Advanced Strafing Guide

A comprehensive guide on the mastery of strafing and longjumps in CS:GO.

Table Of Contents

About The Author

My online alias is UdNeedaMiracle, and I am a fairly well known player in the Kreedz Climbing Mod in Counter-Strike Global Offensive. Many people recognize me for my jumpstats and strafing skills, as well as the large amount of time I have dedicated to practicing them, primarily in the House of Climb 2 NA server.

I am generally considered to be one of the best at strafing in CSGO KZ, and I am currently holding several world records for jumpstats. I am also in the top 5 for no bind and no -forward bind Longjump, no bind and bind single Bunnyhop, bind and no bind backwards longjump, and am a part of the top 10 for nearly all jumpstats categories.

Apart from my jumpstats, I am also the co-creator of the Longjump Crouch Bind along with doc.mad, and I am responsible for the creation of the first (although private) Longjump Practice Server in CSGO.

Why I Wrote This Guide

In order to reach the level of skill that I have, I had to play a very high number of hours. I estimate that I have spent a total of over 2200 hours practicing jumpstats in this game. Unfortunately, I had the time on my hands to do this due to health problems. Most people do not have this kind of time on their hands, nor the desire to spend that long to develop skills in a videogame.

When I first started playing KZ, there were no longjumpers who took it seriously. The best anybody could manage was around 277 units. Now, the world record is nearly 290 units. Practicing jumpstats has come a long way recently.

At the time that I started longjumping, not only were there no serious players yet, but there was no understanding of what led to a particularly high jump. The most common advice given on servers was to try to do more strafes, or try to improve your sync. If you were lucky, you got to hear one of the best responses which you could get to this question, “just jump longer.”

I consider myself the first player to take jumpstats practicing very seriously, in the sense that I was playing the game purely to practice them, with no intentions of ever running maps in KZ, playing competitive, or doing any other thing of the sort. Until players started to come from previous Counter-Strike games, I was the only one. This means that while there were players who were better than me, there were none who were qualified to give meaningful advice. Because of this, I had to learn a lot of things about what was an ideal jump on my own through trial and error, and this took a lot of time and effort. It is the difference between practicing longjumping with no understanding of what you need to improve, and aiming to improve longjumping by practicing the physical execution of concepts that will lead to a good longjump. In my case, it took me over 2000 hours to reach an extreme skill level. Somebody else who is using my knowledge from day one will likely be able to reach my skill level in a quarter of the time.

I am writing this guide because I have lost interest in pushing my jumpstat skills any farther. The small improvements I have been able to make in the past few months show me that I am near the limit of what I am capable of, and I can not justify continuing. At least for the time being, I am done with actively trying to improve my records. However, I do not want my practice to be wasted, and I do not want anybody else to have to play as much as I did to achieve what I did. It is just a game after all.

Who Will This Guide Help?

This guide is targeted towards players of KZ who wish to improve their jumpstats for the purposes of competing with the best for world records, and for those who wish to reach a high enough level of strafing skill to be able to compete in KZ maps for world record times within a short span of time. Keep in mind that there is no substitute for real practice, but practicing with the right mindset, in the right conditions, and with the right prior knowledge will allow you to achieve great things in much less time than it would take somebody lacking these things. Practice does not make perfect, smart practice leads to consistency.

Part 1

The purpose of this section is to introduce newer players to unfamiliar concepts and to provide a deep analysis of these concepts that even experienced players can benefit from. There will also be brief bits of information about the history of the evolution of jumpstats in CSGO. If you are above average in skill or have played for a long time you may wish to skip to Part 2.

Before Even Playing

Jumpstats are something that require you to be in top form to improve your records after a point. This means things outside of the game are going to impact how you do in the game. It is important to be hydrated while you play and to not play while hungry or tired. Another helpful tip is that staying physically active is going to help you play a bit better. Maybe it all sounds a bit ridiculous, but if you want to play your best, I suggest focusing on staying hydrated, eating before you play, and going for a brief ten to fifteen minute walk before you sit down to play.

This applies to more than just practicing jumpstats, however, jumpstats require so much mechanical skill and input, you really have to do everything you can to keep yourself in good performing condition. These are small changes you can make that will boost how well you play.

Regarding Equipment And Setup

Does your equipment and setup alter potential performance? The short answer is no. You do not need to have the best mouse, or keyboard, or have a 144hz refresh rate monitor, or anything of the sorts to become the very best. The only thing that matters is that your setup is comfortable. Many of the best players are using membrane keyboards, a medium tier wired mouse, and a Steelseries QCK+ or QCK heavy mousepad. Some people are under the impression that a mechanical keyboard is ideal for strafing, although I would actually argue that a membrane works much better.

I spent a lot of time worrying about things that turned out to be trivial, such as how I rested my arm on my desk, or how good my keyboard was, and other things of the sort. What I learned after reaching a high skill level is that I could perform basically equally on any setup so long as it was comfortable.

In-Game Settings

First and foremost it is important to set the correct rate settings for 128 tick, which is the primary tickrate servers running KZTimer will be using. Exactly how I have everything set up can be seen in my autoexec which I have linked at the end of this guide.

Secondly, keybindings need to be considered. Generally in KZ you want to be using mousewheel up or down to jump, and the opposite key to duck for the purpose of Countjumps. For those who wish to best apply their jumpstats skills to running maps, I suggest you use the same jump key to practice jumpstats that you run maps with, which is generally mousewheel down. For those who wish to improve at jumpstats primarily, I suggest picking between using Mouse1, Mouse2, Mouse4, or Spacebar to jump.

The reason for this is that the only working version of the crouch bind is an alias, and alias’ can only be bound to keys which have a release. If you want your non bind skills to transfer over to using the bind, it is best to practice on a key that can also be used with the bind. Mousewheel can not be used with the bind for this reason.

Next, you must consider mouse settings. I think as a general rule you will want to have Raw Input turned on. This is because we want to eliminate inconsistent forms of mouse acceleration that are present as a result of how your operating system interprets mouse input. Using acceleration as an in-game setting is a preference, although for what it is worth, none of the top ten strafers or KZ players are using mouse accel as far as I am aware. I think it is generally bad because strafing requires an even movement speed of the mouse and an even travelling distance of the mouse back and forth several times in under one second. Human error is exaggerated by the use of mouse acceleration.

In regards to mouse sensitivity, the sensitivity you use for longjumping is of very little importance. I have done amazing jumps on 800 DPI 1 sensitivity, and done amazing jumps on 800 DPI 2.6 sensitivity, as well as many other sensitivities in between. The only thing that matters is that you pick a sensitivity you can play KZ on and stick with it. One of the biggest mistakes I have made is changing my sensitivity several times. Changing your sensitivity can only ever be a bad thing, and any time you are unable to do things and want to blame your sensitivity, it is likely that you just lack the required skills. I would recommend a sensitivity that makes moving your mouse across the entire usable space of your mousepad no less than 200 degrees turned in game and no more than two full 360s.

Types Of Jumps

The Crouch Bind

The crouch bind is as follows:

alias +LJ "+duck; +jump; -forward; +klook"
alias -LJ "-duck; -jump; -klook"
bind key +LJ

What this bind does is crouch at the same time you jump. This gives a player an additional 1.1 units of height, which leads to an additional three to four units of distance on jumpstats. In addition to that, the bind cancels out the forward key as soon as you jump. It is impossible to gain speed if your forward key is held in the air, so this gives players an extra level of consistency that is otherwise impossible by removing the need to perfectly time the release of the W key when jumping.

The idea to bind -forward to your jump key belongs to the player doc.mad. I came up with the idea to bind crouch and jump together to improve my Longjump distances after talking to one of the House of Climb Co-Owners, Chuckles, about how height could impact distance travelled. Conversations between doc.mad and I led to the creation of the original longjump bind, which was used until a patch to CSGO changed air accelerate values while a player crouched, making it no longer effective how it was written. After that time, players had to switch to the alias version of the bind. The first person to write an alias version of this bind was a user by the name of Logos on the House of Climb forums, who also had the idea to include klook in the bind, which is a similar command to -forward.

Problems Introduced By The Crouch Bind

A massive problem that the crouch bind has introduced is the reduction in the potential skill gap for both jumpstats and KZ maps. By making it possible to consistently reach the max height, some jumps in maps have become extremely consistent, making times on those maps possible that were once highly unlikely to be achieved. Relevant to this guide, however, is the massive decline in the skill gap it has caused amongst Longjumpers. Releasing W is the most important part of any Longjump, because releasing W too early can lead to low prestrafe speed, and releasing it too late prevents you from gaining speed in the most critical part of the jump, the first strafe. The bind has removed this aspect of jumping, therefore allowing many players of several different true skill levels to jump at a similar actual level, and making the distinguishment between the “good” players and the “very elite” players less pronounced.

The bind has also caused a massive block in the development of player skill due to the release of W becoming automated. Many players are quick to rush into using the longjump bind when they start playing because they are unhappy with their relative lack of skill compared to other players, not realizing that the best players in the KZ mod have been playing for years and years to get to their skill level. Chasing immediate satisfaction and success, players turn on the bind, see the results that they desired, and then continue seeking longer distances by continuing to use the bind, under the impression that they could never beat their record again if they did not continue using it.

This action and mindset leads to a decline in rate of progression for a few reasons. The first problem is that it eliminates the carry over into running maps. When running maps, using the version of the longjump bind that includes -forward and +klook is suboptimal, because the type of jumps that benefit most from the crouch bind are high jumps that require the use of the W key to get over the edge of upon contact with it. When players attempt to use the bind for actual longjumps or skill jumps within maps, they have no correct muscle memory for releasing the W key, and therefore are unable to easily complete the jump.

The problem that this bind causes in practicing longjumping stems from the same root cause. Players lack fundamental skills in releasing W, which seems unimportant to many because they think that the bind will always just take care of it for them. That much is true, the bind will always ensure a perfect W release. However, learning to manually release W correctly benefits the timings and airtimes of the first and second strafe, and helps players complete the release of W and every subsequent strafe as an individual action, rather than blending them together. This leads to a more ideal flight path. Players who are so quick to use the bind are robbing themselves of progress because they are too caught up in immediate satisfaction rather than long term results.

The pLekz Height Trick

The pLekz Height Trick is a method of increasing Multi Bunnyhop distances discovered by a House of Climb Admin, pLekz. He made the discovery that if the last bunnyhop a player does before strafing in a multi bunnyhop is done from a platform higher up than the rest of the bunnyhops in the sequence were done from, the player receives additional height that is otherwise not possible, leading to increased distance travelled.

Fundamentals Of Travel Distance

In Counter-Strike Global Offensive, it is important to understand how your player moves without airstrafing. If you run to 250 U/S speed and then jump, you will travel approximately 217 units. If you hold crouch before landing, you will travel 225 units. If you use the crouch jump bind or manually crouch before takeoff, and then crouch again at landing (Double Ducking), you will travel 227.3 Units. This is a clear illustration of the minimum amount that the crouch bind helps. The potential that the crouch jump bind helps is not a fixed amount, however. The crouch jump bind works by adding time in the air. A player who uses the additional time they get in the air more effectively will travel farther as a result. A player travelling at a higher speed in the same duration of additional airtime will go farther than a slower moving counterpart so long as their flight path is a constant.

I first realized this was the case when the player eightbO, one of the top KZ players, widely regarded as having the most technical skill in all of KZ, pointed this out to me in regards to general strafing through the use of analogy. While it seems so simple, his analogy was that if two cars are travelling the same amount of time on a linear and fixed path, but one of them is travelling faster than the other, that car will cover more distance. The exact same is true in regards to movement in KZ.


Prestrafe is a method of acquiring more prespeed. Prespeed is simply the speed you have before you jump in regards to any jumpstat. Prestrafing is the use of the forward key + the left or right movement key combined with a gradual turn of the mouse in the appropriate direction to gain speed on the ground. The maximum prestrafe possible is 276.145, although anything above 274 is acceptable. Prestrafe is the single largest contributing factor to the distance a player travels. A jump on 270 prespeed will never be as good as a jump on 276.145 prespeed where everything else was the same. If a player prestrafes to 276 speed and double ducks, they can reach as far as 259 units, compared to 227.3 when not prestrafing.

If you are unsure about how to prestrafe, many video tutorials are available that would be of more help to you than what I can offer in the form of text. Also, many players would be glad to help you learn in any KZ server you go to.

I am frequently asked what the best way to prestrafe is when using a block. The answer to that question is without a doubt the way that gets you 274-276 prespeed the most often. There is no difference in effectiveness among different methods of prestrafing.


The idea of airstrafing is that you are using the left and right movement keys in the air, synched up with left and right mouse movements to collect speed. Most players already understand what it is, but most do not truly have an understanding of why we are doing it. The most ideal flight path is the most straight one. We are obviously airstrafing to increase the distance we travel, and we are doing it through collecting speed. So, in theory, the best way to collect speed is to simply turn in one direction in the air at the optimal rate for speed gain. The problem is that doing that would take us way off the ideal flight path and we would actually travel nearly no distance from our starting point. We strafe in alternating directions in rapid succession collecting speed in order to continuously cross our player model over the most ideal flight path with the intent of travelling in a straight line. This is why a seven strafe jump is going to have the potential to travel farther than a five strafe jump, for example, because you are staying closer to the overall ideal flight path, and you can collect more speed with less deviation from the flight path.

Jumpstats Components

Part 2

The aim of this part is to introduce the advanced concepts that contribute towards great strafing based on what we can see in the provided in-game stats

Role Of Sync

Sync is a very important component of a jumpstat on a basic level. A person who can only sync for 40% of their airtime is going to have a much harder time going as far as somebody who can sync for 80% of their time in the air unless the person with the higher sync has a much worse flight path. Although, sync is not a true indicator of how good a jump should be. I am known for having extremely high sync, and have even managed 100% on 9 strafes before. However, people who can not sync as well as I can are still able to go farther than me because they are stronger in other areas.

Sync has a strong diminishing return due to one simple and often overlooked fact in CSGO, which is, the more speed you already have, the lower the rate of speed gain is. While this is not a problem experienced on early strafes, where you still have not started to approach the fastest movement you will attain during the jump, it matters on the last few strafes. On the last couple of strafes you perform, while they still drag the average sync down if they are out of sync, it has a lower overall impact both to speed gain and distance travelled.

This occurrence in the CSGO Source Engine can be observed when comparing longjumps with no prestrafe versus longjumps using prestrafe. When the starting speed is 250 U/S, I have managed to get as much as 365 max speed. When the starting speed is 276 U/S, I have only managed around 383 max speed as my personal best. If you took 365 max speed and added the 26 units per second that the prestrafe contributes however, you would reach 391 max speed. 391 Max speed is an absolutely insane value of speed that only one player has ever surpassed. The level of consistency with which I can get 358 or higher max speed without prestrafe, which would be 384 max speed with prestrafe, shows pretty clearly that the less speed you have, the more you can gain. The reason for this is that I have spent almost no time ever practicing no prestrafe longjumps, and still gain higher comparatively when doing them.

The reason that this quality of the CSGO version of the source engine leads to a diminishing return in sync is because of the factor that sync is actually related to, which is potential speed gain. Everybody has always known that more sync is better, but not many people realize precisely why this is the case. The higher sync you have, the more of the potential speed you will have already gained, which means that additional sync percentage will contribute much less to total max speed. It is for this reason that sync stops being important once you can do 85% consistently. I cover more on potential speed gain later.

Not only is more than 85% sync significantly less beneficial, it is considerably more difficult to improve at. The reason is not necessarily that humans can not sync their left and right hand movements perfectly. You can, just not consistently. The really difficult part comes in the last 5%. Almost everybody will achieve 95% sync within the first 6 months of seriously practicing in KZ. Most people will never hit 100% though. The reason for this is overlapping key presses. Most people are only focusing on alternating hitting their keys, but what the best results will come from is also focusing on releasing one strafe key almost entirely before depressing the other.

This is part of the reason why it becomes so much more challenging to sync higher amounts of strafes. The more strafes you are doing, the more difficult it will be to not overlap key presses while still being fast enough to complete all of your strafes. Any time where keypresses overlap you have dead airtime, where it is not possible to be in sync because you can not move a mouse in two directions at once, and where your character can not gain speed, but will only maintain the speed they already are travelling at.


Airtimes are a valuable tool in determining why certain jumps are not as good as others. First things first, what are good airtimes? Good airtimes are even airtimes. The actual number of airtime is not necessarily important. There is a point where airtimes are too low, and you could not have possibly gained enough speed on a given strafe no matter how much you moved your mouse. There is also a point where airtimes are too large, and too much airtime was dedicated moving away from the ideal flight path. Keep in mind, the ideal flight path is straight. Strafing is moving away from the ideal flight path. If I have to give a concrete figure as to what a good airtime per strafe is, I would say 10% to 12% depending on the number of strafes done.

The problem with suggesting that airtimes should always be even causes is that it is not true when performing low amounts of strafes.. If you are not doing nine strafes or above, perfectly even airtimes will not work out for you. The reason is that the airtime on each strafe will without a doubt end up being too high, and you will spend too much time strafing away from a straight line of flight. In the case of doing less strafes, aim for the 10-12% range on each strafe and pile up a large amount of airtime on the last strafe, without doing much mouse movement before landing. You will sacrifice sync on the last strafe but you will gain distance because of it.

The ratio of airtime to gain is a valuable tool also because we are able to see how much mouse movement a person is using and how ideal their movement is. As a general rule, gain should always be higher than airtime except on the last strafe. It it possible to strafe too wide or too fast with the mouse and increase how far away you go from the ideal flight path without actually gaining additional speed, however. Be aware of how wide you strafe.

Gain And Loss

There is very little to say about gain as a stat. More gain is better in every case, and the more you can gain early in the jump, the better. What actually controls potential and actual gain will be discussed later, as it can not be represented purely as a stat.

Loss is always a bad thing, however, not as bad as some people like to imagine. Loss is bad in the sense that the potential of the jump’s distance was not reached due to travelling at a lower speed. The earlier in the jump the loss is, the more damaging it is. This is the case due to the car analogy referenced earlier.

A big mystery in KZ has always been the question, “What causes loss?” The answer to that question remains unresolved in a complete fashion. What is known for sure is that when two strafes overlap and the mouse is moved in the wrong direction while strafing direction is being changed, you will get loss. Simplified, a slow keyboard hand and a comparatively faster mouse hand results in loss. It is hard to determine what other factors contribute to loss.

A jump that has loss will not be any different from a jump where all other variables are the same where the strafe prior to the one with loss simply never gained the lost speed.

As long as a player understands gain and loss on a per strafe basis, it is not necessary to discuss max speed for a jump. More max speed always increases the potential for the jump to be far.


Height is a reflection of at which point in the jump the player used the crouch key. The minimum possible on a normal jump is 55.8, and the maximum possible is 66.0. A player using the Plekz Height Trick will be able to achieve slightly higher than 66 height on a multi bunnyhop if they use the crouch bind for their final hop and hold duck through the entirety of the jump.

55.8 Height occurs when a player runs and jumps, regardless of whether or not they crouch. 56.9 Height occurs when a player ducks when they jump and then releases the duck key immediately, regardless of whether they duck again at the end of the jump or not. 65.9-66.0 height occurs when a player crouches as they jump and then continues to hold crouch throughout the jump. 64 Height occurs when a player does not jump at takeoff, but immediately ducks after jumping and continues to hold duck throughout the entirety of the jump.

Every value of height between 55.8 and 66.0 is possible to attain by altering the timing at which you press duck in the jump, but only certain values of height are useful. A player who reaches 56.9 height by ducking before takeoff will actually get more time in the air. Above 56.9 height provides no benefit (except in the case of multi bunnyhops where a player uses the Plekz Height Trick). To achieve more height than that a player must hold the duck key while in the air, which is never a good decision. Holding duck in the air drastically lowers the rate at which you can gain speed. Therefore, it is never beneficial in regards to travelling distances on a horizontal plane to have above 56.9 height.

Likewise, it is not beneficial to jumping higher on a vertical plane. A player with 56.9 height is jumping higher than a player with 55.8 height by using the mechanic of ducking before takeoff. However, a player with above 56.9 measured height is not going any higher than a player with only 56.9. The plugin simply interprets the character as having more height, however the distance between the character model’s feet and the ground are identical in any case.

When Is The Proper Time To Release W?

Every player who starts in KZ is immediately made aware of the fact that W is rarely ever necessary to be held while in the air, and it is not possible to gain speed while strafing if you are holding W unless you are strafing sideways using W and S. However, many players have told me that they never learned when the proper time to let go of the W key was.

Many players will say that it is best to release W at the same exact time you press your jump key, but I would disagree. While that technically should work, humans can not do that consistently enough. To get the most distance out of your jumps, I am of the opinion that you should focus on releasing W a fraction of a second before pressing the jump key. With practice, this will ensure that you get the most W release checkmarks possible. Unfortunately, the downside to this is that you will lose some of your jumps by releasing it too soon and not getting the full prestrafe. However, the amount of perfect 275+ prestrafe jumps you get with a W release checkmark will be overall higher.

Part 3

The purpose of this section is to talk about what overall leads to the farthest jumpstats based on what we can not necessarily see in stats.

Does Strafe Style Matter?

There are many different styles of strafing: circle strafes, diagonal strafes, horizontal strafes. The question is, does it matter? Many players who are new worry a lot about their direction of strafing with the mouse. There is absolutely no penalty for vertical movement in jumpstats or strafing. You can go just as far. What is true, however, is that circular strafes are much harder to sync generally speaking. Although some players may prefer them, I have not seen any truly skilled player who specialized in using circle strafing.

There is always going to be some form of diagonal mouse movement in strafing, as nobody can hold a mouse perfectly straight while rapidly moving it. As you practice and become more skilled, it will become less difficult to strafe in a mostly horizontal fashion. Do not worry too much about this as it is something that simply comes with practice.

Two Types Of Strafing

This is something that I have never discussed with another player in the past, and I do not know if other players consciously change how they strafe for different circumstances. However, what I can say for sure, is that there are two types of strafing that are visually nearly identical and difficult to distinguish between that work better in different circumstances.

It is all in how the mouse is moved. Even more than that, it is about how you think about strafing as a concept. The first way to think about strafing is as snapping a mouse between two fixed points in virtual space in-game. The second way you can think about it is gliding the mouse across the mouse pad smoothly in alternating directions. The first style of strafing is a more rigid way to strafe and the second style appears much more smooth visually.

There are different circumstances in which these two methods of strafing work better, but I first must discuss potential speed and actual speed, and how gaining speed actually works as far as we as players can understand it.

Potential Speed Versus Actual Speed - How Gain Works

Speed gain is impacted by several factors. First it’s important to understand the difference between potential speed and actual speed.

Potential Speed is how much speed you could have had if you executed everything perfectly.

Actual speed is the amount of speed you actually got based on how you perform.

For example, if a strafe has 80% sync, 10 airtime, and 13 gain, you actually gained 13. However, just looking at the sync, you can already tell you could have potentially gained more. This means that without changing your flight path in any way, just by increasing your manual input accuracy, you could have gained more speed. Gaining more speed on the same flight path means that without a doubt you would have gone farther, because every other variable is the same.

Potential gain naturally exists through the faults in your jump. Nobody is ever going to gain the absolute maximum they could have because nobody is ever going to have a jump without flaws.

Actual gain is controlled by a few things. First and foremost, sync. Higher sync will always mean higher gain. Secondly, the speed and width of your strafe. A wide strafe has the potential to gain more speed than a strafe that is narrow. The speed at which you execute the wider strafe will determine what you get out of it in terms of both distance and speed, though.

All servers using the KZTimer plugin have the server setting, SV_Airaccelerate, set to 100. Airaccelerate is what determines the rate at which speed can be gained. If the speed of your mouse movements is too high for the amount that you have moved based on the AA value, you will not gain additional speed for strafing that wide.

This means that to strafe wider, you have to strafe at the ideal rate of speed. The ideal rate of speed is the maximum speed at which you can move your mouse without exceeding the limits of the AA setting. Even if you can figure out how fast you approximately need to move, it becomes harder to execute that rate of movement over longer distances of mouse travel. Therefore, the wider you strafe, the harder it becomes to benefit from it. Furthermore, the wider you strafe, the more time you will spend away from the ideal flight path. Because of this, the question arises, how wide is too wide, and how fast is too fast? The opposite question also has to be pondered, how narrow is too narrow, and how slow is too slow? Unfortunately, I have no system of measurement by which I can tell you what these rates are. Through trial and error, you will quickly be able to tell what works best.

Another contributing factor to potential speed gain relates to what I previously talked about above, strafing smoothly. While I can not support it with any evidence that is not purely anecdotal, I have always felt that a smooth mouse movement generally could generate more speed. The reason I feel this way is due to the concept of dead airtime, where you are neither gaining nor losing speed. Earlier I talked about this in the case of overlapping key presses, but there is another area in which dead airtime can exist.

Dead Airtime

To reiterate, dead airtime is time in the air where you are not changing your speed in any way. I discussed earlier how this can exist as a result of actions taken with the keyboard hand, but this can also exist as a result of mouse movement.

At the point in which a strafe reaches full extension, as in, a left strafe reaches its left-most point, or a right strafe reaches its right-most point, there is an extremely small and immeasurable point of dead airtime. As the mouse hand transitions from strafing in one direction to strafing in the other, there is a very small period when there is no actual movement occurring. This is where dead airtime exists as a result of mouse movement.

It is for this reason that I feel like a smoothly executed strafe could gain more speed. It appears that a strafe that focuses on a smooth movement of the mouse will spend less time at full extension and therefore have lower dead airtime. While I have always strafed using purely movement from my forearm, I hypothesize that the smoothest movement and most ideal strafing would be from a combination of using wrist and forearm motion. As an example, if a player is strafing towards the left, as they approach the full extension of their strafe, they can continue to move left with their forearm while starting to move towards the right by pivoting at the wrist. However, this would take an intense amount of focus and practice to perfectly execute and would likely only provide minimal benefits.

When To Use Snapping Strafes Versus Smooth Strafes

While I do not hold longjumping world records, I do hold multi bunnyhop and weirdjumping world records. If you read this guide to its entirety, you will understand how the strafing mechanics of weirdjumps are not that different from longjumping, but multi bunnyhops are a totally different thing, for one small reason. What I have discovered is that for some reason, multi bunnyhops tend to go much farther if you strafe in a snapping motion. Weirdjumps, longjumps, and countjumps go much farther if you use a smooth mouse movement.

In addition to that, multi bunnyhops go farthest on smaller strafes, where longjumps, weirdjumps, and countjumps tend to go farthest on medium to wide strafes where more speed is collected. The most logical explanation I can come up with for this falls back to the principal I introduced earlier about speed gain in the CSGO version of the Source game engine. The more speed you have, the slower you can gain speed. Even beyond that, it appears that when you have a very high amount of speed, such as in multi bunnyhops, the optimal angle formed between left and right strafing should be smaller to gain more speed. Try strafing very wide on a multi bunnyhop and you will gain about the same amount of speed as if you strafed very narrow. Although, you should be going farther if you are strafing much smaller,because you will have less deviation from the straightest flight path.

Strafing Evenly

It is imperative to strafe as evenly as possible, excluding the last strafe (I address the last strafe in its own section). To strafe evenly means making airtimes even and consistent, making mouse movement width even, and making mouse movement speed even. If you do not strafe evenly, you will not get as far as somebody else who does that is otherwise of equal skill.

The First Strafe

As I have explained, speed gained early in the jump is the most valuable, because you will have that speed for the longest period of time. For this reason, it is commonly advised to have a large first strafe. Unfortunately, people are confused by what exactly this means. Many players are only looking at their gain on the first strafe, and not how it relates to the airtime they have.

It is important to make a distinction between size of a strafe, and duration. The size of a strafe is what is controlled by your mouse hand, and the duration of the strafe is controlled by your keyboard hand.

It is beneficial to have a large strafe, but not a long one. Having both a large first strafe, and a first strafe with high duration, is seriously harmful to the distance of a jump. The reason for this is that you are collecting speed while strafing away from an ideal flight path. The more speed you can gain, the better. However, when you are gaining a lot of speed and strafing away from the ideal flight path, you are travelling farther off course. The farther off course you go in one strafe, means the longer your next one has to be to get you back on course. Even if you do make it long enough to get back on course, the damage is already done, because you spent so much time going the wrong way.

Try to gain as much as you can in relation to your airtime on a strafe, not as much as you can overall.

While I feel like players who read this guide will already understand this, since it is such a common question, I will address it anyways. Many people struggle to get high sync on their first strafe. Aside from travelling in the wrong direction, which is highly unlikely because your prestrafe goes in the same direction as your first strafe, the sole contributor to low sync is a late W release.

Your first strafe can not technically start in the eyes of the plugin and game until you let go of W. The low sync on your first strafe lowers actual gain, and therefore is the most detrimental occurrence that is possibly seen in jumpstats. This is a big part of the reason I earlier called the W release the most critical part of a jump.

The Last Strafe

The last strafe is a commonly misunderstood strafe. Many people have asked me about how to improve the gain on their last strafe, or how to improve the sync. While you want to be in control of the last strafe, those two aspects are completely unimportant past a point.

What most people fail to consider is that there are two ways to think about the last strafe. You could continue to strafe, and collect a little bit more speed, although you are already at the end of the jump so extra speed gained is negligible. Alternatively, you could stop strafing and continue to fly through the air in a straight line without strafing. It is better if you just stop the strafe early, executing more of a half strafe than a full one. The benefit of the slight speed gain in the end of the jump is almost never worth it.

You can test this for yourself by trying three-strafe jumps. Obviously the first and second strafe must be much faster than the last one for reasons described earlier in this guide. For the last one, try to have mouse movement throughout the entirety of the strafe and see how far you get. Then, try a few times where your third strafe is more like half of a strafe. The reason this works better has to do with how the takeoff point relates to the ending point. Remember, we are trying to strafe along a straight line.

This concept is not something you will have to worry a lot about once you become more advanced and can land nine or ten strafes, because when doing so many strafes, you will never have a long last strafe.

Takeoff Direction When Jumping From A Longjump Block

When jumping off of a longjump block or any straight-edged block, you must take into consideration where you are facing when you jump off of the block. The correct way to be facing is nearly straight off the block, at a slight angle away from facing directly off. It is hard to give an exact figure as to how far you should be looking away from a straight line off the block because this changes depending on how wide you intend on strafing. Focusing on this is only important to make sure you cross the gap in an ideal fashion. There is no benefit to this for pure distance so long as your strafing follows the principles described in this guide.

You can visualize the straight path that a jump should be executed across as the midpoint of your first strafe if you correctly jump off the block. You can think of the direction you are facing as you step off the block as an indicator as to approximately where your second, fourth, sixth, eighth, tenth, etc strafe should be.

Part 4

In this section I will discuss how to train jumpstats to improve as fast as you possibly can.

Grouping Jumptypes To Train Efficiently

Something most people who are beginning in KZ fail to understand is that the only main difference between any two given jumpstat types is the prespeed. The speed that you start at is the basically sole contributor to making them different. Some jump types do require more mechanical skill than others, particularly longjump and countjump being the hardest. However, once a player has mastered the execution of the technical aspects of those jumptypes, and applied the different types of strafing I have described above, skill between jumpstats can have a huge carry over. For example, the way that you strafe between countjumps, longjumps, and weirdjumps is largely the same. This is because their prespeed caps are extremely similar. Countjumps and longjumps have a relatively higher mechanical input requirement than weirdjumps. Therefore, anybody who practices lots of longjumps and countjumps, will naturally be able to get a good weirdjump. It is unnecessary to train weirdjumps at all until you are ready to break a weirdjump record. For the most part, your skills from countjumps and longjumps will carry over.

This same idea can be transferred over to the comparison between single bunnyhops and drop bunnyhops. They both have the same prespeed cap and relatively similar technique, so you only ever need to train one of them to be good at both.

Players only need to practice longjumps, countjumps, bunnyhops, and multi bunnyhops to master jumpstat strafing. Most of a player’s time should be spent on longjumping and multi bunnyhopping because they are the most drastically different styles of jumping.

How Much Should A Player Practice?

When trying to become great at strafing, it is more important that you practice well and practice often than to practice a lot. I suggest everybody practices at least thirty minutes every day, at least five times every week. I also recommend that you only play in the ideal conditions I stated above, where you are hydrated, have recently been physically active in some way, and are not hungry or tired.

If your hands are tired from a previous day of practicing, or you are unable to get warmed up within thirty minutes of sitting down to play, you are probably never going to manage to get warmed up for that day. In this case I recommend players take a day off. Further, whenever your hands start to get tired, it is time to quit. Do not continue to practice after becoming fatigued. You will only make yourself worse by practicing in a way that can lead to the development of incorrect muscle memory.

I would say that the ideal amount of practice is approximately 1 hour and 15 minutes of longjumping per day and twenty minutes for the other three primary jumpstats, bunnyhop, countjump, and multi bunnyhop. I would suggest that this amount of practice should be applied six days per week at most, and three times per week at minimum. However, as I already stated, it is much better to get less practice every day than a lot all at once. If you can not get this much practice every day, try to get in whatever you can.

Start From The Bottom And Build Strong Fundamentals

A lot of players in KZ come in and try to emulate what the best players are doing right away when they are unprepared to do these things. Before practicing strafing at all, a player should be able to reach at least 274 prespeed on longjumps consistently. Once a player has done that, they should practice with only three strafes for several days or possibly weeks until they are able to get a successful W release nearly every time (indicated in the jump panel by going to !options, and choosing to enable the speed panel and the advanced jump panel). Then, move on to five strafes. Practice five strafes until you feel like you have absolute mastery of it, and then move on to six. Master every number of strafes before moving on to the next one. You cannot set out to build a house by laying the bricks for the top floor first. You must first build a base on which to lay those bricks. The same applies to strafing. If you want to be good at seven strafes, you have to be able to do the first six of them perfectly. You do not have to do a high amount of strafes to be able to go far. You can do more than 270 units with a longjump on just three strafes, and more than 280 units with just 4. While some people like to try to do 11 or 12 strafes, they are not getting any farther than somebody else who is doing only eight, but doing them perfectly.

Should You Train Both Prestrafe Directions?

For the purpose of just improving in jumpstats, the answer is no. There is no benefit to training yourself to be able to execute jumps from both sides. However, anybody who wants to be able to transfer their strafing skills into map running should practice both directions evenly on both longjump and multi bunnyhop. There is no need to practice both sides on other jump types if you do not wish to.

It’s also important to know that while you may feel like your jumps are particularly bad on some days, you may still be able to jump normally on your non-dominant prestrafe side.

Should You Train Backwards And Sideways?

There is no use in maps to being able to strafe sideways or backwards. At this time, there are no officially kept leaderboards for either of these techniques. The only purpose for training them as it is right now is to impress people. Considering how easy it is for somebody who is already good at jumping normally to learn them, there is no merit in practicing them regularly.

How Long Does It Take To Become Good?

Do not become discouraged quickly when pursuing skills in strafing. It takes a large amount of practice and time to be truly skilled. With the help of this guide and regular practice, you will be able to become very good within months to a year.

Something I frequently hear is people saying they are having a bad day strafing, or that they were performing very well for a day or two, and then they have several bad days in a row afterwards. What is happening here is that people are under the impression that the days where they were playing great are the measure of their skill level. The reality is that they were having a lucky day, and their normal skill level is what they are seeing on their so called “bad” days. Do not let this discourage you. It takes a lot of time to be truly consistent to the point where you are better than the majority of other players even when you are not “on fire.”


It took me well over 2000 hours practicing several jumpstats types to get to my level of skill, but it did not have to. The only reason it took so long was because I had no idea what I was doing when I started, and there was nobody to show me. Hopefully, as a result of this guide, others will be able to do what I have done in just a couple hundred hours instead. I hope this guide increases the competition in jumpstats in KZ and brings some of the good strafers up to a “great” level. Thanks to everybody who reads this, and I hope it helps.