Tuesday was the third race of ZRL Season 5, but thanks to a broken first race and being out of town for the second race, this would be my season debut. And I was stoked, because although I knew the racing would be painful on NYC’s never-flat Park Perimeter Loop, I felt I had a good chance of hanging with the front pack to the finish while grabbing some sprint points along the way.
But bike races never seem to go the way we plan. And ironically enough, on the very day Zwift Insider published Luciano’s PowerDowns post, I would experience my own personal PowerDown. Let’s get to the story…
The Warmup
Since this was my first race of the season, on a course where I foresaw a decent result, I was careful to do everything I could to be prepped heading into the event:
- Recon race in the days before, just to refresh my knowledge of the route
- Easy(ish) effort the day before (rode with Boone on Climber’s Gambit, where he unlocked his Tron bike!)
- Plenty of sleep
- A clean, carby oatmeal breakfast
- Beet juice two hours before the race
- Caffeine gum – three pieces, one hour before the race. That’s 300mg of go-go juice coursing through my veins!
- PR lotion on my legs about 45 minutes before the race. More bicarb=less burning.
- A nice 30-minute warmup with the C. Cadence crew
Even though my lungs didn’t feel 100% (thanks, Omicron), the legs felt good and I was ready to race.
The Start
The start wasn’t too crazy – ZRL racers have gotten smart, with most learning there’s no point in attacking hard on a flat start when there’s a key climb just up the road.
That first time up Harlem Hill would see the first selection of the race, I was sure. So I tried to maintain a good drafting position a few bikes from the front for the slack initial section, then I hammered on the steeper finish, reaching the top with just a handful of riders ahead.
Making it over the first climb in the front group put me at ease a bit. In the recon race I’d done over the weekend (which was a large field with all categories combined) the pace had been crazy up Harlem Hill, and I got dropped on the second lap. But it looked like I might just survive in the front group of today’s race.
My plan was to go for FAL points on the first sprint, but riders attacked from way out, and when I started hammering, the top-end power just wasn’t there. So I sucked wheels and recovered. I’d be chasing finish points today.
The PowerDowns
I made it over lap two’s Harlem Hill in the front group, which had been whittled down to ~45 riders. Although I didn’t feel like I was on the rivet, my heart rate was higher than normal (another Omicron by-product?) so I told my team I’d be taking it easy on the sprint, just surfing the wheels to stay in the front pack.
Then disaster struck.
As we began climbing the short kicker into the sprint, I saw my avatar sliding backward quickly in the pack. I instinctively began hammering harder, thinking they were just jumping a bit faster than planned… but I still slipped from the front of the pack to the back in just a few seconds.
I got out of the saddle and started hammering for all I was worth to get back on.
Somehow I managed to catch back onto the pack just after we passed through the sprint banner. Then, a second surprise – my avatar kept flying through the pack, all the way to the front, even though I had eased off the power considerably!
Call it race brain or whatever you’d like, but at this point, I hadn’t realized anything was actually wrong. I figured I had just miss-timed the sprint, then chased back to the front of the group thanks to my excellent fitness.
But then the pack started flying past me again. And as I hammered the power, my watts were showing much lower than they should. Glancing at the numbers, it felt like the watts on screen were about 1/2 of what they should be!
The pack flew past. The unplanned efforts to chase back on had sapped my legs. I was dropped from the front group.
Hitting the “A” Brake
The same problem happened once more at the start of lap three as I revved up the watts to grab the wheels of two riders who were chasing the front group.
I’d had enough – I pulled the parachute and hit “A” to confirm that I was paired correctly. And I was – my Wahoo KICKR was paired via Bluetooth, as usual.
Since I was riding alone, with a sizeable group 30 seconds behind, I decided to pair my Assioma Duo pedals as my power source to see if it improved the situation. Then I took it easy and waited for the group to catch me. This group represented places ~40-60, and I sat in with them for the remaining two laps. My power numbers seemed just fine.
The Finish
The finish of this route is always painful, with its slight uphill start then a steep pitch to line. I’ve lost many races in the final 50 meters of NYC!
With an aero powerup in my pocket, I sat in the wheels for the flatter section, then activated the powerup and hammered as it got steep. I finished second in the pack, but a disappointing 35th overall.
See activity on Zwift.com >
See results on ZwiftPower >
See activity on Strava >
Watch my race recording
Final Result
In the end, the finishing positions of the Roosters were much worse than we had anticipated. But our saving grace (once again) was an abundance of sprint points. We took first place overall, despite my best efforts:
What Happened to My Power?
I went into investigation mode once I’d recovered from the race. Uploaded my dual recording to ZwiftPower (see it here), analyzed my log file using Zwiftalizer (no drops detected), watched my race video, and even reached out to Shane Miller and Wes Salmon for their input on what could have gone wrong.
Not knowing how Zwift processes power data programmatically, I’m can’t say exactly what went wrong. But looking at my dual recording data tells me why my power numbers seemed off. Note: my primary power source at the start of the race was my KICKR v5, and I was recording the power from my Assioma pedals to my Wahoo Bolt computer as the backup source.
Below you can see the key section of my dual recording chart from ZwiftPower (purple is the KICKR, blue is the Assiomas). You can see the power numbers are in sync with each other initially, but around 2350 the KICKR’s numbers begin lagging behind the Assiomas. And it gets worse as time goes on!
This time sync issue continues until things magically fix themselves around 2660, just before I pulled the plug and swapped my power source in Zwift to the Assiomas.
So there was some sort of power reading delay either on my KICKR’s side, or on Zwift’s side. That’s why, when I began hammering up that first pre-sprint climb, my numbers didn’t respond. The delay looks to be around 15 seconds, which is an eternity at a key moment in a Zwift race!
So when we hit that little climb, I started hammering, but Zwift didn’t “see” or process my increased wattage until ~15 seconds later. That was enough time for the group to drop me – but then when they eased up after the sprint, my sprint watts were finally being processed, which is why I zoomed to the front even though I wasn’t pushing the pedals at that point.
Then the pack started swarming past me again, and I started hammering… but my power was still delayed, so the hammering didn’t take effect until 15s later. After I’d been dropped a second time.
Delayed power is a really confusing thing. I don’t think I’ve ever experienced it before, and I definitely didn’t understand what was happening in the heat of the moment.
I’m curious: have any of you dealt with delayed power issues on Zwift? Comment below if so!
Takeaways
First: I’m still feeling some effects from having Covid two weeks ago. It felt like a very mild flu to me, but it seems to still be impacting my breathing and power somewhat. Not a huge amount, though, and I comfort myself with the knowledge that half the racers on Zwift are probably dealing with the same thing.
Second: never give up. Even when you feel like your race is lost, it’s not about you. Keep working to get as many points as possible for your team, no matter where you’re at in the race. Your team might just win, and if you gave it your all, you’re a part of that win.
Third: I need to figure out what happened with my power numbers. And that frustrates me, because I’m really at a loss as to what can be done to make sure this situation doesn’t repeat itself.
Your Comments
Any advice to help me fix laggy power? And how did your ZRL race go? Share your comments below!
I had this exact issue for two races at the start of last season. With the problem, I recieved zero points for FAL and FTS, even though the first FAL did not have the issue going in at that time and it also stood as the second top FTS for the night. I only received the one measly finishing point. The races were absolutely horrible. I was going back and forth through the pack thinking I was simply hammering too hard on the catch up and not easing off soon enough, to ohh my god, this group is trying to… Read more »
I ride a KICKR v5, which auto-calibrates.
I have had the same issue with the KICKR v5 as well. Very frustrating when you don’t expect to be launching through the group off the front. I have also had issues with it reading power way under what I was actually putting out. A firmware update seems to have improved that though.
Yes, I ride a v5 as well. It auto-calibrates within Wahoo, but it does not auto-calibrate within Zwift. You still need to perform the spin down.
I had the Wahoo firmware up to date and had Zwift up to date. The spin down in Zwift fixed the problem.
I am using the Kicker v5. It auto-calibrates within Wahoo, but it does not auto-calibrate within Zwift. You still need to perform the spin down within Zwift.
I’d be leery of using Zwift for any trainer calibrating. The last time I did that it ruined my old Cycleops Hammer.
That’s not true, Robert.
First: calibration is stored on the trainer, not the software. So once a trainer is calibrated, it’s calibrated across any software using it.
Secondly: the v5 auto-calibrates all the time. So you never need to worry about it.
I’ve got something like 100 dual power recordings with my Assioma pedals and the KICKR 5 which show very consistent power between the two (KICKR always 1-2% lower than the Assiomas in terms of average power).
Hello Eric, All I know is I spent a week talking back and forth with Zwift and WTRL to determine what the problem was. Even spoke with one of the Zwift software engineers on the phone. I suffered this problem over two WTRL race days, one Crit City race, and then an individual Road to Sky. Zwift did at one point to calibrate and I said, but the V5 does auto-calibration, so I didn’t do it initially thinking, how could that be the problem. The technology should work, right? In the end, I tried everything and the auto-calibration was the… Read more »
“Zwift said it is a problem they are working on, but they have no idea how to fix it.”
Gee, what a surprise!
I have had this twice before. When my Mac has played up twice. Dying with battery spare and then randomly starting back up. Whilst in the mean time connecting Zwift on my phone and then suddenly having Zwift run in 2 different devices. Didn’t knock the app on your phone or anything?
Another time it’s felt weird but not really the same problem, having my bolt on to to tell me my power numbers but being on the resistance page instead of auto kickr page. Wow that made my trainer difficulty shoot all over the place!
I had Companion open on my phone, but the phone was locked (screen blank). Zwift definitely wasn’t running on the phone…
I just had a normal power dropout (about the same place your lag started in lap 2). Fortunately, mine only lasted about a second. I was looking down at my cue sheet at the time, so I didn’t even see the dropout, just looked up and saw the front group go flying away from me as if I was standing still. I was very confused, but managed to sprint back on (burning some matches I didn’t have for the sprint or subsequent hills). I only saw the dropout later when I looked at the stream to try to figure out… Read more »
This happened to me a couple of months ago, almost exactly as you describe: the group began to move faster so I upped the power but no response on screen, then about 10-15 seconds later I get a surge on screen. And like you, I got dropped as a result. It carried on like this for the rest of the ride. I have never worked out what the problem was.
Well, it’s good to know I’m not alone! What trainer were you using?
Eric,
Same thing happened to me about 2 months ago. Happened two events in a row and after some research and digesting of info I switched from Bluetooth to ANT+ and haven’t had any connection issues since besides any unfortunate internet issues which unfortunately I dealt with in last nights race.
As for Covid I’m right with you, had it a month ago and still don’t have the sustained power of prior. Last night I held it together for the first lap but just couldn’t hold the wheels starting Harlem lap 2.
Interesting.
I’m loath to switch to ANT+ for my power connection, because I’ve had lots of ANT+ dropout issues since moving to this house we’re at. I assume it’s a neighbor’s wifi…
BLE, in contrast, has been super solid for me.
BLE was ok for me to begin with but I had two races where it failed utterly. I did find it a little more responsive than ANT+ when it was working.
Switched to ANT+ dongle with a long extension cable so it’s 5cm from the kickr and it’s never dropped out since. I guess a BT dongle (rather than internal) may be the best of both worlds ? Might have to try.
I was in the group you fell back to. Well done on the sprint.
I used to get power delays when using my laptop connecting via Ant+. Used to annoy me so much when doing group rides with intervals. I’d start sprinting and nothing happened for a few seconds, then the power would kick in and I’d sprint way beyond the banner and past the group!
In the end changed to an iPad over Bluetooth and no issues.
Interesting. So the question is: what is the PC to iPad change, or the ANT+ to BLE change that fixed it?
I used an Ant+ USB dongle plugged into the laptop.
The power delays were demoralising so I bought a USB extension cable so the Ant+ dongle would be closer to the trainer. No luck with that either.
This isn’t answering your question Eric.
Maybe I needed to pay more money for a better quality dongle, I don’t know. Was it the Ant+ connection on my Garmin watch causing the problems? I don’t know.
But now I’ve switched to Bluetooth its perfect!
Interesting, I will use my iPad today then!
Eric I assume nothing in your setup uses a battery? My Stages SB20/uses batteries for both pedal power meters and mine needed replacing just two days ago.
Also, I had really strange power like that once and the problem was that my bike software needed updating.
I always do a zero reset on my bike before I ride and find that this helps avoid weird power issues.
I’m also thinking excessive caffeine or beet juice(?)
I’m replying to myself because I thought of something else. It sounds like you connect directly from your KICKR to your computer. Pairing my bike directly to my laptop via Bluetooth wasn’t quick or easy. Shane Miller suggested using the companion app on your phone as a bridge to pair. I tried that and everything paired much quicker. Riding I noticed the power was more immediate and I could feel the road much better, every little rise and fall. You might try this too if you don’t already.
eric, are you on kickr v5 with latest firmware? many forum readers have had to downgrade firmware to avoid this out-of-sync bluetooth FTMS issue. i had exactly this issue and the downgrading of firmware fixed it.
Hey Dan,
I’m on 4.2.5… it’s my understanding from this thread (https://forums.zwift.com/t/zwift-wont-adjust-power-and-terrain-after-latest-wahoo-kickr-firmware-v4-21/534357/145) that those firmware issues were back in 4.2.2.
I thought about the firmware issue, but from what I’d seen, that just caused gradients to be delayed, right? Or did it cause power to actually be delayed in being sent to the game? Those are two very different things.
i actually had to think a lot harder about this than it appeared, but i think you are right — it’s a different problem. when i had delayed gradient, i was getting a similar effect to what you describe — on a hill, i’d suddenly start dropping back in the pack because i wasn’t feeling the gradient, only to feel it on the flat after the hill and suddenly push through the pack. in other words, i was putting the effort in at the wrong time. but looking at your graph, it’s clear that you put the effort in at… Read more »
this happened to me last year when i was out of town staying at a friends house. However it was a super old apple TV, and that’s what i blamed it on. not sure your situation is the same. First time comment, long time reader, keep up the good work.
My hunch is this is some sort of a Zwift bug, perhaps tied to PC/device performance, where it’s lagging the processing of the power for some reason. If that’s the case, your old AppleTV issue may be similar.
Off topic: How long after your Omicron infection did you do your first race?
The first day I noticed symptoms was a Tuesday, I think. And I had just that one day feeling obviously sick – a bit feverish, tired, sore. Then the next day I woke up feeling almost perfectly fine – just a slight cough.
First race back was on Saturday, and I did really well! But I kept it short – a Crit City 8-lapper, as I recall.
Wow, that was a fast recovery. Thanks for the answer. Good luck for the next races
Like I said – it felt like a VERY mild flu. But tested positive! I’m vaxxed (not boosted) which anecdotally seems to help… most non-vaxxed friends seem to have it dragging on 10-15 days….
Eric, I know it’s off topic, but you mentioned it. Being vaccinated doesn’t anecdotally help, it does help fact! That’s the point it significantly reduces chance of, hospitalisation, need for ICU admission and death. Sorry can’t help with power lag issues though. Keep up the great posts
no, it significantly does nothing. See Dr. Malone.
Jeff, would that be the Dr Malone, who’s view’s are highly discredited. Bared from twitter due to spreading disinformation. The international evidence is clear that vaccination reduces severity of disease, hospital admissions and icu admissions. In high vaccination areas like the UK, the link between case numbers and icu admissions has been broken, but not in the US where icu rates still very high, and vaccination rate relatively low. Because you or your friends had a mild illness (most people do) it doesn’t mean that’s the case for all. Not sure this is the right place for a vaccination debate… Read more »
me, wife, and a few friends all non vaxxed had omicron with very little symptoms…never even took a day off the bike. All tested postitve.
when i had it (first variant), it didnt affect me at all, HR still could max, breathing fine; didnt loose strength – but i only got headaches and minor phlegm
This happened to me during the Champs Elysee ZRL race several seasons ago. After seeing the pack start slipping away, I sprinted which shot me through like I was breaking away. My team thought I was trying to break away but my watts continued to show 600+ watts for 10 secs after I stopped pedaling. I worried I was cheating and tried unpairing and pairing again. However, the effect lasted through the whole race which was infuriating as I wasn’t sure what was happening in real time! I was on a Wahoo V5 and connected to Apple TV by bluetooth.… Read more »
I had major issues with this last week in London. It made me look really stupid, sprinting super hard after the first sprint banner. I struggled, surging back and forth, but after two laps it sorted itself out. This week, I lost the front group 3km from finnish because my watts suddenly froze at low numbers. At least I think they froze, but might have been delayed. I actually have dual recording from yesterday (just have never tried to upload it before), so will look into it.
Same thing happened to me a couple of weeks ago on my WTRL TTT race around Watopia’s Hilly Route, I was pedalling for all my worth at the start and the power came about 7 or 8 seconds later, was the same the whole race which made it impossible to judge and ruined it really as every gradient change I lost the draft or went flying past my team mates. I’ve never had this before and am now paranoid it’ll happen again when I’m racing. I run a Saris/Hammer H2, latest firmware, Zwift running on my Win10 PC updated with… Read more »
Gah. Yeah, I feel your paranoia!
I haven’t seen this issue, but I do have an unrelated question—Why do you wear gloves indoors? Don’t they get hot?
I hate the feeling of my sweaty hands on my bars… gloves just feel much better to me. Maybe because that’s what I’m used to outdoors? But no, my hands never feel hot.
I have the same issue, but i prefer a towel over my bars, that way i can adjust it to find dry spot once it gets too wet where my hands are, and it catches a lot of the sweat that would drip onto your handlebar area. it may seem in the way at first but i find it awkward now without one.
I’ve done the towel on the bars thing when I don’t have gloves, but I don’t like how the towel blocks the flow of air from my fan, which is on the ground just ahead of my front wheel.
Had exactly the same problem with the kickr bike in last nights race
Have you considered the issue was at a PC IO resource level. There was a delay in the input from the Kickr being processed into a data record as it was dealing with something else at the same time. It would explain why it got worst over time and nearly back in synch when the power went to zero. I would investigate if the PC’s resources were maxed out. Anti virus/firewall update, windows update, one drive sync etc. I had the same problem as Dan with the gradient delay. It caused the same effect as yoyoing through the pack. I… Read more »
I had similar issue with my Kickr and I believe it was just the no name USB Ant+ dongle causing the issue…… I replaced it with the more expensive Garmin Ant+ that has the extension to get it closer to my trainer as well.
I’ve yet to have delayed power readings but I was getting delayed trainer resistance when simulating grade with my KICKR V5. I switched from Bluetooth to ANT+ and haven’t had any issues since.
I experienced the same problem yesterday starting the first Harlem Hill until the finish. At one point I could only hold 80 watt while everybody was flying by. I completely “destroyed” my legs to keep up and got so frustrated. I’m using a Wahoo Kickr Core with firmware version 1.1.3 and the Wahoo is connected to the notebook with a Tacx ANT+ dongle. I did a Wahoo spindown right before that start. Zwiftanalyser didn’t show any dropouts. I didn’t think about connecting my powermeter as a second source, but I’ll definitely do that next time (what will hopefully be never).… Read more »
I had the same issue with the delay, riding a Zycle Smart ZBike.
I get this from time to time, maybe once every 20 rides or so? I don’t think it’s a trainer issue as I’ve had it with my Vector 3 pedals, Powertap pedals, Kickr4(? second newest model anyway) and Elite Suito. Also had it on iMac Pro, Apple TV, PC, Ipad Pro, so probably not platform specific. I would think either BT or Zwift internal bug.
I don’t race, so it’s only mildly annoying and I don’t mind. (Though I probably look really dumb climbing like mad on the downhills…)
This happens to me a lot in races. Keeping the same pace as the group, keeping the watts steady, and next thing I know I’m falling off the back and having to push extra hard to get back on. I’m only using my v5 Kickr for power, no dual recording, and my laptop is Ethernet connected, so there shouldn’t be any lag issues. But with Zwift, who the hell know what’s going on. They have virtual drivetrains slowing down bikes, so it’s anybody’s guess what unknown bugs and gremlins could be lurking and complicating things.
I had the same issue, with far fewer consequences, on a recent ride with the C pace group. Power dropped off inexplicably on my avatar even though I was at steady watts, had to hammer to catch back on, then flew through the group. I have a Wahoo v2 that has been calibrated and was Zwifting through my iMac. This has got to be a Zwift bug.
This is what happens when a founder stops paying attention to their core business. New developers come in and don’t understand all of the code, so they introduce weird bugs as they push up new features.
Hi Eric. Francois here from South Africa. I had the Kickr Snap up to the end of December 21 but then decided to move to direct drive and got the Tacx Flux S. Started using it the first time this year. My power drops have been so severe and consistent that I’ve considered taking the trainer back to Garmin. Feels like I’m ripping my legs and the power reading is the probably 60% of what I think it should be. Calibrating on Zwift also doesn’t solve it. Sometimes when calibrating my avatar shoots on holding a fixed power for probably… Read more »
I was also coming back from covid in this race. I did a test ride on sunday and realized my fitness felt ok but my burst power and recovery was very much not there.
So what I did was that I set a watt ceiling for my self to and made sure I didnt go over it. That obviously got me dropped in the Harlem Hill but otherwise I did feel quite ok.
Post COVID racing is horrible. Had issues myself this week, although I was last 77th place , my race was recorded by Zwift and I didn’t get my 4 points. For some reason I lost connection at 38.7km, but when it reconnected I was at 42km and told I had 1 more lap to go. Any ideas how I can claim my points?