)\
**Post date:** [August 30, 2023, 3:55pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/11 "2023-08-30T15:55:58Z")

</div>

This is such a painful debate.

I got beaten by a 16 year old in a real life pan flat TT a couple of weeks ago. He averaged 280w. I averaged 317w. I’m 72kg. He is a child. My CdA is around 0.19.

You can’t take either raw watts or w/kg as an absolute, and to do so is a total misunderstanding of both real life and Zwift physics.

If you want to know what a successful power metric to rank riders looks like, check out Compound Score, or even better, zrCS from ZwiftRacing.app. Statistically (with lots of data) proven extremely high correlation with cycling success in both real life and Zwift, which suggests the physics for weight/power must be reasonably close.

---

<div class="post-metadata">

**Author:** ![S\_A\_ccc](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/s_a_ccc/32/529686_2.png) [@S\_A\_ccc](https://forums.zwift.com/u/S_A_ccc)\
**Post date:** [August 30, 2023, 4:08pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/12 "2023-08-30T16:08:04Z")

</div>

compound score is probably as good as it’ll ever get. personally i will settle for any metric that i can explain to a normal person at this point

---

<div class="post-metadata">

**Author:** ![pmbdk](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/pmbdk/32/493694_2.png) [@pmbdk](https://forums.zwift.com/u/pmbdk)\
**Post date:** [August 30, 2023, 4:12pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/13 "2023-08-30T16:12:05Z")

</div>

Unless “compound score” (whatever that is) takes into account the actual race terrain, it will still not be able to predict “win rate”…

---

<div class="post-metadata">

**Author:** ![Steve\_Cl](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/steve_cl/32/19534_2.png) [@Steve\_Cl](https://forums.zwift.com/u/Steve_Cl)\
**Post date:** [August 30, 2023, 4:15pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/14 "2023-08-30T16:15:49Z")

</div>

Compound Score is 5mins pure Watts x 5mins W/Kg.

From link on Intervals.icu website [https://www.researchgate.net/publication/361643363\_The\_Compound\_Score\_in\_elite\_road\_cycling](https://www.researchgate.net/publication/361643363_The_Compound_Score_in_elite_road_cycling)

---

<div class="post-metadata">

**Author:** ![gloscherrybomb](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/gloscherrybomb/32/493046_2.png) [@gloscherrybomb](https://forums.zwift.com/u/gloscherrybomb)\
**Post date:** [August 30, 2023, 7:05pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/15 "2023-08-30T19:05:03Z")

</div>

It wouldn’t be able to predict win rate on a particular course, no. But we’re not trying to do that, so it’s not really relevant.

---

<div class="post-metadata">

**Author:** ![gloscherrybomb](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/gloscherrybomb/32/493046_2.png) [@gloscherrybomb](https://forums.zwift.com/u/gloscherrybomb)\
**Post date:** [August 30, 2023, 7:06pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/16 "2023-08-30T19:06:03Z")

</div>

Yup, and the ZR version is tweaked to account for riders with weights not typical of elite cyclists, so is even more robust.

---

<div class="post-metadata">

**Author:** ![pmbdk](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/pmbdk/32/493694_2.png) [@pmbdk](https://forums.zwift.com/u/pmbdk)\
**Post date:** [August 30, 2023, 7:27pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/17 "2023-08-30T19:27:03Z")

</div>

> [@gloscherrybomb](#):
>
> It wouldn’t be able to predict win rate on a particular course, no. But we’re not trying to do that, so it’s not really relevant.

What are we trying to do?

---

<div class="post-metadata">

**Author:** ![gloscherrybomb](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/gloscherrybomb/32/493046_2.png) [@gloscherrybomb](https://forums.zwift.com/u/gloscherrybomb)\
**Post date:** [August 30, 2023, 8:48pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/18 "2023-08-30T20:48:31Z")

</div>

Rank riders from top to bottom globally. But that’s not the point of this thread in particular, it’s the suggestion that W/kg shouldn’t beat W, or vice versa.

---

<div class="post-metadata">

**Author:** ![pmbdk](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/pmbdk/32/493694_2.png) [@pmbdk](https://forums.zwift.com/u/pmbdk)\
**Post date:** [September 1, 2023, 3:19pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/19 "2023-09-01T15:19:13Z")

</div>

> [@gloscherrybomb](#):
>
> Rank riders from top to bottom globally.

Exactly, and no score can do that without taking into account the route.

---

<div class="post-metadata">

**Author:** ![S\_A\_ccc](https://sea2.discourse-cdn.com/gozwift/user_avatar/forums.zwift.com/s_a_ccc/32/529686_2.png) [@S\_A\_ccc](https://forums.zwift.com/u/S_A_ccc)\
**Post date:** [September 1, 2023, 4:01pm UTC](https://forums.zwift.com/t/totally-unrealistic/611267/20 "2023-09-01T16:01:41Z")

</div>

i used to be convinced of this but having seen it in practise i’m not so sure… if you like to sprint, but suddenly pick up a newfound love for getting dumpstered in alp races, a good ranking system should be able to adjust itself pretty quickly as long as the fundamental algorithm is sound

the bigger challenge, and it’s actually a really hard problem to solve IMO, is finding ways to make sure people who race infrequently or take long breaks from indoor cycling before returning to it are seeded appropriately

[Next page](https://forums.zwift.com/t/totally-unrealistic/611267.md?page=2)
