# Enginerding: What Is Anti-Squat & How Does It Actually Affect Mountain Bike Performance?

Feb 14, 2020 at 3:16

Words: Dan Roberts
Illustrations: Taj Mihelich

Anti-squat has become common term in discussions about bike design. But what does it mean? As simply as possible, it’s how your bike’s suspension reacts when you accelerate. Although there are other factors at play as well, anti-squat most often gets oversimplified to pedalling performance. While it's a big factor there, we need to start off with some other terms and build our understanding to properly understand its effects on bikes.

Riding bikes is a hugely dynamic process, so to analyse what’s going on we need to strip real life back to two dimensions. That way we can understand what is happening and make some calculations before we start to, bit by bit, add in real world elements and scenarios to arrive back at real life with an analysis that matches it as closely as possible.

First, we start with some definitions of mass and weight. As a rider we have a certain mass, measured in kg. As most of us live on Earth, we are subject to gravity accelerating our mass towards the center of the Earth. This accelerating of our mass exerts a force on the ground where we contact it. This is our weight, measured in Newtons.

The mass of the rider and bike is accelerating towards the center of the Earth creating the weight force. This is then split between the two contact patches creating a load at each. There is an equal and opposite reaction force pushing back up against this downwards load.

For the analysis of anti-squat, we need to use our center of gravity. Every particle of the bike and rider is being attracted to the center of the Earth due to gravity, but to simplify it for calculation we concentrate all these forces to one singular point; the center of gravity or CoG. As a bike can’t ride itself and it is similarly not as much fun to run down a trail the center of gravity needs to be that of the bike and rider combined.

This combined weight of the rider and bike creates a load acting at the ground that is split between the two contact patches. Neither acceleration nor braking can cause our actual weight to transfer. During acceleration or deceleration, it’s a load transfer that happens. Our weight can remain in a constant position while the load is transferring underneath us.

The CoG of a giraffe demands a completely different suspension design to allow it to have an efficient pedalling bike, amongst other considerations.

When you accelerate the amount of load transferred to the rear contact patch must be balanced with a reduction in load at the front. The weight always remains the same, just that if you somehow manage to wheelie your bike, all the weight is being supported by the rear contact patch.

In mountain biking the magnitude of this load transfer isn’t to the same degree as in cars or motorbikes. We simply can’t achieve the same acceleration forces, no matter how big your thighs are. However, it’s still hugely important in the system.

This load transfer happens in all bikes, no matter if you have suspension or not. But it’s with the inclusion of suspension, especially rear suspension, that we encounter its tangible effects.

With constant acceleration of the bike and rider forwards the CoG of the system is accelerated in the opposite direction and creates a moment due to the CoG height above the ground. This moment causes an increase in load at the rear accompanied by a decrease in load at the front. The weight has remained in the same place the whole time, but the load has transferred.

Under constant acceleration, the load is transferred to the rear contact patch, and in turn load is reduced at the front contact patch. This acceleration causes a rearward rotation of the bikes sprung mass and un-attended to will cause the front suspension to rise and the rear to squat.

Anti-squat therefore seeks to counter this suspension compression from load transfer. It is simply an expression of how the suspension responds to driving forces, and is expressed as a percentage.

To illustrate anti-squat and make sense of these percentages we can draw the graphical method for calculating anti-squat.

Step 1: Draw in the ground between the two contact patches.
Step 2: Draw a line up from the front contact patch through the front axle, perpendicular to the ground.
Step 3: Determine the CoG position and get a measurement for its height above the ground.
Step 4: Find the instant center of your bike.
Step 5: Draw a line from your rear axle to your IC.
Step 6: Where this line and the line of your chain cross is called the instantaneous force center or IFC.
Step 7: Draw a line from your rear contact patch through this IFC and cross over the front contact patch line. This is called the anti-squat force line.
Step 8: Measure the height, from the ground, that this line crosses the front contact patch line.
Step 9: Compare this measurement to the CoG height measurement in terms of a percentage.

Seeing as the load transfer is trying to rotate the sprung mass of the bike and rider, if your anti-squat percentage is 100% then the anti-squat forces exactly balance out the tendency of the suspension to squat under the load transfer from acceleration and the net result is that there is no suspension compression or extension.

If the anti-squat force line crosses the front axle line at the ground, we can call this 0%, as none of the tendency to squat is counteracted and the suspension compresses solely due to the load transfer.

Crossing the front contact patch line half way between 100% and 0% results in 50% anti-squat and half of the force needed to combat the squat tendencies is combatted. The net result being some suspension compression, but not as much as if we had 0%.

Going under 0% the anti-squat forces actually work with the suspension compression caused by load transfer and put us in a pro-squat zone. In this region the suspension will compress further than at 0% with the help of the pro-squat forces.

Above our 100% mark, where we’ve effectively separated the drive forces from the suspension response, the anti-squat forces will actually be more than enough to combat the squat tendencies, and the rear suspension will extend with that excess force.

The percentage of anti-squat is all in relation to the CoG. If the anti-squat force line and the CoG cross the front axle at the same height it's 100%. If the anti-squat force line crosses at the ground, it's 0%. You can then go below or above these limits with negative percentage values or above 100% values respectively.

The percentage of anti-squat tells us how much of the squat force, caused by the load transfer from acceleration, is counteracted. 100% means the two forces are equal and the net result is no suspension compression. 0% on the other hand, means none of the squat force is counteracted and the net result is the suspension compresses.

Anti-squat isn’t a fixed figure for a bike, it changes as the bike goes through its suspension travel. What we can do is draw out the graphical calculation at incremental steps in the travel until we have a curve that shows us what’s going on from zero to full travel.

Calculating the anti-squat percentage for multiple steps in the travel of the bike gives us an anti-squat curve. You can then see how the anti-squat changes as a function of rear wheel travel. Some anti-squat curves are a straight line, others have far more curvature to them.

When you pedal a bike, you’re investing energy in the system to move you forwards. Some of that energy will always be transferred to somewhere that isn’t going to help the forwards movement, like friction in your drivetrain. But ideally, you’d want as much of that effort you put in to be getting you up the hill. A system that doesn’t combat any of the squat from the load transfer is going to result in unwanted suspension compression and you bobbing your way up a climb.

Anti-squat is a good thing then, to combat the natural want of the bike to squat and compress the suspension. While the acceleration of your bike might not change with or without anti-squat, if you have some combating then you’re more likely to keep your CoG in a constant position, maintain a more favourable bike geometry and keep the load at your rear contact patch at a more constant level.

With anti-squat being good, too much of a good thing can be bad. The anti-squat needs to be within a good range and applicable to the bike’s intentions. Designers can adjust the anti-squat with pivot locations, changes in the geometry, wheel size and chain line.

The chain plays a big role in anti-squat, beginning with its duty to transfer the driving forces from the mainframe to the rear wheel. It’s from these driving forces that the whole load transfer story starts. However, a common misconception is that anti-squat is the chain force and nothing else. It’s in the combination of the chain line with the suspension pivot layout that we get our anti-squat force line. While the suspension layout can remain static, we can actually move from an anti-squat case to a pro-squat case just by changing the gears and so changing the chain line. So, the chain’s effect in the anti-squat calculation is in addition to the inherent anti-squat properties of the linkage system. The chain can then either add, do nothing or work against the anti-squat of the linkage system. There's also more ways than one to accelerate a bike.

Looking at high pivot bikes with idler pulleys we can see that they have anti-squat percentages often over 100%. This is despite there being only a small proportion of the anti-squat forces coming from the chain line, which is often positioned close to or directly through the main pivot. That means it’s then the pivot location which is providing the high value of anti-squat, the location of the pivot dragging up the anti-squat force line.

Pedal kick back is caused from the growth of the chain as you go through your travel. Some bikes have more and some less, but generally as a bike has more anti-squat it will exhibit more pedal kick back. Pedal kick back might actually never be felt at the pedals, as this chain extension can result in a rotation at the rear wheel.

It’s also with the chain that we find one of the potential downsides of anti-squat. The amount of anti-squat is followed closely by the amount of pedal kickback for most suspension designs. The aforementioned high pivot idler pulley designs being an exception. As the suspension moves through its travel the distance between the rear axle and the chainring increases. The top portion of the chain then needs to compensate for this change in distance and causes a rotation at the cranks or a rotation at the rear wheel. It may even do nothing at all if the angular velocity of the rear wheel is higher than the angular velocity of the cassette and freehub when the chain extension gives it a good tug. Pedal kickback gets measured as an amount in degrees and its magnitude increases or decreases as the amount of anti-squat increases or decreases.

Pedal kickback is a whole other topic, with research already done into its relevance, how much is acceptable and in what scenarios the theoretical degree of motion of the cranks will actually happen. But as mentioned, the anti-squat works best in a reasonable window when it doesn’t stray into extremes and if that is the case then the amount of pedal kickback will also not enter into an extreme amount.

Now that we understand what anti-squat is, how we calculate it and what its percentages mean, we can add in more and more pieces to bring the analysis back to real life and start to consider how we can really use anti-squat.

With the anti-squat calculation only considering forces and moments from the load transfer due to acceleration we could perceive that 100% is the amount to aim for to combat all the squat tendencies. However, in real life we have more loads trying to compress the suspension than just that from the load transfer.

The stripped back nature of the anti-squat calculation only looks at the effects of load transfer. In reality there are many more forces acting to compress the suspension when we pedal or accelerate. When you push on the pedals some of that force is going to want to directly compress the suspension, so maybe it's good to counteract that too, as well as the squat from load transfer.

Even if you remain rigidly still while pedalling your CoG will want to rise up as you push down on the pedal. This acceleration of the CoG upwards is balanced by a downwards force on the bike, which will want to compress the suspension. So, pedalling a bike with that ideal 100% anti-squat figure will still result in the suspension compressing. Raising the anti-squat percentage above 100% means that we extend the suspension slightly with the extra anti-squat force. This extra extension can be used to counteract that extra vertical load put on the bike while pedalling. There is again a limit to how much extra anti-squat force we generate.

Coming back to the amount of acceleration that a human can achieve on a bike, ebikes have upped that acceleration amount. For anyone that has ridden one they will know the hugely increased sense of gathering speed. This up in the amount of acceleration also ups the amount of load transfer and can make it happen in a shorter time. While the power delivery might be smoother on an ebike compared to a non ebike, the detrimental effects of the load transfer are more profound and tyre loading and unloading can be more severe. Added to this the system’s CoG is in a different place with the addition of the battery and motor. So perhaps on ebikes more than any the design of the anti-squat is of high importance to ensure an efficiently powered bike with good traction.

The CoG position of a rider can change horizontally, meaning the 100% anti-squat limit would remain the same. But the changes in load at the front and rear tyres from this shift in CoG could result in the effects of the load transfer and anti-squat forces being more or less prevalent.

Once we add more and more real-world factors into the anti-squat equation, we can start to see why it becomes difficult to design bikes. Once we start to see also the influence in CoG position that rider size, riding position (pedalling seated or pedalling stood up), terrain (riding on flat ground or riding up a steep hill) and the fact that a mountain bike rider is a wildly moving quarter horse power engine during even the least rowdiest rides we can start to see why we needed to strip the analysis back to a more simple beginning and add in some assumptions.

With current analyses made visible to the public, and without a lack of a standard set of rules for each manufacturer to abide by when analysing anti-squat, there are quite a few assumptions that need to be stated and understood.

Most of the time the CoG height is fixed, and no consideration taken into its varying position with rider height and seated or stood pedalling. The analysis is also conducted on flat ground, where the distribution in load between the two contact patches is closer to equal. If we were to tilt the bike uphill we would naturally start the analysis with a higher proportion of the load on the rear contact patch and any bikes that would squat with the load transfer would come dangerously close or completely remove any load on the front and cause the bike to start to loop out. For our analysis in 2D we also consider the bike to be perfectly vertical.

When climbing a steep hill our weight is acting in a different direction to when we are on the flat. Firstly, there is a component of our weight pulling us back down the hill and as the total needs to remain the same it means we have reduced overall load at the tyres. Our load distribution between the tyres is different too, with more load on the rear and much less on the front. If in this situation your bike has a tendency to squat then it won't take much acceleration for the effects of the squat to remove all the load at the front contact patch and you're left with zero traction.

Riders of different sizes will have different CoG heights and so the 100% anti-squat line will be different. It then makes sense, in an ideal world, to adapt the suspension of each size of bike to the size of rider that will be on it. A larger rider would need the anti-squat force line to cross the front axle higher up to achieve the same amount of anti-squat. It does however increase the number of variables in manufacturing a bike, but some brands do it.

As we compress the suspension to analyse the anti-squat throughout the full range of travel, we must also consider what the front suspension is doing. To keep things simpler, we can say that the fork stays extended. This will generate an anti-squat graph using lines to describe what’s going on. We could add in a degree more complexity and compress the front suspension in time with the rear, allowing the front and rear to arrive at full travel at the same time. This too would generate an anti-squat line for us to interpret. What we can also do is to analyse the anti-squat at the two extremes of zero front travel and full travel and plot two lines on the graph which would define a window of anti-squat that we would be working in. All three methods can be done, but the assumption of which one should be stated in the analysis.

We can further increase either the number of curves or windows on our graph by accounting for multiple chain lines. On most mountain bikes we have a varying chain line even with a fixed chain ring size.

A single anti-squat curve also depends on the position of the front wheel. If we draw anti-squat curves, one for zero front wheel travel and one for full travel then we generate an anti-squat window that our bike will operate in. But it can't just be taken at face value. The real world situations need to be considered again, as you'd rarely be pedalling your bike with the fork bottom out.

Analysing anti-squat needs us to simplify reality to understand how we’re combating the tendency of our suspension to squat with load transfer. Hopefully the article has gone some way to explaining how we do that. But hopefully more, our re-insertion of that analysis back into the real world and the additions of other loads, situations and considerations helps give some further understanding and food for thought into analysing our bikes.

• 115 1
@dan-roberts really interesting article that explains very clearly balancing technical content without going too geek and making it confusing. What I think would be great for a future article from you would be a piece explaining the different engineering terms in bike design, how they balance each other and where the big trade offs occur. Top work and I really enjoy your articles
• 26 0
Can't agree more : Sweat graphics, excellent explanations. Thanks
• 13 2
I'm totally anti squat, it was squatting with a guy on my back in exercise warm ups that damaged my knee... Brilliant article though!
• 18 0
Agreed, I consider myself a bike geometry nerd and this adds to my understanding substantially. Thanks for great work.

I’m also 6’3”, and this makes it really clear why a single suspension design, STA and chain stay length across all sizes of a given frame will behave totally differently for different riders. Let’s home more brands start varying these numbers with size, Norco-style!
• 5 8
great suspension article: enduro-mtb.com/en/mtb-suspension-systems

little less nerdy
• 2 0
Totally agree, thanks for the article.
• 1 1
@jaydawg69: little less nerdy but also a little less correct. They tried to put things in boxes and it’s not what I would call a good thing.

The present article is brilliantly clear despite being very detailed and very technical. Thanks Dan. Looking forward to the anti-dive article as this is a little less clear to me.
• 1 0
@EnduroManiac: what’s incorrect on the article I posted?
• 3 0
@EnduroManiac: I helped out with that Enduro Mag article. It just had different drivers resulting in a different outcome. The information is the same, but here we went really deep into just one topic.
• 4 0
@dan-roberts: there’s (from memory) one thing that stroke me and is linked to the + or - of a given system. For horst link it’s said something like « most designs have a low anti-squat ». I find it very much misleading, it will just get in people’ s mind that horst link = low anti-squat. You know better than me that any system can have high or low AS. Just like you can have several leverage ratio curve on any given suspension design.
I remember an editor who kept saying that VPP’s drawback is that the leverage curve can only be regressive-progressive. Not only it’s not the case for all Santa Cruz bikes now, but it was already the case for the V10.
There’s enough wrong statements made in the industry (and repeated by many) that’s why I disliked this simplified article as it may fuel this trend on a few points. Can’t remember all of them.
• 2 0
@EnduroManiac: good points. The new Enduro and Demo are Horst Link and have great AS numbers.
• 86 2
This is all well and good, Dan. But it still doesn't help me explain to M'lady in a clear, concise and understandable way why I need a couple more bikes.

I tried telling her that the forks on my current ride aren't slack enough. She told me I was slack enough to make up for it...how do I come back from that?
• 22 0
Tell her you need to get longer. (And don't tell her that this will help you get through the fun bits more quickly)
• 11 0
Safety bro. Its all about your safety. New bikes are more confident, have droppers, less likely to endo, have better brakes and tires....
You wouldn't want anyone to get hurt out in the chunder, right?
• 3 0
You're on your own with that one!
• 15 0
You would think that a topic utilizing a Bunny would reduce the content down to the lowest common denominator ,or equivalent grade level,but you would be wrong.
• 21 0
Ha ha! That giraffe is on a bike!
• 9 0
Great read, I've tried calculating anti-squat manually when designing my own bikes instead of just using the linkage suspension software but always came up short on how to work of the CoG? Is there a general rule when working out where the CoG is?
• 2 2
Since CoG will be just in front & above of the saddle seated on flat ground, so can lower CoG by dropping saddle or leaning forward or back, but since ground angle & body position change that only really useful for static testing or predicting the movement of CoG as terrain change
• 5 0
~600mm above BB is fairly common in the linkage models. Small deviations won't make a massive difference.
• 4 0
On most people, their CoG is roughly their belly button.
• 3 0
since you are going custom, just measure yours. All you need is a scale, 30cm-ish block and a buddy

the calculator: tinyurl.com/v52e4rb

I always use the BB as a ref, not the floor
• 1 0
Thanks guys
• 6 0
This is the first article I've ever seen that actually accepts that mountainbikes may be ridden on slopes and with a body that can move with respect to the bike. So to me this is the first article on the subject that seems worth reading. Thanks Taj!
• 7 0
Taj did the illustrations, Dan did the enginerding.
• 5 0
Bravi!
I'm just sorry I don't know English very well. Translate the article into Italian, here bikers need fun! we are all at home
The right suspension setting remains essential!
• 7 3
"Pedal kickback is a whole other topic, with research already done into its relevance, how much is acceptable and in what scenarios the theoretical degree of motion of the cranks will actually happen. But as mentioned, the anti-squat works best in a reasonable window when it doesn’t stray into extremes and if that is the case then the amount of pedal kickback will also not enter into an extreme amount"

How much is acceptable? This is a very good question!!! Do you try to ride without pedal kickback?

We think that pedal kickback is a big enemy of riding.
• 5 2
Oh yeah, PK is the enemy !
There is a true miseducation around bike suspension kinematic.
Too much long-travel bikes are designed with huge AS/PK value to feel firm at pedalling in parking lot test but are a shame from suspension efficiency point of view. And PK values are never displayed, even in Dan Roberts article.
Bike will pedal well (high AS/PK values) OR have an "active" (i.e. working) suspension (low AS/PK).
There isn't any bike that can pedal well and have a true working suspension kinematic unless you switch a lever on the shock, until they got an ideler pulley that cancel chain growth.
• 3 0
@gnralized: www.instagram.com/p/B8yqM0giWha

High AS values with zero chain growth
• 2 0
That’s an interesting idea, does it effect how quickly you engage when you begin to pedal? @OchainFactory:
• 6 0
@gnralized: I think PK is entirely blown out of proportion. I ride one of the highest anti-squat bikes on the market (Orbea Rallon) and have never noticed PK to be an issue. Some riders of Rallons seem to think they aren't very sensitive as a result, but having ridden mine back to back against Santa Cruz, Yetis, Evils, Pivots, Norcos, etc. I can't say I've noticed much of a difference. Perhaps its because I'm a fairly large rider (95kg)? Not sure how the various models account for rider weight input to the pedals influencing all this.
• 6 0
@ratedgg13: Totally correct. Rider weight negates the effect and transfers it to the wheel.

Pedalling efficiency of the rider is a huge factor too. Someone who can pedal smoothly at 100rpm without actuating the suspension would have no issue with a 50% AS bike and can gain in suspension sensitivity. A guy who mashes out the saddle at 70rpm needs 120% or it “doesn’t pedal well bro”.
• 3 0
@ratedgg13: When I had a Turner 5spot DW, the PK was driving me nuts on climbs. It was not so much the PK, but the feeling you got when the suspension extends. All that PK is released back and the cranks lurch forwards. Really annoying feeling. Don't know the AS on that bike though.
• 10 1
Just run cheap OEM hubs with low engagement.... problem solved
• 1 0
@ratedgg13: Rider weight is definitely a factor, possibly a crucial one. The pedal kickback force is independent of rider weight, while the force opposing it is precisely rider weight, so the heavier you are the less it affects you.
• 1 1
@twhart20: yes, you have the same degree when you start to pedal. But in 6° setup (you can swich from 6° to 12°) it is almost impossible to notice. The 12° degree is designed for DH race on track like Blakesnake Val di sole (extreme rought track) :-)
• 2 0
@Gmang: with a 10 degree freehub you have a random availability of grades (from zero to ten) which implies a "random suspension feeling"
Ride chainless is different
• 2 0
@R-M-R whats your thoughts? You've done more thinking on this than almost anyone (including most major brands...).
• 2 0
@ratedgg13: I've done a couple of chainless downhill races and the suspension is much more reactive without the chain. Spend a day at the bike park without a chain, and I guarantee you'll notice a big difference.
• 7 0
@ratedgg13: It's true, I've recently gone down quite a rabbit hole researching and modeling it. I still need to think through a few things, but I'm fairly confident pedal kickback WHILE COASTING is almost nonexistent. It's definitely real while pedaling, though it doesn't cause reversal of the rotation, just changes in the forward rate that feel like a reversal, but are nowhere close to actually reversing the rotation.

When coasting:

On low kickback bikes (kickback is closely linked to anti-squat), it's impossible. Your bike and your legs would break from the severity of the impact before kickback occurs.

On very high kickback bikes with high engagement hubs, it may be possible on severe impacts in the outermost sprockets; still have to work out how severe (probably not very) and how frequent (probably not very) it's likely to be.

On typical bikes with typical hubs, if it occurs at all, my current thinking is it's so minor and so infrequent as to not really be a thing. I'll discuss it in the forums when I'm more confident in the new way I've set up the equations; for now, it looks like the conditions to produce kickback on a typical bike are so extreme in every parameter that it should be rare and minor - and if it occurs, it's the least of your problems. I'll update when I feel good about the numbers.
• 3 0
@OchainFactory: This assumes no forward motion of the bike. When coasting, the hub is "spooling out" chain while the chainstay elongation is "taking up chain". If the rate of spooling out is greater than the rate of taking up, kickback cannot occur.

It appears you may have neglected to account for the forward motion of the bike. I hope we can discuss this further.
• 1 0
Consider this anecdotal: I recently went from a Pivot 429 Trail (DW link) to an Orbea Rallon (Split pivot). I’ve noticed that on long, rowdy descents my trailing leg gets very fatigued, which never happened before. Perhaps this is occurring from the increased kickback of the Rallon’s design? Or just from the increased amount of travel?
• 7 0
@Paddock22: Do you have a longer dropper that's allowing you to squat lower? Kickback shouldn't fatigue the trailing leg, but squatting low sure will. I went to a 185 mm dropper and my trailing leg immediately felt it!
• 6 0
@R-M-R: I did! I hadn’t thought of that but it makes total sense! Thanks!
• 3 0
Interesting discussion, gents. I hadn't considered a lot of this.
• 3 0
@ratedgg13:
Yes, maybe. I'm 75kg and never thought of the rider weight effects as a sensibility factor, so thanks for that !

I would like to explain the long (empirical) way that makes me conclude that high AS/PK ratio was not compatible with optimal suspension operation i.e maintaining traction (tire contact with ground) and damping shocks - and nothing else.

All the bikes you're referring to have pretty high AS/PK, this can explain that you didn't felt a big difference with the Rallon.
Particularly the Norco sight and range (particularly the 650b first gen) got way more AS/PK than your Rallon and in a lesser way, the Evils.
I owned a lot of these bikes from 2013 : split pivots (Devinci Troy, Trek Slash), 4-bar ( Norco range and sight 650b 1st gen - still own the sight in 3x setup), sliding monopivot ( yeti SB5C). All of these are in the same window for PK, i.e. 22-27° at full travel in 32-46, with pretty linear PK ratio (i.e. PK growing constantly along travel), except the yeti at the end of travel.
I didn't ride bike parks or groomed trail, mostly surrounding natural trails in Quebec then south of France with lots of stones and roots, and (to my regret) few flowy sections. I mostly grin the upper section of my cassette, didn't go frequently below 18 (32 tooth ring).
Most of bikes I rode used the ubiquitous 210*51-57 "legacy" shocks, and most of these bikes got a progressive suspension ratio (which exception to the yeti), and I always swapped the manufacturer shock to high-end 4-ways shocks, CCDBA or Fox X2 circa 2015, with standard to light compression tune and always ride with compression almost fully open.

On all of these bikes, given my riding trails and bike setups, I ever felt a strong feedback (like a hardtail feeling) on small bump/square edges/stairs-likes steps, and a very bad traction on loose terrain, be it climb or downhill.
Something that lasted for almost 5 years, that makes me thought that my suspension wasn't optimally working, or something was wrong with my setup. I was constantly looking for improvement, so my multiple bikes swapping.
I finally went to the nerd thing, bought the Linkage software and start to compare sus kimematics curves. I'm not a mechanical engineer by any mean but got a PhD in fluid mechanic, so not completely dumb with figures and physical theory. I ended isolating the 2 constants of all the bikes sus kinematics I rode: high AS/PK figures and progressive ratio.
So to test my theory I looked for and bought a bike with low AS/PK and almost linear ratio with the simplest linkage : a 1st gen monopivot Kona Process 134 - I acknowledge that it was not a very rational methodology since I needed to isolate each one of the variables I would to test for sensibility, instead of having to discriminate the effects of both variables.
But it gives good results and that was the first time I can said that my suspension was actually working. Bike got a lot of traction, no foot feedback and a sensation that defined what small bump compliance should be. But it blow through it travel. Constantly. Despite numerous shock swap and fine tuning. So I made the assumption that it was the effect of the very linear ratio in a context of low chain growth/minimum transmission-suspension interactions.
So I finally looked for a bike with low AS/PK and a progressive ratio with constant progression to isolate hydraulic effects. To my knowledge there was very few of this kind, in 2017 there was almost only one brand doing this kind of stuff: Knolly.
I bought a Warden, swapped the DBIL for my long lasting 2015 X2 and it worked just fine. This setup had a huge small bump compliance, traction is first-class and the bike that didn't blow through its travel at the first bump…
So I concluded that low AS/PK and progressive ratio was the way to go given my usual trails, habits, age, and now, weight.

Sorry for the long story.

Side note - it was rarely written but some of the 2013-2017 bikes suffers a lot in term of AS/PK going from 2x - 3x setups (for which they were designed - Norcos, Yeti SB 1st gen) to 1x setup since bigger plates minimize PK. Having more than one plate give designers one more freedom degree in designing full sus kinematics.
• 1 0
@gnralized: Fair enough - I'd just also note that @R-M-R is a very serious engineer (not the armchair variety) who has worked in the bike design space as an engineer and has tested all of his claims both in modeling and with prototype bikes. So he's got a lot of weight behind what hes saying.
• 2 0
@ratedgg13: I'm blushing! I don't disagree with gnralized in principle, but I think many variables need to be separated in that analysis, a few sensations are being incorrectly attributed to other parameters, and the claim of the Norco models having higher kickback than the Rallon appears to be inaccurate when measured in comparable sprocket combinations, accounting for the different configurations of a multi-chainring bike vs. a single-ring bike.

Damper properties, axle path, tire size and pressure, spring curve, leverage curve ... many variables need to be considered before kickback can be isolated as the cause of what a rider feels through his or her feet. Modeling of kickback shows that only some bikes will ever experience it, only in some sprockets, and only sometimes. Only a small number of bikes will experience it frequently with certain hubs and in certain sprockets, and even then I'm not convinced it's a significant issue. Many bikes will never experience it.
• 1 0
@R-M-R: I'll add that gnarlized seems to be right in avg weight (75kg) category for what most bikes are designed around. So it's interesting that he couldn't get dialed in on some of the bikes he mentioned.

Could it be that riding style also has a significant impact on what a rider feels and how the bike reacts? I tend to be a "hover" type rider where I'm slightly weighted on the seat when I climb this lets me shift my weight around as needed and to be frank, I don't really notice pedal bob on most bikes when climbing. I've noticed that some people stay firmly seated for maximum traction but they tend to complain about pedal bob, even on bikes that are supposedly good pedaling platforms
• 2 0
@Paddock22: I can't comment on gnralized's set-up or the trails he rides. Maybe there's some unusual combination of factors in play on his trails, like sections that make sense to be coasted at low speeds in a high ratio. Uncommon observations may have uncommon, but perfectly reasonable, explanations. With what we currently know, I can comment only on what I can model.

Pedal bob is not the same as kickback. Bobbing is either excessive squat - or, less likely, rise - due to pedaling forces or a resonance in the suspension due to your pedaling. Kickback is an unwanted change in pedal rate of rotation due to suspension movement. A bike could have a considerable amount of one and a low amount of the other.

Yes, riding style and bike set-up have some effect. If you use less travel than other people over a given terrain, you're less likely to experience kickback.

When pedaling, all bikes with rear suspension are subject to kickback and/or a moving bottom bracket (the latter in the case of GT's i-Drive (etc.) or URT designs). If you haven't noticed kickback while pedaling, it could be that your bikes experience only a small amount, your trails don't cause a great deal of it, or you're mostly unbothered by it.

Kickback while coasting is a different situation that does not affect all bikes, is effectively eliminated in some sprocket combinations, and does not occur often, even on bikes that are prone to it - and if it does occur, it may not be enough to notice or cause any noteworthy effects.
• 1 0
@R-M-R:
I'm mot a mechanical engineer and I just haven't enough time to dig more theoretically on what I'm feeling.

Most of my observations of "alleged" kickback effects are during pedalling in high gear/low speed or coasting at low/middle speed in middle cassette range.

I'm really picky in sus feeling, on a warden prepared for having very few pivot/bushing friction and an X2 coil, climbing a coarse gravel trail in 32/42 I can tell the difference of 4 click less in low speed compression on traction feeling (4 from full open to close).

I didn't do the enveloppe calculations, but assuming an instantaneous engagement freewheel, is it right to say that if instantaneous chain growth velocity is higher than the instantaneous wheel velocity at the engaged gear radius position (i.e. instantaneous engaged gear velocity), then then upper chain is in tension between ring and gear ?

I was talking about the 1st gen 650b sight and range who used a kinematic layout close to giant NRS, lapierre OST 1st gen, etc... with horst link pivot far below rear wheel axis (chainstay pivot).
The range 650b 1st gen got AS/PK figure really close to the rallon. The range 2017 is not so far, too, in the 22-27° PK range in 32/50@150mm, although PH decrease more in the lower gear than rallon and Range 1st gen.
I cannot share Linkage software graphs but here is the publication of Antonio Osuna's blog.
From the software:
PK for 32/50@150mm - 32/10@150mm:
- Rallon 2018: 26.79° - 9.04°
- range 650b 2013: 27.75° - 8.82°
- range 29 2017 : 22.98° - 5.19°
• 2 0
@gnralized: We should separate kickback while pedaling from kickback while coasting. Kickback while pedaling is real, is unavoidable, and is well represented by the outputs from Linkage. Kickback while coasting is the effect that's difficult to quantify and turns out to be a myth on many bikes.

When comparing the Range and Rallon, instead of using the same sprocket combinations on each bike, I tried to use the most realistic matches. For example, several models of the Range were spec'd with 2×10 drivetrains, so I used the 36T chainring vs. a 30T on the Rallon. Cassette sprockets were chosen to give the closest possible rollout (distance traveled per crank revolution). Not exact matches, but close. If I use the 30T single chainring spec'd on some models of the 2015 Range, the kickback on the Rallon is very similar to that of the Range. My Linkage model is labeled "model year 2015", corresponding to the model update released either late 2014 or early 2015; I'm unsure how this compares to a prior model, but a quick visual inspection suggests they're similar. In any case, the point is that you said "Particularly the Norco sight and range (particularly the 650b first gen) got way more AS/PK than your Rallon" and I wanted to show this isn't the case. The values are similar between these bikes, with the Rallon sometimes being higher, depending on the sprocket selection. Regardless of the exact configuration, the Norco isn't "way more". This does not discredit your arguments, it's just a point of clarification.

It's reasonable that you would notice the difference from four clicks of low-speed compression.

If your bike took a large impact while coasting at low speeds, this is a scenario that could produce pedal kickback on some bikes. Not a Knolly, though, as they have extremely low kickback.

If I interpret your statement correctly, then yes, your condition for the onset of coasting kickback is correct. You can use various inputs for the portion of the travel used during an impact and the rate of shock compression to determine the actual kickback experienced. This page provides some models for vertical wheel velocity while passing over an obstacle, assuming a rigid tire and static chassis: www.shimrestackor.com/Code/User_Manual/Sections/Suspension_Velocity/suspension-velocity.htm.
• 2 0
@R-M-R:

You're right, they are similar, I confused myself, no matter sprocket range ;-)
I never rode the Rallon but when I checked its figures without comparing I stayed on the perception that its kickback cannot be higher than that of the 1st gen range (which was huge for me with a 1x 32/11-42 !), which proved to be wrong...
Didn't get that the Rallon had this much PK...

Thanks for sharing your knowledge, I will have a close look at it.
• 4 0
here's an experiment from Antonio Osuna's blog where he concludes that kickback has no effect when coasting downhill and is only worth considering for climbing. A bit more nerdy blog, but really worth reading.
• 3 3
"There are many other factors that affect bicycle performance than AntiSquat. Geometry, derailleur, overall chain growth, rider style, tires, frame stiffness, leverage ratio, travel, shock type, and aerodynamics. All this together makes the bike but the most important thing is the rider. We can make a bicycle which is easy to ride but we can not make the rider change their way of riding. The AS is good and bad at the same time but if we can control it’s effects reasonable it can be good. Modern bicycles are very effective but still Aron Gwin won in Leogang chainless = without AS and pedaling" from Pole website. (polebicycles.com/what-is-anti-squat-and-pedal-kickback)

I believe more in a good bike builder!
I ask the question again: Do you try to ride without pedal kickback? (Just remove the chain)
• 3 0
@OchainFactory: never w/o chain, nope, .ight try just to see.
I had a SC vp free, tho, lost of kick back, and an Intense 5.5 (a bit as well), and only felt the pedals pushing my feet when pedaling uphill (hub already engaged). What Osuna says about angular speed of freehub due to kickback vs. angular speed of wheel makes total sense to me. And that guy knows a bit on the subject too
• 1 1
@ismasan: ok I understand. But why riding without chain feels totally smoother?
• 3 4
@OchainFactory: chainslap comes to my mind. Feels smoother not due to better suspension action, but to the lack of chainslap, wich vibrations can be felt through the pedals.
• 1 2
@ismasan: No it’s obviously the chain cogging around the freewheel reducing sensitivity.

If people want grip they need the most supple and sensitive suspension possible. Removal of friction in all forms is essential. Go watch Bruni’s runs.
• 1 2
@jclnv: well said
• 5 0
@jclnv: apart from many things quite unique feature of Loic and Amaury is their management of stretch reflex when going between eccentric and concentric contraction. They are banging the bike into G outs and they can brake very hard. Hart is also a good example of it. If you want to get faster on the bike you better focus on what your body is doing, your won suspension, rather than some tech details... Gwinny has reached what is possible with staying hard and tense over everything. The new era of DH is ruled by plyometrics.
• 1 1
An easy experiment you can do yourself is to spend a day at the bike park without the chain. I've done a couple of chainless DH races, and it really does make a massive difference in how responsive the suspension is.
• 1 0
Years ago, I had a single pivot bike without linkage that was designed for a big chaining (around 42t) but I switched to a much smaller one (around 36t, can't remember exactly). On slow big drops (a step after a turn), due to the linearity of the suspension, the KB was pretty high (deep in travel). It was horrible - even more due to a hub with a lot of engagement points - as you want to absorb as much as possible in this kind of situation. So I think Antonio is right, most of the time is not so relevant, but do not forget that there is a person on a bike that should feel confident.
• 2 0
@jclnv: The friction of moving a few degrees a few links of a chain is next to nothig compared to suspension forces,
or are you being ironic? 0_o
• 1 0
@ismasan: Have you calculated it?

Have you tried a needle bearing DU bush compared to a normal one?
• 3 1
@WAKIdesigns: Nonsense! Those two have totally opposite riding styles. Bruni is basically static compared to Pierron. I highly doubt Bruni’s bike would work as well if he didn’t ride like that. The spring rates and weight distribution are so perfectly balanced that he only has to use minor inputs to change front to back grip levels. Pierron rides like a madman to make up for a bike that is nowhere near as well set up.

Gwinn ran firm suspension and to keep the geo static but he had to muscle the bike down the hill which worked to a point until some faster, smarter, kids came along and that philosophy runs out of legs pretty quick.
• 1 0
@jclnv: no, but holding a chain in the air I can move the links by blowing with my mouth, that gives me an idea of it's massive friction.

The needle bearing vs bushing is not a valid example, shock mount bushinghs work loaded with hundreds of kg
• 2 0
@labourde: A bike designed for a 42t is a rarity and in any case, it would have to be a very old model...
• 2 0
@dan-roberts I have always wondered whether it is possible to quantify how much of the anti squat is inherent to the suspension design (as if there was a motor in the rear hub) and how much is attributable to chain forces. The part explaining how anti squat changes with chain line looks like a nice place to start, but then I wonder, are we still just considering the variation within the AS spectrum which takes the combined action of suspension design and chain forces into account?

I mean I also understand that talking about weight transfer without considering the chain is kinda pointless, but still...
• 2 5
we have done many tests in recent years. And we have come to the conclusion that eliminating pedal kickback is the right way. Without the pedal Kickback it is possible to design a suspension with very high values of Antisquat without affecting the performance of the suspension in descent. it is always useful to remove a variable from problems ...
• 6 0
High-pivot designs with idlers are effectively a form of 'linkage-only' anti-squat. (or close enough)

A bit simplified; but you can say that the 'linkage-anti-squat' is proportional to the inclination of the line between rear axle and IC (IC higher than axle gives anti-squat, IC lower than axle gives pro-squat), and the 'chain-anti-squat' is proportional to the inclination between the chainline and the line between the rear-axle and IC. (chain-line point down relative to rear-axle/IC-line gives anti-squat, chain-line pointing up relative to rear-axle/IC-line gives pro-squat)

For high-pivot designs, the line between axle and IC is strongly inclined, leading to a high linkage-antisquat.
Usually, for such designs the chainline is (almost) perfectly parallel to the line between axle and IC, then you will effectively have no contribution of the chain force to the anti-squat. This also gives approx. zero chain-growth and thus zero pedal kickback. However, I am also of the opinion that pedal kickback is not really an issue, especially when descending. (and thus coasting)
• 1 0
@gillish: thank you, that was a good explanation.
• 1 0
@gillish: It seems like most high pivots on the market do drop the idler pulley position a bit below the pivot, introducing some chain force anti-squat back into the system,
• 1 0
@dthomp325: You're right. In fact to have 100% 'linkage-only' AS, the main pivot would have to be so high that the line from rear axle to main-pivot (or IC) is parallel to the line from rearwheel contact patch to cog-height above the front axle. (the famous 'anti-squat line' that is shown a few times in Dan's article) For this your main-pivot would need to be around the seattube/toptube junction. I guess this is not desireably in terms of axle-path, which would become too rearward.

Things become even more interesting when the idler is mounted to the rear triangle. (like on the Deviate Highlander) Now the 'chainline' that matters for AS is actually the chainline from crank to idler, as that is what will exert a force on the rear triangle that could generate AS.
• 2 0
the vertical forces applied to the suspension from the rider of a bicycle pressing downward on the pedals is a much bigger contributor to unwanted suspension squish than load transfer due to acceleration.

After all, humans are weak "engines" that don't accelerate with much force relative to the resistance of the suspension with a decent amount of spring resistance and low speed damping.

pretty sure you could measure the affect scientifically or anecdotally of acceleration weight shift induced squat vs pushing pedal downward whilst pedaling induced squat pretty easily. consider how much bob there is when your moving at a constant speed but hammering on the pedals to maintain it, the acceleration is out of play but the lions share of the pedal bob is still happening.

This article mostly glosses over the real nemisis of efficient pedaling on a full suspension " Ol' Pedal Bob" and misleads us to believe it has something more to do with acceleration. Anti squat could be considered the way that it is mostly focused on in this article, but the real culprit to which chain growth is engineered to combat is Pedal Bob
• 2 0
This is a great point. The whole anti-squat analysis was developed for motorcycles. High accelerations (relative to bicycles) with a fixed center of gravity. These calculations make much more sense in that world.

However, even though I agree with you that mashing on the pedals and all of the up and down movement that goes with that has as much (or more) to do with pedal bob, I think the analysis still holds. I just think we would be better served by talking about it in a way that was more "this is how much your pedaling forces will resist pedal bob" sort of way than in a "this is the perfect calculation for exactly how much pedal bob you are going to get" way that most people phrase it.
• 1 0
This partly true. But largely this is true because bikes a so good these days.

If you take 70% of the problem away and 20% (pedal bob from leg mass movement) is left, that 20% is now 66% of what is left. And so it is way more obvious.
• 1 0
@dirktanzarian: i agree. The wjole thing is far from cut and dry, there are too many wild variables with peoples pedaling characteristics; not to mention their equilment & setup
• 1 0
I though Dan discusses the downward pedaling force a bit, including them in the umbrella of total downward forces acting on suspension? Admittedly it grabs riders' attention more to think of suspension design as a battle between pedaling efficiency and suspension suppleness... whereas rider technique comes down to individuals. As @jclnv pointed out above re: smooth high rpm's vs. pedal mashing.
• 2 0
"So, the chain’s effect in the anti-squat calculation is in addition to the inherent anti-squat properties of the linkage system."

You haven't mentioned anything about any inherent properties of any linkage system. The only anti-squat formula we've been shown _needs_ the chainline in order to compute the anti-squat.
• 1 0
Is this why mfgs are reluctant to change CS length at the dropouts for FS bikes, perhaps doing it Norco style by moving the BB forward?

Does the front center increase in larger sizes cover CoG height change of taller riders adequately?

I believe the bigger prob is the weight distro. On an old school bike like the Remedy 29, while seated, 70% of your bodyweight is on the rear tire on level ground; while standing, only 55% is on the rear. This might sound good, since you want weight on the rear for slip-free propulsion, and weight on the front for wash-out-free cornering, but doesn't this lead to why long travel bikes suck at climbing, since riders are less likely to compromise on DH susp tuning? There's excessive sag on the rear due to so much weight on the back on an incline (even more than 70%), due to tuning rear to be supple out of the saddle. Some brands might opt for a longer CS to improve seated climbing (like on ebikes), but that's known to be detrimental to airtime-worthiness. Some are going even steeper with STA and reducing the difference between sitting and standing. New topic worth covering?
• 2 0
The main (kinetmatic) reason for not increasing CS length at the dropout is it increases leverage. You end up with a softer suspension for longer chainstays which works against the taller (and heavier) riders on bigger bikes.

Also its quite a bit more complicated to execute on mechanically.
• 1 0
@RoboDuck: Exactly!

Why don’t people get that?

It makes a huge difference.
• 1 0
To qualify that statement you can also build a low-leverage long chainstay bike. You just can't lengthen the chainstay on an existing linkage without raising leverage.
• 1 0
@RoboDuck: Leverage increase means travel increase, right? How much increase for 430 to 440mm CS (e.g. Tallboy V4)? You make it sound like a big deal, but is it? I doubt it's as much of an effect as a RM Ride 9 chip, nor something that can't be tuned for, such as adding air to the shock.
• 1 0
@Varaxis: Little bit of travel difference but the big difference is that it moves through your spring and damper quicker so you need more air pressure and sometimes firmer damper tunes. This could mean the heavier guys are at or above max pressure.

Its not massive for 5mm of change but it starts to add up if you are increase 10-15mm.
• 1 0
@RoboDuck: there seems to be a consensus among taller riders on Tallboy V4, Megatower, etc. on size XL+ who much prefer the longer CS setting (+10mm). They say it feels more balanced. That's why I get the idea that weight distro matters more, and isn't getting enough attention..
• 1 0
@RoboDuck: ep1.pinkbike.org/p5pb17533361/p5pb17533361.jpg

Higher average leverage ratio means there's a lot more movement at the axle than at the shock, I thought, moving through spring and damper slower.

Real numbers to show how small it is on a '19 Kona Operator for 15mm CS length change at dropout. 5-10mm extra travel means you feel more travel per bump, and leverage ratio increases from 2.5 to 2.6 (@ 100mm)
• 1 0
@Varaxis: Yep it does move slower. Less compression support from the damper and you need to run higher spring rates. Overall a softer setup. You will get a bit more travel though.

You debate how much to care about the difference but it is certainly perceptible on trail.
• 1 0
@RoboDuck: what are you even debating? That really heavy people exist and a bike shouldn't have adjustable chainstay length, since these people are already at the limit of how stiff they can make it? Cater bikes more to such people?

You want people to pay attention to micro differences, some that they can tune for on the trail, over more meaningful ones that need to be handled at the design stage?

You like short CS, are afraid such bikes will go extinct and just want to BS to deter the idea of longer CS in larger sizes from taking off?
• 1 0
There must be a manufacturer out there that's made some prototypes with different anti squat values. It would be great for a PB tester to have a ride and give us some feedback on the real world differences from varied anti squat.
• 1 0
If you've got a design with some kind of floating IC it is probably doable to have one link with a flip chip to change the instant center quite dramatically. It would probably be a lot harder with a single pivot design, unless it has got an idler pulley. Then you'd just shift the idler using a flip chip. Provided the chain is long enough of course.
• 1 0
@vinay: I was thinking that someone like Trek would have made a number of frames that were identical other than AS (different main pivot height), but I'm sure a burly main pivot chip on something like an Orange or a Kona could do the job. Shame they didn't add something like that to GrimDonut
• 1 0
Cannondale did size specific tuning with their latest Habit.
• 3 2
Anti rise is so much more important. The is the measure of how the suspension responds to weight shift when not under power (and braking).

Our body is all over the pace when pedalling on an Enduro or DH race run. We can deal with small changes to suspension behaviour and it's only for such short period of time. But when freewheeling, carrying speed, cornering, we want the bike to remain as neutral as possible.

What's more important: pedalling ability when seated or Body English when descending. I know what I design my bikes for!
• 16 0
Totally, anti-rise is important. But we can only take a deep dive into one aspect at a time. Anti-rise will come at some point in the future.
• 1 0
@dan-roberts: Righto! I look forward to reading it...

It's just anti-rise often gets totally forgotten. Probably because much of the literature is based around motorbikes where indeed the power application is way more important.
• 1 0
@phutphutend: Same too with anti-squat. Which is where it needs a good dose of chin scratching to re-apply it to our MTB world. There's no preferential treatment with doing anti-squat first. I also enjoy anti-rise!
• 1 0
Your bikes pedal pretty well when seated anyway.
• 1 0
No Anti-Squat is just as important. It’s all important.
• 2 0
Anti rise is the most over-sold number in bike design. It has no effect while freewheeling, as you imply. There is no objective good, better, or best in antirise numbers, just different. Different riders often prefer more or less brake squat. Keeping the rear suspension highly active during braking can raise the COG and cause the fork to dive and be over-worked while descending. The opposite is true on the other extreme. Braking can and should be adapted to the terrain, suspension design, and riding style. Any racer will tell you that braking through rock gardens and corners is not advisable. Ask Amaury how much high antirise numbers affect his descending speed, or anyone riding a high pivot bike for that matter. The rear end of the bike only does a small percentage of the actual braking while descending, which reduces importance of anti rise even further. Fork, tire, and wheel setup become much more important during those scenarios.
• 1 0
@grizzlyatom: Anti rise is the parameter to do with any weight shifts that don't involve drive mechanics. Its not just braking. As such, how your suspension work when you shift your weight around on a bike is super important.

Obviously different bike designers can have different views on what is right and wrong, otherwise the world would be very boring! Personally I think neutrality is key. Anti rise (and anti-squat) as close as possible to 100% through the full stroke (or maybe as unchanging as possible noting cofGs vary).

Bikes aren't fast, riders are. Having a bike that is as intuitive as possible for the rider is always going to be the best outcome...
• 1 0
@phutphutend: the term anti-rise as used in these articles is purely related to how the bike's suspsension reacts to mass transfer DURING BRAKING. If you're talking about overall kinematics and suspenion dynamics, yes, many other factors that have to do with rider weight shifts come into play. However, that's not what's being discussed. This is from @dan-roberts first article about suspension dynamics on PB:

"When we talk about anti-rise, we talk about how the bike will react to mass transfer while braking, and so having the mass shift forwards. This pitching forwards effect would usually have the suspension system extending and having the bike rise out of its travel."
• 1 0
@phutphutend: To me, neutrality is definetly wrong. Riders need feedback. An obvious proof is the first generation of BMW telever, that the german company had to refine because most people had difficulties to know how much their motorbike brakes. This lead to some crashes. Also changing the mass load of the bike can have some advantages like giving more grip to the front tire and slightly reduce the wheelbase before cornering. This is typically used in moto GP.
• 1 0
@grizzlyatom: @dan-roberts am I wrong? If so, what is parameter to do with weight shifts other than braking and accelerating? Anti-rise does a pretty good job of being that parameter...
• 1 0
Compromises bump sensitivity too much. We still aren't even close to getting to the real potential in suspension design. There needs to be a fundamental concept change like we are seeing in front suspension but nobody has thought of it yet.
• 1 0
You mean Trust?

I’m pretty rear suspension design advances preceded the Trust fork, ie DW.

All things considered, a rear suspension built with the express intent to be combined with a Trust fork could make sense.

The only variables I’d like to see tweaked specifically for a Trust fork application is front center and HTA.

I suspect Trust has done this and likely found an optimized application.
• 1 0
Nah. Motorcycles have had the same suspension design for years, including telescopic forks.
• 1 0
The answer is Forbidden Druid, Deviate Highlander.
• 1 0
Awesome article, well written and fun graphics. I made it through most of it and realized that analyzing this stuff is fun but reaches a point where I have to ask myself, does any of it matter after a certain point? Fractional gains vs rider skill....I dunno.
• 1 0
"Above our 100% mark, where we’ve effectively separated the drive forces from the suspension response, the anti-squat forces will actually be more than enough to combat the squat tendencies, and the rear suspension will extend with that excess force."

But they're not separated above 100%... unless you mean they're separated _exactly_ at 100%... which is not clear in that sentence.
• 1 0
"A system that doesn’t combat any of the squat from the load transfer is going to result in unwanted suspension compression and you bobbing your way up a climb."

Suspension compression, yes, but bobbing no. Earlier you assumed constant acceleration, in which case any % anti-squat would not cause bobbing.
• 2 0
100% AS counters the load transfer from acceleration, not counter load transfer from anything else. The load transfer from acceleration is something that can always be expected; the amount of loading a rider adds from their own action varies. DW tunes in extra AS for rider movement, to make it so the shock is less relied upon.
• 2 0
@Varaxis: There was zero mention of rider inputs up to this point in the article. Everything so far has assumed constant inputs, so to throw in this mention of bobbing here is potentially confusing because it does not match the givens so far.
• 1 0
That is exactly what happens with a bike that has less than 100% AS or isn't equipped with a shock capable of compensating for squat with copious amounts of low speed compression damping. When going up hill the bike necessarily slows in the lulls between the strongest acceleration pulses imparted by the riders pedal strokes. For a low anti-squat bike (without other shock tricks) the suspension squats during acceleration pulses (and squats to a greater degree while climbing hills) and returns to Sag between these pulses when acceleration forces (squat and anti-squat) die away. It is suspension bob that we experience as we are riding, not squat or anti-squat. The disequilibrating (suspension compressing and extending) acceleration forces that we call squat and anti-squat acting on the suspension are capable of producing mayhem but on a bike that mayhem is always expressed as suspension bob. Suspension bob is not something you experience riding a bike with a well designed suspension configuration. But achieving a bob free ride on a suspension bike requires either i) balancing squatting force with anti-squat or ii) the use of relatively advanced shock capabilities like dialing in additional low speed compression damping or using some form of inertial valving. Another option is iii) physically altering the drivetrain relationship to the suspension geometry during the course of travel - GT and Lapierre both offered takes on the floating bottom bracket idea - but these attempts to create a bob free suspension using a floating BB seemed to offer no advantages over the other already mentioned solutions which were, in any case simpler. Also, getting these drivetrain varying tricks to work still required carefully balancing compressive and extensive acceleration forces and/or use of a precisely tuned shock.

If you are getting suspension compression only while climbing a hill (n.b. all suspensions compress to a point beyond Sag while climbing hills for the reason mentioned in the article - due to the changed body position of the rider relative to the bike more weight shifts onto the rear wheel) that is because your bike has a good balance between squat and anti-squat or because you have a very well tuned shock. But that doesn't mean the writers comments were incorrect. The writer was referring to bikes with an imbalance of squat and anti-squat. Those bike do bob (assuming no special shock controls) and they bob worse when climbing hills.
• 1 0
This just blew my freaking mind:
”Anti-squat is a good thing then, to combat the natural want of the bike to squat and compress the suspension. *While the acceleration of your bike might not change with or without anti-squat*, if you have some combating then you’re more likely to keep your CoG in a constant position, maintain a more favourable bike geometry and keep the load at your rear contact patch at a more constant level.”

Holy crap, all this time I've been working under the assumption that higher anti-squat values somehow translated into more pedal forces contributing to moving the bike forward, but this seems to be saying it's just making me feel better about the lost energy by minimizing bob?! I'm an idiot!
• 1 0
My Anti-Squat position lies on the couch where my COG sinks in where my butt dips the deepest in the cushion. My second anti-squat position, sort of, is over the toilet when I have to take a dump. The COG on that one is directly over the center of the bowl and my feet rested flat on the ground.
• 1 0
@dan-roberts Thanks for this great article!
I am trying to calcultate the AS of a bike single pivot + idler but i higly doubt the standard graphical method can be used.

In case of Idler, I read somewhere that the line through the upper chain line must be replaced by the chainline from the rear spocket to the idler.
Doing like this would lead to weird stuff on the graphical method:
The point of intersection (chainline with idler) / (line from the rear axle to the IC) could be much farther than the vertical line of the front wheel axle (DidlerDrear_spocket). It could also be behind the rear wheel axle if Didler >Drear_spocket. Finally this point does not exists when Didler=Drear_spocket.

What method/calculation would you recommand to determinate the AS of a frame using an idler?
• 1 0
@R-M-R
Any chance I get your point of view on this topic ? cheers
• 2 0
@o-clem-o: Depends whether the idler is on the same element as the BB (typically the front triangle) or on a linkage element. If the idler is on the same element as the BB, use the usual method, but use the chainline between the cassette and idler. If the idler is on a suspension element, save yourself a lot of suffering and just use Linkage software!
• 2 0
@R-M-R: thank you for your time.
In my case the idler is on the front triangle and is concentric to the pivot. That's why I cannot figure out when the diameter of the cassette is very close to the diameter of the idler. It is becoming a strange configuration as the chainline and the line from the rear axle to IC could be parallel and never intersect.
I like linkage but I really want to understand what's behind and not just take the results especially with an idler.
• 2 0
@o-clem-o: It's fine if they don't intersect: the lines just extend to infinity. You can approach the limit from either side and, necessarily, get the same result, so just imagine convergence on either side of the infinite case and split the difference between these scenarios.
• 1 0
I think Dan Roberts nails the topic of anti-squat and does so in terms that people without specialised training can understand. But I did find one dubious statement in the article. Speaking of a rider climbing a hill Dan said,"If in this situation your bike has a tendency to squat then it won't take much acceleration for the effects of the squat to remove all the load at the front contact patch and you're left with zero traction." I think that is somewhat misleading and I would like to try to explain why.

Dan, it should be said, sets the scene very well. Of course there is danger of losing traction at the front wheel while climbing. And, at the moment of the acceleration pulses it is conceivable that the front tyre might lift. And, the experience of suspension bob while riding up hill is certainly disheartening. Just when you are hoping for a bit of acceleration the bike depresses under your pedal stokes and it moves lazily and in a wallowing way up the hill. Your energy is being lost into this bobbing mess with the cyclical compression of the suspension happening just when you put in your most vigorous exertions meant to make the bike pick up some pace. That is truly a terrible riding experience. Personally, I can't stand suspension bob.

Now, despite correctly conveying the terrible riding experience that will be offered by a bike with a tendency to squat when riding up a hill it is misleading to imply that bikes with a tendency to squat are especially prone to traction problems. On the contrary bikes with a tendency to squat may offer better overall traction compared to bikes without that tendency. If we forget about the front wheel for the moment and only consider rear wheel traction it is not hard to see that squat to some extent (despite the poor riding experience it offers) is a useful modality to reduce traction robbing cyclical pressure (down force) variations at the rear contact patch. When climbing on loose or rough surfaces the greater amplitude of the pressure variation at the rear tyre contact patch of a bike with 100% AS or more makes such a bike more prone towards momentary losses of traction with the wheel spinning slightly and intermittently as the rider tries to propel the bike up the hill. Dan is well aware of the traction robbing nature of pressure variation at the rear contact patch but fails to weigh up the full significance of that insight for the climbing context. Additionally, when riding up a hill on a surface that offers good traction it is in fact the higher anti-squat bikes that will have a greater tendency to lift the front wheel. Why? Not losing any energy into squat the higher anti-squat bikes will offer greater acceleration potential and as acceleration is what lifts the front wheel of the bike that will be something that the high anti-squat bikes are more prone to.

None of this means that 100% AS bikes will offer a poor climbing experience. There will be very little suspension bob and that is a big gain. Furthermore, human beings have limited strength so we do not only need bikes offering good climbing traction but also efficient climbing. A bike should get a rider up a hill without exhausting them due to energy being lost into squatting of the suspension. Riders want forward motion not squat. Additionally, most human being are unwilling to devote more time than is necessary to making their way up a hill. A bobbing mess that seems to want to prolong your climbing suffering is experienced as a undeserved punishment by many riders. And, as it happens, squat is only one way of improving traction and not a very good way. Traction is also improved by better tyre compounds and design. A good selection of gears to deal with different riding scenarios is also a great help to balance traction and efficiency.
• 4 0
Wow. Great read. And Taj rules!!
• 4 0
All the bunnies are naeked!
• 5 0
Quarantine Mechanics
• 4 1
I feel like this is just getting us prepared for how good or how not good the Grim Donut with ride...
• 2 0
Yep, grim donut video tomorrow. I
• 4 0
Best explanation i’ve ever read on this topic. Well done!
• 3 0
Just release the new wreckoning already!
• 2 0
Here's another one by Leo Kokkonen / Pole Bicycles:
polebicycles.com/what-is-anti-squat-and-pedal-kickback
• 3 0
Its a better explanation as well. There seems to be a confusion in this article between acceleration related anti squat, and pure suspension pivot related anti squat.

For example, motorcycles all have %100 anti squat when you consider only suspension, because the chain line goes above the pivot of the swingarm. But because the acceleration is significant, and the CG is significantly lower, the applied force acts to create some anti squat on acceleration.

For bicycles, the acceleration forces are much smaller, and the CG is much, much higher, so you are never going to achieve acceleration related anti squat. The only anti squat that matters is pivot related.

Also, one thing that they don't mention is that gear selection affects anti squat as well.
• 1 0
Great informative article. That is not the frame I would illustrate in the graphic (lol), but nevertheless, a great article!!
• 1 1
Riding a hard tail is the best way to avoid anti-squat issues and to read that complicated article. Moreover, I don`t trust a rabbit that rides a bike. Rabbit are better in a hoven with mustard and polenta.
• 1 0
It’s a great article but a good rider shreds on anything does he/she not ?
‘ All the gear no idea’ is far too often applicable these days.
• 3 0
Hi Taj!
• 2 0
I need to go lie down after reading all that , gravity has got me !
• 2 0
This rabbit looks awesome
• 2 0
When do we get a video of Taj sending it?
• 2 0
And this is why I teach social studies....
• 1 0
Maybe we need more free coaster-hubs? That could possibly negate pedal feed back on high pivots etc.
• 2 0
Canfield Balance Formula rules all.
• 1 0
You know it!
• 1 1
b23.tv/av95121174
• 2 0
The use of an Orange bike for the graphics says it all.
• 1 0
Trying to read these articles makes me pick up Thrasher and go back to skateboarding.
• 2 0
i ride a hardtail, what is this magic you speak of
• 1 0
Yeah Bro, leave them having headaches with their full-suspended witchcraft
• 1 0
Well done! But I still remain with one question: How to find the Center of Gravity? Is it alvays close to my... liver?
• 1 0
If you have to explain to people that they need a new bike cause their current one clearly sucks, this would be it!
• 1 1
In professional MTB coaching, we refer to it as COM. Center Of Mass. Think about where the center of gravity is. It most certainly isn't in your body!
• 1 0
I wish I'd taken the time to learn how to properly quote and bold text before posting, dammit.
• 2 1
Sheldon Cooper rides bike.
• 1 0
Great article, Dan. Thanks!
• 1 1
Abbreviation of center of gravity is just CG, hadn't seen it referred to as CoG
• 1 1
Thank you for this pinkbike!! But if you guys could also ad videos that would awesome
• 1 0
Could you provide more details?
• 2 0
Well done.
• 1 0
Great article! I like the drawn charts
• 1 0
Great article, really improved my understanding!
• 1 0
So, the rabbit is riding a capra..... got it
• 1 0
MX + XC = E-MTB....hahaha
• 1 0
BITD, we'd just pedal more evenly. Dark days.
• 1 0
Good article! Very interesting and informative!
• 1 0
Does anyone remember Bob from the 90s?
• 1 0
Hi I’m 6ft8, but you can call me... the Giraffe...
• 1 0
Came for illustrations by Taj. Was not disappointed.
• 1 0
total load remains the same. must be younger than me.
• 1 0
Didn't realize I knew SQUAT about SQUAT!
• 1 0
Great article. Hopefully theres more EngiNERDing coming our way!
• 1 0
This is an epic article, but I'm too stupid to get it.
• 1 0
Does squat / anti squat have an effect on tyre wear?
• 2 2
Why rabbit does not ride grim donut?
• 2 1
session
• 1 0
"MX + XC = EMTB" Nice!
• 1 0
Pointy but syndrome
• 1 0
Wait...what?
• 11 13
anyone else just not care about that stuff and just wants to ride bikes
• 7 0
I just want to ride bikes too!
• 7 1
That's what hardtails are for.
• 3 1
@mtb-scotland: I'm interested in this stuff and want to ride bikes.
• 2 0
yes, but be thankful there's people who cares
• 3 2
@ismasan: take my comment as tongue in cheek.
• 1 0
@mtb-scotland: So I guess we need an "issajokestoopid" icon