How the (Scored) Race Was Lost: Testing Zwift Racing Score on France’s R.G.V.

When I saw the news that Zwift was rolling out their new Racing Score metric in various test events, I knew I had to give it a go.

My first thought was to jump into one of the 5-category ZRacing events that were using Racing Score. The prospect was especially appealing because my initial “seed” score was 570, putting me right at the top of the middle category, which was for racers scored 400-575. Maybe I could find my way onto the podium!

But it wasn’t meant to be. Zwift reset the seed scores to fix a bug soon after launch, and mine went up to 588. Plus, schedule constraints put me into the race starting at 10:10am PDT, and this race was using Zwift’s experimental 10-category setup. In this scheme, there are actually two events – a “high” and a “low” – due to Zwift’s current (and silly) limitation of 5 categories per event. Each event has 5 categories, and my score put me in the lowest category of the “high” event, with riders between 550-600.

So I was still near the top of my category. On paper, at least.

The cool thing about results-based rankings, though, is that when done right, they reflect your race craft, not just your pure power numbers. Everyone was initially seeded based on their power numbers, but in theory, someone like myself with (*checks ZwiftPower*) 1141 races under their belt should have the smarts to outperform others who have the same physical capabilities but much less experience.

So how would I perform in my first-ever scored Zwift race? There was only one way to find out…

Warmup

I chewed two pieces of caffeine gum as I jumped into the Maria Robopacer group which had recently been relocated from Makuri Islands to France as part of Zwift’s big TdF July focus. After warming up a bit with Maria it was over to Coco to ramp up the effort a bit, put in a few harder pulls, and generally break the ice so my heart and legs wouldn’t be surprised by the race effort.

Then I headed to the pens for one lap of France’s R.G.V. route. We had 43 riders ready to go.

The Start

The start was rather tame, with nobody pushing hard from the line.

The attacks would come soon enough, though. One of my DIRT teammates (Captain Diogo) went early, moving to the front for a hard pull, ramping the group up from 40kph to 46.

The Middle

This pattern would repeat itself throughout the race. Someone would move to the front and push hard, others would chase, and our speed would ramp up to 46-48kph. Eventually everything would come back together, we’d catch our breath, and a few minutes later another attack would go.

The group wasn’t thinning much, though. This was a flat course, and as long as you stayed in the draft, you could stay in the group holding around 3.3 W/kg. Still, every attack dropped a few riders, which meant we had gone from 43 to 33 by the time we rode through the Pavé Sprint.

The biggest push (other than the finish) came, unsurprisingly, up the Aqueduc KOM Reverse. It’s really more of an uphill sprint than a climb, and while I’ve been dropped here before in mixed-category races (*shakes fist at A-cat*), I can usually hang with a group of B’s up and over this rise.

The same was true today: I stayed well-positioned mid-pack, triggered my draft boost powerup, then recovered on the descent that followed. We went from 33 to 31 riders, and everyone settled in for the final push through Les Intestines to the line.

The Finish

Les Intestins are always a tough push at the end of a race: 3km of twisty, rolling road, followed by an all-out sprint! I stayed in the draft, steered as much as possible to shorten my lines, and generally tried to save my legs as much as possible for the final sprint.

With 500 meters to go the road flattened, the twisties ended, and 31 riders were tightly packed into the front group.

I triggered my draft boost as we turned right onto the main road and avatars got out of their saddles. Then I dug deep, pushing with all I had to the line. Riders just ahead took the corner wide, so I chose to stay in the draft on their wheels instead of cutting the corner with steering.

My legs struggled to hold high power, though. I crossed the line in 8th.

See results on zwift.com >
See results on ZwiftPower >
See activity on Strava >

Watch the Video

Zwift Racing Score Results

The in-game results screen (shown above) didn’t reference Racing Score in any way. But I could see my Racing Score result on the event’s page on zwift.com, as well as in the Companion app. Here’s are the zwift.com results:

My score heading into the race was 588. Finishing 8th earned me a score upgrade to 594. (It’s worth noting that this isn’t like ZwiftPower, where your ranking is derived from your 5 best results in the last 90 days. With Zwift Racing Score every race counts, and your score is calculated at the end of each race based on that race history.)

Checking the results on zwift.com, it appeared as though every single rider in my race got a Racing Score upgrade from their result. That is, they all had an up arrow beneath their result! I reached out to Zwift to ask how that could be possible, and it turns out… it’s not. The wrong arrow is showing some cases, so while some rider’s scores did decrease, it’s only showing up arrows for everyone on the web results. Companion app results show the arrows accurately.

One thing I did note when browsing results in Companion is how startlingly accurate Zwift’s seed scores seem to have predicted the results. In my race, in all 5 categories, there isn’t a mix of up and down arrows as you scroll down the list. Instead, all the upper finishers get up arrows, and all the lower ones got down arrows. This would only happen if people finished right around where their seed scores would have predicted them to finish.

(I did find one rider in the highest category who had an up arrow instead of a down. But just one rider.) Zwift says they’re checking into this, because they noticed the same pattern and think it’s a bit suspicious.

Takeaways

What was my actual experience in the race? It was good. It felt like I was pushing at a typical B-category pace, but it also felt different than a typical B race, because riders were putting in digs that forced us to chase. It was less of a boring, steady effort and more like several hard efforts with recovery in between. It felt (and I’m too lazy to do the homework to confirm this) like we had several riders with high FTPs but poor sprints who were trying to get away off the front, or at least stretch out the group to thin the herd before the final push. And that’s good, straight-up bike racing.

In the end, our starting group of 43 became a finishing pack of 31 sprinting for the win. That seems like a sensible level of attrition for a flat route like R.G.V.

One final takeaway: I forgot the draft powerup now lasts 40 seconds instead of 30. This change was made over a year ago, so I really should have it locked in by now. It’s a crucial mistake, because I could have triggered the powerup 10-15 seconds sooner and enjoyed its benefits heading into my sprint effort.

Your Thoughts

Have you tried a scored race yet? What do you think of Zwift’s new Racing Score? Share below!

Bike Racing Tactics: 9 Tips for Zwift Races

Racing tactics. I see posts regularly about riders sitting in and then sprinting at the end, and how that isn’t right, or variations of that.

I think people don’t realize that’s how racing works. I’ve been racing a long time (40+ seasons). This is how racing works: you go as easy as possible until you have to go hard. And when it’s hard you absolutely have to stay in contact. If you can’t, your race is over, because once you lose the draft you’re done. If you can’t hold the wheels while saving 30% energy from the draft, there’s no way you can go faster than that while NOT drafting to catch back on.

Tip #1: Don’t Get Dropped

Therefore, the first priority in a race is to “not get dropped”.

In Zwift, that generally means two things.

Tip #2: Go Hard To Start

First, go hard the first minute, and do NOT let gaps open up from the starting pen. Even though the course may not start for a bit, the race starts as soon as you are moving. There are so many riders who average higher than the lead group but because they didn’t go hard the first minute, they finish many minutes behind. You have to go hard at the start and stay in the front main group (let the one or two super-aggressive attackers go). After a minute, maybe at most two minutes for the lower level races, everyone eases up, and now the pace will become more real. I’m guessing you want to go 2x as hard as the advertised pace. D race? You need to do 3-4 w/kg for much of that first minute. Even if you think it’s murdering you, do it. If you can’t stay with the group for that first minute, I’d actually recommend quitting and joining the next race and trying again. It is absolutely that important, it will absolutely change your racing. You have to learn to go really hard and still be able to pedal after.

Tip #3: Push Over the Top

You have to keep going “through” the top of rises and hills. This is where the majority of people who shouldn’t get dropped end up getting dropped, just like the ones getting dropped at the start. These poor souls ease at the top of the hill or even a very short rise, let a 5-meter gap open, and it quickly becomes 10-15 meters and their race is over. If you climb “through” the top, and stay within 2-3 meters of the group, you can relax 15-20 seconds later when it flattens out or the downhill increases your speed. A good tip here is to surge as you hit any rise – if you’re doing 1.5 w/kg sitting in, surge to 2.5 as you hit the rise. you’ll see riders around you go from 1.5 to 3.5+ trying to hold position, then ease back a bit, the whole time you’re doing a steady 2.5.

(Note: if you get dropped during a hill then that’s just a climbing thing. If you get dropped on descents once you’re up to speed, it’s probably a weight thing. Both require being stronger, which is not really about tactics but more about fitness.)

Tip #4: Save It For When You’ll Need It

The thing with racing is that “real racing” is not about “working together” and “sharing the load.” It’s about breaking the other riders, cracking them, so that they are no longer in contention. Where and how you break the others is up to you. The basic premise though is that you need to save everything you have until someone tries to crack the group, then you need your reserves to outlast theirs.

It doesn’t do a racer any good (at least to benefit themselves) to pull any harder than “easy.” For me, that’s about 1-1.5 w/kg (80-120w). In more competitive races I might pull harder to help a teammate or force a momentarily distracted opponent to chase back on (almost always after cresting a rise, if I notice a competitor 5m off the back, I’m immediately surging to force them to chase an extra 10-15-30 seconds). In less competitive races I’ll pull harder to tire myself out. And to be fair, other times I’m going 1.0 w/kg because I got shelled and now I’m alone, I can’t do 4 w/kg for 5 or 10 minutes to catch, so my race is over.

Tip #5: Pull At 1/5 Your Jump

When the attacks come, you should be 100% ready to respond at max power. I tell my teammates they should pull at 1/5 of what they can jump, so if their jump is 800w, they shouldn’t pull harder than 160w. It’s reasonable for people to jump at 5x an easy effort. If your jump is 500w, don’t pull harder than 100w. If you can respond to two attacks in a row, even better, because the common tactic is to attack after someone else’s attack is neutralized.

Tip #6: Be Strong When It’s Hardest

When you pull at 1/5 your jump power you realize, wait, that’s really, really easy. Like really easy, like it’s not racing! What’s the point? Well, the point is that eventually, someone will attack at 5x their easy, and when they do, if you can’t respond, you’re going to be dropped, and now your race is over. If you’re racing to race, you must be strong at the hardest parts of the race, not the easiest.

As racers get more experienced, they’ll be able to judge a little better what they can do when “building reserves” and still respond to the big attacks. Maybe you can try going 1/4 your jump, and see what happens. Or 1/2 your jump. For me that’s a no-go, at best maybe 1/4 if I have to pull a minute or three, 1/3 if I’m following moves and can’t let a gap open. 1/2 means I’m desperately hanging on and hoping everyone else blows up before I do – I probably have a 20-second fuse at that power before I’m cooked.

Tip #7: Develop Your Peak

The other thing is to develop your peak power. Not everyone can break 10 w/kg, but 10 and 7 w/kg are realistic, attainable numbers by many men and women.

Tip #8: Training Isn’t Everything

Peak power is NOT all about training, it’s more about coordination, gearing, proper position on the bike (most Zwift videos I see have riders sitting far too upright to recruit their most powerful muscles), and, critically, being fresh. You cannot hit good peak power if you’re completely blown up or if you’re sitting upright, and the more you ride in your more powerful position, the more you’ll improve your useful peak power.

Tip #9: Ride Racier

On Zwift you don’t have to be low for aero like you do outside, but you’ll notice that you lean forward when working hard (like on a hill). That’s because you’re instinctively recruiting the muscles from the back part of your body, your glutes especially. If you’re not using these muscles, you’re probably leaving 30-50% (I’m guessing) of your peak power off the table. Learn to recruit these muscles by riding in a “racier” position regularly. If your glutes are sore after a ride, you have not been riding in that lower position consistently enough.