As a general rule, once you’re dropped in a Zwift race, you’ve lost. And if you’re dropped on a double draft race, then you’ve really lost!

But that isn’t always the case; it depends on how the race is set up in terms of categories and timing. I recently hopped into a Critty Critty Bang Bang race over 14 laps of Crit City’s Downtown Dolphin and had a blast pushing hard, getting dropped, and winning! Here’s how it happened…

Warmup

I began my race prep by chewing two pieces of caffeine gum and rubbing some PR lotion on my legs. Then I hopped onto the Richmond course for a ~20 minute warmup, because I hadn’t ridden Richmond in a while. I’ve always loved this race course.

My Wahoo TICKR version 2 wasn’t behaving – it has been reading low or erratically lately, even with a new battery. So I grabbed my 4iiii Viiiiva instead and headed to the start pens. Let’s race!

The Start

I knew going into this race that it was a mixed-category affair. But there were only a handful of A riders signed up, so I figured the pace wouldn’t be unbearable.

We only had 20 B riders in the pens and 70 riders total, which is wild considering the last time I did this exact race we had 402 total riders participating! It’s definitely outdoor season for much of the cycling world.

The race started unremarkably, with the typical hard first lap. It wasn’t too tough, though – I averaged 324 watts in order to hang with the front. We were quickly shedding riders, and by the time we neared the end of lap 3 we had a front group of ~20.

The zPower Pull

One A rider in particular was pulling the group around – and of course, “N Bergamini” was on zPower. I mentioned it early on when he kept riding off the front, but the mixed category setup means the A’s probably won’t let a flyer go, because his power isn’t too outlandish compared to theirs – and it’s easy to miss that he’s on zPower, since the only indication is the tiny missing lightning bolt in the rider list.

ZPower wasn’t allowed for the race anyway, so everyone could have let Bergamini go and it wouldn’t have impacted the final results. But that only works if everyone lets him go – and that wasn’t happening. So we all held on tight and kept pushing!

Bergamini would end up finishing 3rd according to Zwift, averaging 5.5 w/kg – a full 1.1 w/kg higher than his closest competitor. He would be DQ’d from the official ZwiftPower results.

Getting Dropped

Hanging with the 7 A riders wasn’t easy, but I wasn’t on the rivet either. We made good time zipping around the circuit, and I used a powerup to maximum benefit every chance I got!

This was a double draft event, and the pace was pretty high thanks to the A riders (especially Bergamini). I didn’t feel I had enough extra wattage in my legs to attack, and with the increased pack speeds from double draft I didn’t fancy my chances of staying away even if I got off the front with a couple riders. So I tried to sit in and conserve as much as possible.

But I was tailgunning a bit too much, and somewhere around the beginning of lap 11 a gap formed up ahead and I missed the break. I’m still not sure what happened (I wish Zwift had a way to replay events) but I think our group was lapping another group, things got messy, and before I realized it the fast guys were in a breakaway.

It’s important not to panic here, though. The big question is, “Who is in the breakaway?” I’m racing against the B’s, and there were only 5 of us left in the lead group. Did any B’s make it into the breakaway? I started looking around and counting, then breathed a sigh of relief. No B’s in the break – I still had a chance to win this!

The Finish

With Bergamini and the strongest A’s off the front, our group’s pace dropped – which was fine by me. (Average wattage for lap 12: 271. Average wattage on the lap before the breakaway: 308!)

This let my legs recover a bit. I crossed my fingers, hoping the Zwift gods would smile upon me with a useful sprint powerup for the finale. But alas, I was gifted a feather! I used it on the final prime brick climb to give myself a bit more rest, then sat near the front of the group, ready to jump and follow the wheel of anyone foolish enough to attack early. Nobody was.

As we neared the bottom of the twisty descent I went all-in from the middle of our group, using the extra-strong draft to slingshot me into the empty, painful chaos that is the front of any Zwift sprint finish. It was another seated sprint, which I’ve been using to good effect lately.

As I neared the finish line I glanced at the screen and saw my competitors 5, then 10, then 15 meters back. They weren’t gaining! I crossed the line in first for B’s. My first back to back race win on Zwift… ever! (Read about last week’s win here.)

See activity on Zwift >
See activity on Strava >
See results on ZwiftPower >

Takeaways

With field sizes down significantly from a few months ago, it’s much more possible to place well in Zwift races these days. That’s fine with me – I prefer to race in fields of 40-80 rather than 200+ anyway. For me, at least, the race is more interesting when you can get to know the individual riders instead of feeling like it’s “me vs the blob”.

So while I’m typically not a fan of mixed-category races, at this time of the year I’m finding them much more tolerable as they give me a chance to mix it up with the A’s without getting shelled within the first few minutes.

My lesson for today is one I should have already learned – tailgunning is dangerous. Just when you think you’re safe hanging onto the back of the group, a gap will form and you’ll get dropped! It’s safer to ride up near the front, but still out of the wind. If I had been positioned correctly I think I could have gone with the breakaway – then I would have been the only B in the group. That would have been sweet!

The other lesson is that getting dropped isn’t always the end. I’ve been in mixed races before where just one or two B’s got into the winning break, which left a podium spot in my chase group. Don’t give up! And pay attention, so you know if the group ahead contains your competition (meaning you may want to chase that group) or not (meaning you can sit in). Getting dropped can actually help you, as it did me when it reduced the group’s pace so I could recover a bit and be ready for a big final sprint.

Your Thoughts

Have you noticed smaller field sizes on Zwift lately? What sort of field size do you prefer? Chime in below!

I had been riding off and on since I had kids about 8 years ago. Two years ago, I found out about Zwift and was immediately hooked. I was determined to be the top Category A+ racer. I was a Division I runner in college and I figured “how hard can it be to beat these people?” Of course I am old, I have two kids and a full time job and I really wasn’t a very good Division I runner, so I quickly pivoted to “What does it take to be the top CATEGORY B racer?”. Below is what I found after cranking through the data.

If you don’t know about ZwiftPower and how they rank races, check out the top 20 riders in Cat B, which I used as a basis for analysis. (Only race data considered; group rides and tours were excluded in the analysis.)

After analyzing the data, it is clear that if you’re looking to move into the top positions, you should try to emulate the below:

• Race Experience: These people race a lot and get very good at it. Zwift is a skill. It also gives them more chances to pray to the powerup gods.
• Pretty High 15-Second Sprint Power: 10+ W/kg
• Very High 1-Minute Power: 6.5+ W/kg (most likely the 45 seconds before the sprint, then 10+ W/kg at the end; you have to be able to get into a position to let your 10+ W/kg sprint rip)
• Top of Category 20-Minute Power: 3.7 to 4.9 W/kg; overall average 4.1 W/kg
• Be Light: ZwiftPower states that to move to Cat A you have to have three races/group rides above 4.0 W/kg 20-Minute Power and your power from those rides must be above 250 Watts. So if you are light you can exceed 4.0 W/kg and remain in Cat B.
• Race in Category A: Higher Category = lower points

If you’re curious about the specific data I used, check out all the data details at the end of the article. Otherwise, let’s jump into the number details.

The Numbers Breakdown

So let’s start here… the data table:

20-Minute Power

This one isn’t a mystery. You must have a high power to weight ratio. All of these riders except one are capable of riding 45-60 minute races at 4.0 W/kg or above (except that pesky Rider 11). They may not do it every race, but they can, and I am guessing they are all experts at sitting in the draft.

Six of the 20 riders have W/kg that averaged over 4.0 W/kg. That is because they are small enough not to trigger the category upgrade where you need to be above a 250 W FTP. 

5-Minute Power

You need to be able to hang with the group up the hills. About 4.7 W/kg will do the trick on average. I did some spot checks on races and that is about right, except for the light people who average higher. Hopefully, I will be able to figure out a better way to get actual data for individual races and I can expand on this.

1-Minute Power

I think this is the most important part of placing highly. As you can see in the table above, you have to be very good at 1-Minute power. If you can’t average 6.5 – 7.0 W/kg across the final minute of the race you won’t even be in a position to use your sprint. This is skewed up because it likely includes the 15-second sprint power – to correct for this I extrapolated 45-Second power in the table.

Or more succinctly: you have to sprint for 45 seconds at the 45-second power and then 15 seconds at the 15-second power. 

15-Second Power

My initial hypothesis was that the person with the max 15-second power was going to win the race. This is partially true, but it seems like a combination of 1-minute and 15-second is the key. You have to be good at both. As you can see in the table, you need to be able to average at least 10 W/kg, and realistically 11 W/kg to get into this upper echelon of B category racers. If you want to be the top B racer you need 12 W/kg to hang with the sprint in the A category where you need to race to get a low enough ranking.

Races Per Week

If you want to get a good rank you have to race a lot. Don’t underestimate the power of learning how to race well and the luck of needing an aero to place high enough to get that race ranking.

Podium Percentage

This is how often a racer is in the top three of their race. Lots of these people race in the A category so their ranking gets better while still classified as a B rider. In general, the lower podium percentage racers here usually race in Cat A, and the higher percentages race more in B. I was surprised the percentages were not higher – it might go to show how hard it is to get in position and have the right powerup for a Zwift sprint.

Other Tidbits

• The racers seem to be clustered around 50 and 75 kg. 
  • The lighter side makes sense – they can exceed Cat B W/kg with an FTP right below 250. They must be just heavy enough where the flats aren’t too bad and the sprints are fast too. It also might just be the lower limit of what men (I believe these are all men) weigh – I found a calculator that said 50 kg was in the 1st percentile of 30-year-old men.
  • I’m not sure why 75 kg. I’m guessing it provides enough strength for a good sprint with a low enough weight to keep a reasonable FTP for normal people. This might also just be the weight most men end up around when they are biking a lot.
• 11 of the 20 riders race primarily in Cat A, 4 of the riders race in a combo of A and B, and the other 5 riders primarily race B. 
• 5 of the 20 riders have fake names. 
• 1 of the riders (Tom Gakes) looks like he used to be a pro and is a verified rider on ZwiftPower.
• 3 of the 20 riders are under 20 years old, 2 are between 23-29, 11 are between 30-39, 2 are between 40-49. 1 person is not registered at ZwiftPower (I don’t get why the person is in there). 1 person mysteriously disappeared from ZwiftPower since I ran the data and 3 days later when I looked up the ages.
• 6 of the people are from Europe, 5 are from North America, 8 are from Asia and one has vanished into the ether. Japan and the US are tied with 4 each.
• A huge portion of the races these people do are short and flat like Crits. 

The Data Details

• Data Source: ZwiftPower
• Data Curation: Manually downloaded. I would love to pull more data and do other analyses (see other articles I’m planning at the end) but I haven’t found an API to download all of the data from ZwiftPower. (Guys who run ZwiftPower – hit me up.)
• Data Filter(s): Past 90 days of racing from the Top 20 category B racers as of July 26th. When you look at the rider profiles you will see higher power than what is shown below because the data I used is only from races where the rider achieved a rank of under 250 for the race ranking. Our goal is to identify how to become a top Category B racer in Zwift, not to see what everyone’s max wattage is during a 2.0 W/kg group ride with a sprint in the middle. Also, cutting off at 250 is arbitrary. Looking at the best five races from some of the riders between 15-20th it looked like a good cutoff. 
• Analysis Tool(s): Google Sheets and Excel. If I can get more data I will do better analysis in R, but this data is easy.
• Additional Notes: Rider 11 seems to be an anomaly. That person has consistently lower power across all categories. 

Let Me Know What Else You’d Like to Know About

Below are additional data-focused articles I plan to write, also based on data from ZwiftPower:

• Cat A+, A, C, D (Next in Line)
• Women Specific article – I’m not sure of format and Categories
• What it takes to win a Zwift race
• How to win on each course
• Zwift Rankings and Category Improvements

Let me know if this has helped you and what you want to hear about next. Just keep in mind that I cry easily and if I don’t see any comments or votes below that my kids will ask me why I am sobbing at my computer looking at Zwift Insider and I will have to explain how cruel a place the world is. So please just comment so I don’t have to burst my kids’ bubbles at 6 and 8 years old.

One thing I really enjoy about Zwift racing is the ability to try new tactics without risking much. Whereas an outdoor race might cost me a $40 entry fee plus several hours of driving (and associated costs), racing on Zwift costs me almost nothing at all. The time commitment is minimal, there is no risk factor, and there’s no entry fee!

Sometimes I come up with hair-brained race tactics beforehand – I’ve written about plenty of those. Other times, the ideas strike mid-race, like some sort of divine lightning bolt. Today’s race was the latter. And the best part is: it worked! Here’s the story…

Warmup

I grabbed just one piece of caffeine gum and rubbed some PR lotion on my legs. Then it was off to Watopia for a ~45 minute warmup before my race at 6:35am. This time around I hopped into a group ride led by my buddy Peter McKenzie because the stated pace (2.7-3 w/kg) was just right for a warmup, once I threw in a sprint or two.

After a nice warmup spin with Pete and crew, I headed to the start pens. Eight laps of “Bell Lap“, which is actually the direction I prefer on Crit City. Let’s race!

The Start

We had a mid-sized group with 42 starters in B cat. As usual, the start was hard and fast, and the selection was quickly made – halfway into the first lap a group of ~20 was off the front.

As we began the second lap, I noticed the rider on the front our group, one z-powered “J Kippen”, was holding steady between 5.5-6.5 w/kg. Some riders were struggling to hold his wheel as he kept pushing unrealistic power numbers, so I messaged the group: “Kippen is a flier. Let him go?”

The group seemed to respond immediately, letting Kippen drift off the front. Brilliant! I felt bad for the riders who had already been dropped off the front group thanks to Kippen’s Z-Power Ranger pull, but what do you do? That’s Zwift racing… for now, at least.

The Middle

We had 18 in our front group, and the pace was pretty manageable. So I started putting in attacks with the help of whatever powerup I received each lap, just testing to see if others would go with me, or if I could thin the herd a bit.

As usual, nobody went with me. And I didn’t see a single rider drop from the pack either! And that’s when the idea struck, like a bolt of lightning sent from the Zwift gods: what if I announced my attack?

Crazy, right? But follow me for a minute.

“Anyone Want to Attack with Me?”

My experience in over 170 Zwift races has taught me that solo attacks rarely have much effect, at least until the final minutes of the race. This isn’t the case outdoors, so what’s different on Zwift? I have some theories:

1. It’s hard to spot an attack on Zwift: compared to outdoor races, it’s harder to see when someone begins an attack on Zwift. Unless you’re looking closely at the right-hand list of power numbers, you won’t notice a rider attacking until they’ve drifted off the front – unless they’re foolish enough to really power up so their avatar changes to the sprinting position. Because attacks are hard to spot, your response to the attack gets delayed, so by the time you notice, it’s just not worth the effort to put in the dig to close the gap to the attacker. So you sit in the pack, knowing the lone attacker will get reeled in soon enough.
2. High pack speeds: pack speeds on Zwift are higher than outdoors, especially in double draft events. This makes it even harder for a single attack to stay away from the group.
3. We’ve been conditioned: after experiencing the lone attacker being unable to stay off the front, Zwifters have been conditioned to not respond to attacks, and not initiate them, either.

Whatever the causes, I think we can all agree that attacks are most fruitful on Zwift (and outdoors) if you can get at least a few riders to join you. But how do you do that, if you don’t have team members in the race on Discord?

What the heck. I’ll just announce my attack, and see if others want to follow.

Near the end of lap 5, I messaged the group: “Anyone want to attack with me?” Nobody responded for a bit, until one rider wrote “What’s the signal?” I didn’t have a signal, but I had a feather, and the twisty climb was coming up. So as soon as we hit it, I just wrote “Now”, activated my powerup, and hit the power hard!

A gap quickly formed several riders back. I messaged the group “PUSH PUSH” and we kept the power up for another minute or so, watching the gap grow from 3 to 5 to 8 seconds. We’d done it! We now had a front group of 8, and the chasers weren’t coming back. I couldn’t help but smile, even as I suffered. Type II fun, right?

The Finish

With two laps left, our breakaway group was firmly established. We locked into a steady pace, and I held onto my aero powerup for the final sprint. No one attacked in any significant way on the final laps, so we moved into the final meters as a unit.

I activated my aero powerup just after the last left-hand turn, ramping up my power in a (rare for me) seated sprint.

I’ve been testing the Axxion Rocker Plate from Traxxion Dynamics, and I’m finding that my seated sprints are really strong due to its design/setup. My power numbers were really good for this sprint, averaging 876w for the final 13 seconds. I even had enough of a lead on the group to snap a couple shots!

See activity on Zwift >
See activity on Strava >
See results on ZwiftPower >

Takeaways

My simple takeaway for this race is announcing your attacks may be a good idea. Certainly not something I would do outside, but on Zwift, I think there’s a case to be made for it, at least in certain circumstances.

I’d love it if some of you could try announcing an attack in an upcoming race, then comment below to let me know how it went.

Your Thoughts

Is “telegraphing your pass” a good idea in Zwift racing? Why or why not? Share your thoughts below!