Have you ever wondered if reaching for the lockout lever is worthwhile? Sure, a firm lockout can make a bike
feel more efficient, but does this really translate into faster climbing? And if so, how much faster can you go for the same effort?
To find out, I've ridden up the same climb way too many times, using a power meter and a stopwatch to try and find some answers. To control variables and keep things repeatable enough to measure this in the real world, I picked a short road climb with 44m of elevation gain and a nice consistent gradient.
The short climb made it possible to do lots of timed runs so I could see if my times were consistently faster with the lockout on or off - if you only do a couple of timed runs you can never know if you would have gone faster anyway.
The climb's consistent gradient is important because changing gear or effort mid-climb can make it harder to stay consistent between runs. I picked a day with barely any wind and kept the same position so changes in aerodynamic drag didn't affect the results. I measured my time and average power output over a marked section of the climb, and aimed for smooth, efficient pedaling at a cadence of around 80 rpm, trying to keep as close as possible to a 315 watt power output. I did a total of twelve timed runs over two bikes, alternating between riding with the lockout on and off.
I rode eight runs on a
Privateer 161 (four with the lockout on and four with it off), then I did four more runs (two on, two off) on a Geometron G1.
The 161 has 161mm of rear wheel travel and 120% anti-squat at sag, making it an efficient peddler, but the RockShox SuperDeluxe shock has a light damping tune which allows some bob when open. The firm lockout virtually eliminates bob when closed, which should increase the difference I'm trying to measure. The Geometron G1 was set with 175mm of rear travel and an EXT Storia Lok shock. This bike has around 100% anti-squat, and bobs noticeably more than the Privateer when fully open.
Anti-squat values are calculated with an arbitrary center of mass height, so because I'm tall I need more than 100% anti-squat to minimize pedal bob.
It's not easy to hit the exact same power output every run; sometimes my average power was a few watts above or below the 315 watt target. Fortunately, it's easy to compensate for this using math. Climbing speed is, to a very good approximation, proportional to power output on a steep climb (I've checked this by riding the same climb at different power), so it's easy to work out a very good estimate of what the times would have been if each one was done with exactly 315 watt.
ResultsOn the Privateer, the average time of the four runs with the lockout off was 3:14.7, while the average of four runs with lockout on was 3:13.9. That works out to 0.4% quicker when locked-out. Meanwhile, the G1 was on average 0.8% faster with its lockout on.
Looking at all the results from both bikes together, there is a
statistically significant difference between the times with the lockout on vs off. In other words, I was
consistently faster with the lockout on, and it's unlikely that would happen by chance if the lockout wasn't having an effect.
So using a lockout is faster, but is it a big deal?Based on this, over an hour-long road climb you'd save 14.4 seconds by using the lockout on the Privateer and 28.8 seconds on the G1. That's not going to make a huge difference to most of us, but if you're at the sharp end of an XC-marathon race, or running late on an EWS liaison, that kind of time could be worth a lot to you.
Put another way, the time taken to dispatch a steep and simple climb is more-or-less proportional to the total weight of the bike plus rider; so if you have a combined weight of 100kg you'd need to save 400g from the Privateer, or 800g from the G1, to make the same difference. To me, this says more about the effectiveness of saving a few hundred grams than it does about the effectiveness of a lockout.
On the other hand, if it takes two seconds to lock and unlock your shock, you'll need to be climbing for about eight minutes before you recoup that time on the Privateer.
What's the bottom line?In a nutshell, the advantage offered by a lockout under these conditions is measurable but modest. It's safe to assume though that when riding out of the saddle, pedaling with poorer technique, or riding a bike with less anti-squat, it will be more pronounced - possibly a lot more. So the results here are probably the lower bound of the effectiveness of a lockout (on an enduro bike). Measuring the difference when out of the saddle might be trickier though, because it's harder to do consistently.
It's worth acknowledging that on a steep and smooth climb the main advantage of a lockout might be the steeper dynamic geometry and the more comfortable position this provides, so any increase in speed may be considered a bonus.
The effect of a lockout on a rough climb is a separate question. I'd be interested to see how rough the surface needs to be before the lockout is no longer an advantage, and if it makes climbing slower at some point. Perhaps I'll do a part two to find out.
Let us know in the comments if this is something you'd like us to investigate further.
1. Remember halfway up a hill to lock out the shocks.
2. Fumble around losing momentum, speed and line trying to get the switches flipped.
3. Get to the top of the hill.
4. Forget I locked out the shocks.
5. Get halfway down the hill wondering why my bike feels like crap.
6. Remember I locked the shocks.
7. Fumble around losing momentum, speed and flow trying to get the switches flipped.
8. Use above as an excuse to my friends as to why I was so slow.
9. Forget lockouts for about 6 weeks, then repeat above steps.
Dunno why you need that much time to open or close it.
I even done it on the bike for the older pro10 Propain shock placement. If you would hit a bumb while doing it you would lost a finger because the chainstay eat it.
Bikes I had before and now the 161 also are simple. Just move your hand from the bar under the top tube near the seat tube and your done. Maybe two seconds while your already going.
On the mega avalanche I found the remote was the most useful thing ever.
I did used different compression settings to save energy or generate more speed.
I would love them if they are not so prone to breaking/don't work and are a hassel to maintain. Never installed them on my 161 and won't do them of they are still that bad.
Instant compression and rebound changes are super nice if your on the run.
Dont care for xc though, only Enduro here...
I remember one of the glowing reviews the bike received at the time said:
"if you need to use the lockout on this thing you should consider trying mountain biking"
1: Probably use something like: en.wikipedia.org/wiki/Two-sample_hypothesis_testing (the t-test gives some guarantees of significance even though you might not have a full distribution).
2: You would have to measure people going downhill (forgetting) to turn their lockout off, or alternatively, simulate this hypothesis by having people leave their lock on/off and then justifying this is similar to forgetting the lockout.
3: Try to control the downhill, so that you have maybe one section that is bumpy, so that you can attribute the metric changes to that feature.
4: If you want to enumerate the situation where you are half-way down, and then remember, you can do that too but you would just have to add that condition to the experiment but would require the pair-t test to be done differently. You would have to decide which specific condition you wanted to test, otherwise you might run into issues around simpson's paradox.
His prices are totally reasonable and he does a great job. I’m about to restore my old F700. He’s rebuilt the 80mm headshok for me and he rebuilds my lefty Max 140mm for me every year. New bikes have their advantages to be sure (I have those too) but the older bikes are honestly a blast to ride. Just like an old MG or Triumph is a blast to drive. Nope they aren’t a McLaren but still so much fun. Honestly I’ve yet to find any telescoping fork that tracks as solidly as an old Headshok or Lefty.
gearjunkie.com/biking/hardtail-vs-full-suspension-mtb-mountain-bike
www.youtube.com/watch?v=s7qjODdT2v0
ojs.cvut.cz/ojs/index.php/ap/article/view/468/300
Hi, interesting test, would have been really nice to see a comparison with standing up. I get that you need to keep variables to a minimum but at the same time i feel that your efforts in doing so somewhat minimized potential resultat.
As a damper always works against a motion there has to be some movment in the linkage for it to steal some energy. So your efforts in keeping everything smooth and stable would have minimized the negative effect of the damper. So i think you could view your test as a good estimate of the minimal positive effect of a lockout.
Or you could view it as a meassurment of your personal ability to pedal smooth.
Anyway i feel you have gotten one end of the spectrum and by doing the same test again standing up you could get an estimate of the potential maximal positive effect of a lockout and catch the other end of the spectrum. Then we all just have to realise that the real world would most likley be some where in between depending on terrain, rider etc.
Your ability to keep a stable power and hit correct average power will probably decrease but as you said but you can easiliy compensate the average power. Also might be interesting to look at the average deviation of power to see if that correlates with efficiancy.
So please load up some more carbs and get climbing again, ill be waiting in my armchair.????
This would probably mean 1. Test on a longer trail with mixed conditions. 2. Pick some sections to stand and mash the pedals 3. Do waaaay more tests and use an xc bike.
Nice to have Seb here on Pb! Let’s get nerdy
The main problem with riding standing is (as mentioned elsewhere) that the vertical movement of the bottom bracket relative to the rider's mass can result in more physiological effort to provide the same power. In other words, the power meter would no longer be a good gauge of physical exertion. Heart rate is too inaccurate so you'd ideally need oxygen monitoring equipment. I might have a go anyway if this article does well to see if i can "feel" a difference in exertion when riding to a constant speed.
But that’s exactly the point of the standing test. Now that you have seated figures, a standing test would show how much work you’re doing against the damper (I.e. bouncing about and producing heat), vs how much against the hill (I.e. raising yourself and the bike up it); and, after turning the lockout on, how this ratio shifts when you’re being prevented from working the damper. I agree that this would be much more enlightening. Otherwise you’re really only testing whether the anti-squat is as effective as the climb switch or not.
Ran it 8 times, alternating locked (rear only) and unlocked. All results were within +/- 2 watts of 320, and +/- 2 seconds of 3:10. Unlocked averaged +1.25 watts higher and 1.5 seconds faster.
Obviously not super scientific, but conclusive enough for me to be comfortable forgoing a lockout for XC racing purposes.
I recently swapped the RS with an DHX2 and there it is way more noticeable when it is open.
I don't see why we should not use the leavers. I also used 3 POS remotes on races to change the compression on the fly to accommodate those parts where you need to pedal more and just not dive to much on stepper stuff / fast corners.
Have you test some coil overs for the forks or will you test them like the Vorsprung Smashpot?
I suspect most Scott riders rarely select the fully locked out position - it's that reduced-travel "traction" mode that makes the system so useful.
We are splitting hairs here, but at some point we need to look at the scientific approach that roadies use. (Try telling a TDF racer that he can lower his seat by 50mm and not notice a power to energy change.)
I think standing-up-pedalling might be a more interesting test.
Also nice work on recognising anti squat is affected by CoG.
Effectively, rear shock compression on a climb will have the opposite effect as the move toward steep seat angles, slammed forward saddles, and nose down saddle angles that have been becoming increasingly popular, due to their claimed benefits for in the saddle climbing.
It is also somewhat like going up a very steep hill, where it becomes progressively harder to pedal effectively as you feel like you are shifted into the "back seat" of the bike and you have to hunch further and further forward over the bars in order to maintain somewhat normal front to rear wheel weight balance. If you take it to an absurd level, just as a thought experiment, you could imagine being rotated so far that your saddle height would be the same, but the biomechanics would actually become those of a recumbent bike, which are totally different, and not noted for being advantageous on climbs (although they do have advantages on the flats).
You mentioned both biomechanics of pedaling and CG location, and I'm saying that for riding trails, one of those is marginal and one of those is huge. The marginal raw power gains of a more hips-over-BB pedal stroke is vastly outweighed by the fact that you don't have to fight the bike's tendency to lift the front and take away steering or potentially loop out, that you can just sit and pedal and deal with the terrain instead of dealing with both the bike and the terrain. With proper sized chainstays and a good seat tube angle, the "effective seated rear center" is plenty good enough on modern bikes, even with extra sag in the shock, for most riders to keep the power on.
You mentioned imagining a rotation of the whole system making it like a recumbent, except the bars and pedals don't move relative to the seat in that rotation, and you're still on a normal bike saddle, so it's not really at all like a recumbent, it's just a normal bike rotated. Recumbents are harder to put power down on because you're literally sitting back in a chair, isolating all of your upper body, most of your back, and a lot of your hips, meaning the vast majority of the power is coming from the legs only, and that causes the drop in power (literally less of the body involved), not just because your pedal stroke direction is less vertical relative to gravitational pull.
locking out makes the bike very rigid, usually more so than a hard tail or even a road bike - which can be quite uncomfortable over long hours of climb.
on a bike that isnt like that, lock out definitely makes a big difference. mainly you can put out more power. when its bobing the suspension absorbs part if the energy, so when the test tries to keep the same power output, its not testing that at all, its testing if the suspension delivers better grip, which.. it does. 0.4% on that amount if time is also within margins of error anyway.
* Fully open fork and shock
* Shock travel restricted down to like 70% and compression tune firmed a bit - this is great for pedalling up non-smooth climbs or riding on trails where you don't need all the travel, or on pump tracks even.
* Full lockout of fork and shock, really only use it for climbing on smooth surface out of saddle
And that's why the twinloc is good idea IMO, people can diss the extra cables all they want but it works.
The only real difference between the two has the way the remote lever operate and whatever tune you get with the bike, the rest is +- the same.
Most people have no idea that you can get these directly from fox, because no bike vendor sells that directly on the bike, except for scott, which have their own lever action for it. The scott remote is fine, but the fox is fine too. A few years ago the fox remote was significantly different from scott (and pretty bulky)
@seb-stott: I second morgiou; Could you please share the statistical tests and/or the data with us?
I realize I might be overthinking this; I also think that for this goal the sample size is just fine if you don't start adjusting for wattage.
Also accuracy of timing comes into scrutiny on such small margins.
I know that's not strong evidence by scientific standards but this is a PinkBike article not a PHD project.
This is a very unfashionable opinion, but I really miss the remote for the Climb Switch on the new Kitsuma shock too. Despite myself I ended up using that all the time on my enduro bike when I was running an Air IL on it. You could keep it firm for smoother section, then open it up for rougher sections of climb for traction just by moving your thumb. Fiddling around between your knees for the lockout switch is a PITA.
I think you already know what you have to do next @seb-stott
And after that, perhaps a head-to-head between the Cane Creek Climb Switch and whatever Fox call the Propedal lever these days.
And this is before considering unwanted geometry changes because the rear suspension sags further than with a lock out activated.
To quantify the losses you would have to measure the heat build up in the shock, but oh boy, think about all the variables. A click of rebound would throw everything off...
Perhaps easier to apply lactate testing on the top after each run to quantify the fatigue from each run? The watts alone will not give the whole truth
I want to know if it's EASIER - how much more energy am I wasting on the vertical bob from an open shock, how much difference in that do certain shocks have? etc.
Speed ain't everything unless I'm racing, which is rare, and I can't imagine many Pinkbikers are "...at the sharp end of an XC-marathon race, or running late on an EWS liaison...". (I'm sure the few who are will chime in claiming that they represent many hundreds of others ;-))
Maybe the end effect is ultimately as much a quantitative thing as qualitative, so its a moot point?
I remember seeing a video or interview a while back where an XC racer who said she chose between a hardtail and a full suspension bike based on how rough the climbs were, and not how rough the descents were. Obviously there is more time spent on climbs, but she clearly felt that having suspension working on the climbs to help with traction would be faster.
I haven't raced in years so I don't care about my times (well I sort of do, but not really), but I often end up on long sufferfest climbs and those are all mental. Flipping switches and the 450g saved by going carbon 'feel' like they are making a difference, therefore I'm choosing to ignore your conclusions and believe switch flipping makes a huge difference.
But really, I know the biggest actual difference I can make in the short term is choosing wisely for breakfast.
Because if the difference between open and full lockout is 0.6% then the difference between a descend mode and a firm climb mode, like my cane creek, must be very modest.
This test says all I need to know to comfortably leave my lockout open all the time.
I can only hope that most people also have access to and utilize trails with similar features. If someone does feel a _need_ to lockout that often (outside of a race), I feel a little sorry that you don't have cooler trails.
For enduros, speed on the transport climbs for me is dictated by how much energy I am trying to save, so keeping my HR in check rather than watts. This means speed varies a lot by weather, how far I am into the race, how many crashes I've had etc.
The big question is how to measure the effect of using lockouts during an XC race. Would like to see the data behind Specializeds Brain development.
Moxie Mx3 29er with a Mattoc 140, Hayes Dominions, xo1 11 speed, HT X2 custom long axle, Revive, Vittoria Mazza and Airliners. Especially the Airliners are such a blast on a hardtail, you can literally hit stuff as hard as you like without flatting or dinging, they dampen the ride too. I have run cushcore before and the protection was way less. Frame, fork and brakes are brilliant too, i am not seeing myself riding an FS outside the park again.
So fast and so much fun.
Thanks Seb! I've been saying this for years. This is exactly why bike weight matters a lot more than people think. For racers anyways.
That's the upper bound of the difference it can make, assuming a linear relationship. That would be the case only if all the work went into raising our potential energy. Accounting for tire rolling resistance, tire-dirt interface slippage, drivetrain friction, air resistance, etc., that 0.5 kg difference will increase the pace of a 100 kg system by less than 0.5%.
My real world experiences don't correlate so well with this though, perhaps because all of these studies and calculations are premised on equal power. I feel like out on the trail far more damage is done by the bike feeling rubbish to pedal - psychologically it's unrewarding to pedal heavy, squishy bike. You don't even feel like trying and so you perhaps don't attack climbs in the same way, preferring to sit back and wind your way up more gently. This perhaps leads to a far bigger difference in the real world, for most users.
Tire rolling resistance and poor ergonomics, for example, can be far more significant than a little bobbing from a generally efficient suspension design. There have also been some inefficient suspension designs that would benefit more from a lockout - or, better yet, a redesign to avoid the need for the lockout.
I also wonder if the metrics that we're using here - % difference and time in seconds have a tendency to sound insignificant - who cares about a few seconds unless you're racing? But actually, emotionally and psychologically a few seconds ahead can mean quite a gap - and your buddies pulling away from you does feel pretty terrible
Also, Seb, great article, thanks for putting this out there. Reading through the comment section there's a lot of people talking about the various things you could have done differently - but this is already great content and I appreciate you taking the time to create this!
Many relationships between variables are linear, but one that definitely does not respond linearly is our emotional state! Doesn't take much to demoralize someone who's oxygen deprived and drowning in lactic acid.
Regarding tires: Some elements of rolling resistance could be tested in a lab. One difference, though, is that mountain bike tires should be tested with a powered wheel, rather than a powered drum. All test equipment of which I'm aware use powered drums; this is adequate for road tires, where shear deformation of the casing and tread is minimal, but mountain tires - especially fatbike tires and low-rebound compounds - require a powered wheel.
www.bicyclerollingresistance.com
Maybe air resistance is a minor factor for you, but at the rate I climb ...
But seriously, it took me quite some time to spot the other non-linear factor. At low speeds, all sprung masses, including the rider, closely track the contours of the ground, i.e. the transmissibility ratio approaches 1 when the frequency approaches zero. As speed increases, the sprung(ish) wheel and fully sprung rider can "skim" the terrain, i.e. transmissibility ratio drops below 1 for each suspension system as the impact frequency of terrain roughness (not trail contours) exceeds √2× the natural frequency for each system.
Steve's Tuesday Tune #23 covers this succinctly near the beginning.
Assuming: hit size distribution, hit geometry, wheel radius, weight distribution. It's easy to then estimate roughly the energy losses due to the horizontal component of he hit.
Suspension locked you can take the potential energy to scale the kinetic energy losses
Suspension unlocked you need to estimate the dissipated energy (knowing the damping forces) and add a small amount of the potential energy depending on how much your suspensions are moving (50% to 80% of initial potential energy remaining)
You'll see that for a single hit, the difference is quite significant, but the hit distribution has quite a dramatic effect on the actual losses when locking the suspension (losses by hits being a fraction of total losses). Actually there is theoretical scenarii where locking the suspension is more beneficial as you can get back quicker some momentum before the next hit, and some where letting the suspension move is a no brainer. You'd need many different measures on different real tracks to have a useful measurement.
That's also my motivation for trying different suspension settings.
But most of all I want to know, which (Maxxis) treads/casings roll faster. There is nothing to find online.
So please make a rolling resistance test for enduro/dh-tires as minions, high rollers, assegais, some Schwalbes and other brands!
Measuring just power through the pedals does not factor in all variables. With the additional grip provided with the lock out off you may he able to apply more power for the same level of fatigue.
So please redo on a dirt road :p.
Do you also take into account the 200 grams or so that a shock remote would add? That would be half the difference already!
So when you are racing marathon it is very likely you are better of with the weight advantage!
To note, if it's just a "firm" lockout and not an actual "locked" lockout, you may be losing equivalent power to damping. Sure, you're bobbing less, but the force put into the damper is higher over that lessened travel. If it's truly "locked" then you're not losing anything to the damper.
How’s about those flashy new carbon high pivot bikes with 0% anti-squat, I’like a rock bar in the shock mount to pedal those for extended climbs...
Honestly, what I typically get from Pink Bike articles and the accompanying comments is just how varied terrain is, yet it seems the bulk of PB comments are always about riding up...then down. Don't people ride trails that do both...a lot? I guess if you're in Squamish you basically ride up, then down, but I don't think that's the reality for most riders. So that being said, I tend to choose between "open" and "trail" (or whatever the middle position is typically called) and never lockout. In fact, I don't even fully lockout on the steep road climb back to my house. I guess I'm too XC and not enough downcountry.
I'd like to see a timed comparison with like a Downcountry bike vs. a nice pedaling Enduro bike, up a more typical climb (for me anyways) on rough terrain.
First comparisons would be as the bikes come.
Second would run the same timed climbs but use the identical wheels/ tires on both bikes to find out how much of the climbing difference is just drivetrain induced drag.
Maybe a third test could be strapping the weight difference to the downtube of the DC bike.
Thanks for doing it.
For the average Joe / Josephine, 30s over a 1hr climb is negligible. Maybe it would be worthwhile for those racing, and this is a good followup:
"The effect of a lockout on a rough climb is a separate question. I'd be interested to see how rough the surface needs to be before the lockout is no longer an advantage, and if it makes climbing slower at some point. Perhaps I'll do a part two to find out."
Maybe MFG's will be able to adapt a lightweight automatic system that will handle lockouts.
For me, it's a no brainer in this sort of scenario however I understand that these days modern suspension design with anti bob and decent damping do mean lockout is less of a requirement. My last two bikes haven't even had lockout due to their modern design mitigating the need for it; the only time this bothered me was when standing up on the pedals uphill.
Maybe Levy was right all along
I have lockouts on other bikes and don't think I've ever used them. I'm not even sure if lockouts are supposed to "release" like that 4x shock does, but I like it.
(Though I'm a bit surprised there wasn't more to be gagging from the lockout, given the bikes you tested this on)
cyclingtips.com/2018/03/fast-chain-lube-that-saves-you-money
If you have a remote it's a bit less of a hassle, but having the extra weight and clutter on the handlebar is again not worth it to me.
And seeing Sebs results from running these bikes that should definitely gain from a remote I'm thinking my bike that has a rather pedaling friendly linkage has even less to gain.
YMMW
(If I was racing XC, hell yeah I'd want a handlebar remote for all those sprints, but I'm not so..
As someone that forgets to turn lockouts off, I now feel much better about leaving my shock fully open, all the time. Which is a nice outcome.
Thanks for doing these sorts of analyses, by the way. Too much of MTBing is based on hype, so some actual research is welcome.
I don't think anyone was ever debating if a lockout was worth it if you had road climbs in your future. The real question is whether it's worth it on a medium to long trail climb where traction might be at a premium and that 0.8% increase could be completely obliterated by a lack of traction at some points.
None of your contact points changed location: your seat is still at the same height relative to the BB, and the bars are still in the same place relative to both of those points. On a well designed modern geometry (steep seattube and decent chainstay length) you shouldn't even have to drop your chest and shoulders all the way down to the bars so even your torso can stay in relatively the same position on all but the steepest climbs (and then you're standing so all that positioning goes out the window anyway).
And I'm just saying that doing with a lockout/climb-switch is silly, because it costs traction. Doing it with geo (steeper seat-tube, appropriate length chainstays) is a much much much better option for bikes made to ride on trails.
It also depends a bunch on pedaling form and suspension design. Shit, on a high-pivot design (with an idler and dialed anti-squat hovering just under 100%), letting the suspension sag and thus lengthen the chainstays would be amazing for helping maintain normal/ideal riding position no matter how steep and just blast the power down. A lockout/climb-switch could be quite detrimental in this case, because it would keep the chainstays short _and_ reduce traction.
I’d be curious to see what the difference would be on an Ibis, Yeti, or something else that climbs better.
Literally a lever on the shock body
I see it the other way. There's only so much you can do to optimise climbing without compromising descending, and vice versa, so lockouts give the designer more freedom to prioritise descending.
And the lockout isn’t just a lever and cable. That cable attaches to either a complex frame lockout system (Cannondale, Scott, Canyon style) or some unneeded valving hard wear in the shock.
Just like 1x is better because it eliminates an entire mechanical system, ditching the lockout also gets rid of a bunch of unneeded stuff. Well, unneeded if your frame design is good.
2- I would like to see a test on an HT comparing the lock on the forks.
At this point, lockouts are just a holdover from when suspension sucked but sponsors made their XC racers ride f/s bikes.
When doe they put an accelerometer in the shock so it locks out automagically?
www.pinkbike.com/news/review-fox-live-valve-suspension.html
Great article! However, I’m pretty sure that’s a G1 in stead of a G16. What’s hour thought on the G1 vs Privateer 161?
Cheers
I use it for when I’m on smooth pavement.
For me suspension is more about reducing fatigue than going faster.
If you want to measure the benefit of a lock, you have to take the same climb, the same speed and then you measure the difference of energy (watts) it took you to make the climb with and without lock.
This is because you measure something that has a very similar absolute value, so linearity isnt an issue, and all other factors like temperature stay almost the same.
With power meters it is even worse - a slight temperature change will change values enough to erase a 0.4% differential. And you did not say anything about sample size and the human factors.
I would not dispute the conclusion of the article if the supposed advantage were larger. 0.4%? Please...
What a truly hideous thought.
For example not wearing that detachable bell helmet
"Siri, open shock"
I find this "testing against the clock" to be a scam.
I ride better not locked out and have more energy climbing since i do not feel every bump or non-smoothed out terrain.
I am also less exhausted climbing without a lockout.
my 3 cents