Much of the aviation related instructions seem clear-cut and science-based.

For some reason landing seems heavy on art and light on science. To explain, I’m told to round out at about “yea” height, which does not register on any instrument. Even Rod Machedo calls it "the Art of Landing".

Depending on how you feel that day about the runway fixed point rushing at your face at 67 kts, “yea-15ft” could mean surprise wheels touching down (before solo), or stalling high in ground effect and plopping onto the runway (my 2nd solo).

Understanding the math in Jacobson’s flare and Rod’s expansion rate, Lindberg-reference, it seems frustratingly unstructured learning when it comes to landing.

Setting the “it’s something you get the feel for” advice aside, is there a better way of learning landing that I’m missing? Really appreciate you help here.


  1. The circuit, approach, and glide-path are fairly repeatable accurately using feedback from the instrument cluster even in variable conditions.
  2. The goal here is to establish a method to allow consistency from pilot to pilot without relying on seat-of-the-pants black-box terms such as human skill, feel, art, etc.

Edit 2

Explaining the issue further

Having seen the responses from community, it appears that we're simply asking PPLs gray-matter to act as the Height Above Terrain instrument. And everyone's OK with this despite the fact we're well trained about looking out visual illusions and [not] trusting the grey-matter in IMC.

If we're honest with ourselves, we should be able to acknowledge the absurdity in this even though it's been fine for 100+ years.

Allow for some humour: The landing skill as it's currently described looks suspiciously like practicing and getting good at guessing how many jellybeans are in a jar with the winnings being your existing nose-gear.

One can feel the stall onset a good bit before the stall-horn goes off or the actual stall happens, but we're not asking the pilots' brain to act as a stall-horn.

Yes, we can train the brain: If you've done any inversion training on the trampoline you'll notice how quickly "air sense" develops and know your exact position and attitude during a flip [the skill]. Yet, we all watch highly trained olympic athletes bumble it, and be inconsistent

The kind of guide/indication the question is seeking

If you apply the 1/60 rule during landing you can then conclude that for every ~20ft of forward travel (at 3-degrees glide-path), you're dropping 1ft in height.

This can be very helpful in knowing when and were to round out and Jacobson's Flare seems based on this. Furthermore, the paint marks on the runway are known lengths so you can tell height loss passing the start and end of a painted segment.

The trouble is: you need to have a known starting height e.g. height at threshold; and the Altimeter, or one's brain aren't the most accurate of tools for this.

The question is investigating the possibility of other visual references queues that could inform the pilot regarding ground-effect and round-out height similar to the not-so-easy-to-miss sign of buffeting before some stalls.

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – fooot
    Commented Dec 28, 2021 at 5:12

8 Answers 8


Landing an airplane is neither art nor science, it is a skill.

Like any skill requiring hand-eye coordination, muscle memory, reflexes, timing and judgement, it requires practice. It doesn’t matter how much you know about the subject or how you feel about it, the only way to get better is to actually do it.

For example, consider the act of hitting a baseball:

  • You can use science to calculate the speed of the pitch, the timing of the swing, the power and the trajectory needed to clear the fence.
  • You can use art to describe the grace of a perfect swing, the crack of the bat, the roar of the crowd, and the feeling of elation from rounding the bases after hitting a home run.

But the only thing… the ONLY thing that gives the batter a chance of hitting that 95 mph fastball is years of coaching and countless hours of batting practice.

I don’t know what kind of “deterministic guide” you think might actually be useful, but to answer your question, there might be... A few respondents here have provided some useful information that perhaps might satisfy you.

But you cannot intellectualize yourself into being a better pilot, the only way improve is through repetition.

ADDENDUM: First let me acknowledge that I haven’t answered the question as it is worded. However, as I have tried to make clear above, landing an airplane is a psychomotor skill that requires a blend of cognitive functions with physical movement. As such, you cannot separate one from the other and hope to achieve success.

In that regard I believe the question is flawed in that it would appear to be asking for a purely cognitive approach. OK, maybe not purely cognitive since the pilot trainee would at some point need to engage in the physical act of moving the controls, but it (apparently) seeks to sever the very feedback loop that helps program the brain to duplicate the motions resulting in a “good” landing while avoiding those that result in a “bad” landing. (i.e. the emphasis in the question is on NOT relying on seat-of-the-pants “feel”, skill, etc.) Our brains are hardwired to process sensory input, so to willfully disregard this tremendous benefit seems ludicrous.

While the cognitive aspect is important, it cannot stand on it's own. Improvement can only occur when it is paired with feedback from physically performing the action. The exception to this is that once the brain has a sufficient database of good and bad feedback it is possible to mentally rehearse and improve through visualization. (i.e. "chair flying") And to further the importance of practice over intellectual cognition, consider the human infant: Knowing absolutely nothing about the incredibly complex equations of motion required to program a robot to balance and walk, they are somehow able, (by physically performing the action and processing the "feel" as feedback...) to learn to balance on their feet and walk.

My personal technique while instructing was to actually emphasize sensory feedback in an attempt to instill a Zen-like oneness with the machine; the airplane as extension of the body... (For example: "Did you sense that sinking feeling right before you noticed with your eyes that we were low? Do you remember how far back you had pulled the power just before that? Do you think you may have overcorrected for the high/fast state? Do you remember the feeling of how much less you corrected for the previous one where you nailed it? The sound of the engine, the relative position of your hand? Yes? Good, then lock it into your brainpan and try to do that again!")

Finally, flight instructors need to be aware of each student’s unique learning style and adapt their teaching as much as possible to optimize results. If the goal is consistency from pilot to pilot a single source deterministic guide runs counter to that principle.

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – fooot
    Commented Dec 28, 2021 at 16:40

Yes indeed, reading Height Above Terrain with a resolution of a couple of feet is not accommodated by the classic aircraft instrumentation set, particularly not the set in aircraft used for initial training.

Learning how to fly is for a large part training the body to control the visual and motion sensory inputs. Not every aeroplane has a HAT instrument, so it is not a bad idea to get trained in making an assessment just based on out-the-window view and the feel of ground effect kicking in.

Reading an instrument takes more time than feeling and seeing motion, and time is critical when touching down. And there is the need for the human to teach the body how to land, pick up landing skills as mentioned in @MichaelHall's answer.

Indeed, autoland touches down algorithmically, with higher repeatability than when humans land. Basis of that are transmitters of signals that we humans cannot pick up, the receivers in the plane can. And now, what we do not want is that the plane can only touch down when the signal receivers are working properly. As ever, in civil aviation the three most pressing factors are safety, safety and safety:

  • Civil aviation is super-safe, but of the relatively few accidents the highest proportion is in landing. From this link:

    Nearly half of all worldwide commercial jet airplane accidents during the 10-year period from 2004 through 2013 occurred during the final approach or landing phases of flight, according to statistics published in September by Boeing Commercial Airplanes.

  • Fully automated flights with no pilots on board are not perceived well with paying passengers.
  • And humans must be able to take over in emergencies, as mentioned in the answers to this question. Captain Sully in the Hudson indeed.

Yes there are algorithms that can land an aeroplane deterministically, but well experienced skilled humans must be present to take over in an emergency and get us safely on the ground when the avionics systems are not in order.

  • 3
    $\begingroup$ That's why aircraft that do have radar altimeter accurate to a feet announce the height with synthetic voice. $\endgroup$
    – Jan Hudec
    Commented Dec 26, 2021 at 22:27
  • 1
    $\begingroup$ And when your RA or speakers are broken, you make the PM give you callouts based on baro alt. $\endgroup$
    – user71659
    Commented Dec 27, 2021 at 2:38
  • 3
    $\begingroup$ And when you don’t have a pm? $\endgroup$ Commented Dec 27, 2021 at 16:28

In general, flight instructors are not going to get this detailed with you for the practical training and tend to just concentrate on what you need to do in landing a particular airplane, but I will offer this analysis for what it’s worth.

Officially, landing is the means by which an aircraft transitions from flying to being supported on the ground by its landing gear. The techniques to do so vary from aircraft type to aircraft type, but all typically involve the aircraft making a very gentle descent towards the ground, thence contacting the ground, mooring mast, etc. on its landing gear and remaining there with the weight of the aircraft transferred from the wings, rotors, or other lifting surfaces, to the landing gear to then taxi to a parking area, shut down, etc.

For airplanes, gliders, and other fixed wing aircraft, this process typically involves two stages: an approach to a runway followed by a roundout. The approach segment involves flying the aircraft along a specific path through space guided either by visual reference to the ground or electronic instruments in the case of an instrument approach. Approaches can be straight in, circling, or involve a traffic pattern. The goal of any approach is to line the airplane up with the runway centerline, along with a vertical guidance, at the minimum airspeed needed to enter the roundout phase. This speed is typically 1.3*Vso or approx ~30% faster than the stalling speed of the airplane in the landing configuration. Approaches will either terminate in a roundout, a go-around for a traffic pattern, or a missed approach procedure in the case of an instrument approach. Go-Arounds and Missed Approach Procedures involve terminating the rate of descent to the airport or runway, applying power, and configuring the aircraft for a maximum rate of climb to either re-enter a traffic pattern or fly an instrument course to set up for another instrument approach.

If the termination of the approach and a go-around or missed approach is not needed, the next phase will be the roundout. The goals of the roundout (sometimes called flare) are:

  1. Arrest the rate of descent of the aircraft toward the runway about 1-2 feet above the runway
  2. Place the aircraft into the correct attitude so it
  3. Makes the appropriate contact, aligned properly with the runway, on its landing gear
  4. Accomplishes all this while covering the minimum amount of runway without excessive ‘floating’

All non-VTOL fixed wing aircraft are landed in this manner. The exact methods and techniques for doing so will vary between aircraft type to aircraft type and depending on the runway type, landing gear on the aircraft, etc.

That’s all great for a textbook, but human beings are three-dimensional creatures who experience life very differently than a textbook description. Pilots are artists, because flying an airplane is a function of the right side of the brain. We think about this problem spatially, taking visual and kinesthetic cues, which we can use to determine whether we are or are not performing this task correctly. And a generic textbook description is not going to cover that. How, for example, do you determine how high you are off the ground in the flare? While propwash effects are a factor in landing a small trainer airplane, it is not a factor in flying a jet. How do you anticipate the length of a flare so the aircraft touches down where you want it to? What are the flight characteristics of the airplane when landing during gusty winds, or crosswind conditions? The process of controlling the airplane itself is an artform.

Given all the variations that go into landing an airplane it’s difficult to create a deterministic models as you put it for getting good landings. But there are some recommendations.

  • On a good VFR day in the traffic pattern, try to envision the traffic pattern itself, including all legs of the traffic pattern in the sky around your airport, then make the airplane fly that path. It sounds a little crazy but it does provide good guidance for where you need to be and visualize where you need to be in order to make the airplane do what you wanted to do.

  • Good landings all start with good traffic patterns, and good consistency is the key to good traffic patterns. One of the best ways to achieve consistency is knowing what power settings for what configurations of the aircraft you will need to achieve a certain rate of descent, airspeed, etc.

  • If you’re operating out of an airport where traffic is light, or if the tower will grant this at a towered airport, asked to fly a larger traffic pattern, and make notes on a knee board of the power settings, trim, flap configurations, etc. that you need to achieve stabilized flight at certain airspeed. As an example, flying a C172 in the downwind, you not that the airplane, flaps up and trimmed just slightly nose down of the takeoff mark requires a power setting of 1900 RPM to maintain level flight at 85 KIAS. Note these metrics for all legs of a traffic pattern and then fly them that way in the airplane. If you have trouble remembering those things on the go, take a session where you chair fly in front of a cockpit poster doing all the steps of a traffic pattern, setting power, configuration, trim, etc. until you have it memorized and it becomes second nature. If you do this you will find the airplane is very controllable in all legs of a traffic pattern and any kind of adjustments, only require slight changes in power and/or trim. I especially like practicing long final approaches this way. This advice is indispensable when transitioning to a new airframe type.

  • Following the steps I described above, start practicing picturing the pattern as a transparent path in the sky and make the airplane fly alongp that path. You will find it reduces a lot of errors in position, altitude, and speed if you do it that way.

  • Base to final turns can be a struggle for students, often times causing the student to overshoot or undershoot the runway centerline. To guard against this, on base, project the runway centerline out to a point where it intersects your base leg flight path and prepare to turn inside of that corner. You should line up right on final as you roll out if you do this right.

  • Timing the start of a round out and flying the round out can be more of an art than a science. It’s a matter of hand-eye coordination combined with kinesthetic fuel for the aircraft. Rod Machado’s runway expansion method works well for light trainers, but this is just something you gain from a lot of time in practice. If you’re making bad landings on the way you’re pretty normal.

  • Another exercise during dual instruction are high speed taxi flares. Basically, you and the flight instructor will go out on a (long!) runway and accelerate the aircraft up to approximately 10 knots below its rotation speed, thence set power to hold this speed and apply back pressure on the yoke to lift the nosewheel off the runway, effectively doing a high-speed taxi wheelie down the runway. This will give you practice with two different skills. The first is gaining an appreciation and kinesthetic feel and response of the flight controls in the roundout as well as the approximate sight picture you want to see in the roundout. Note, if the airplane becomes airborne at this point, immediately ease off some back pressure and gently reduce the throttle while holding that same nose high attitude you would in the flare. This should cause the airplane to touch down again. Always plan on terminating the exercise with at least 2500 feet of paved, dry runway remaining so you can slow down to taxi speeds under positive control without risk of overshoot or loss of control. NOTE WELL: This is an exercise to be performed on dual instruction flights only. NEVER attempt this on a solo flight as a student pilot.

  • Still another exercise is to have the student fly a low approach 1-2 feet above the runway on Vref (approach speed). This, too, will provide an excellent site picture in kinesthetic experience of the roundout. If crosswind conditions exist, the student should be applying the correct cross control technique to keep the aircraft aligned on the centerline of the runway. With approximately 3000 ft of runway remaining, a proper go-around technique should be initiated to allow the airplane to establish a best angle of climb to clear obstacle at the departure end of the runway. I favor this technique with new students who have little or no aviation experience as the transition to fully landing the airplane. All that’s required to land the plane in this low approach exercise, is to smoothly pull the throttle to idle while simultaneously holding the airplane off from touching down using back pressure. NOTE WELL: This is an exercise to be performed on dual instruction flights only. NEVER attempt this on a solo flight as a student pilot.

These techniques aren’t magic but they might just make it easier to flare and land by taking some of the mystery out of it.

  • 1
    $\begingroup$ thank you for the detailed post and the exercises. It's a great answer for the human skill development part. $\endgroup$
    – MandoMando
    Commented Dec 27, 2021 at 17:34

Autopilots regularly perform landings, so yes, landing can be described as an algorithm. But since human brains are unable to perform such an algorithm, we depend on our neurons to build pathways that approximate the "perfect" solution - these pathways we perceive as experience and gut-feeling.


As mentioned in other answers, computer programs exist that can land a plane. That shows that algorithms exist to land a plane. That doesn't mean that a human can mechanically implement said algorithms.

OP compares landing with parking, saying that they apply an algorithm to park. The difference with landing a plane is that, compared to parking, there many more variables to control. The pilot needs to control attitude (in particular if there is turbulence) horizontal speed, vertical speed, centering the runway, gliding slope. And the flare, which requires reducing speed and putting the plane in the right attitude for touchdown. In addition, the plane's behaviour will vary depending on pressure altitude, plus one needs to take the wind into account; both the headwind part, that affects the glide, and the side wind, that affects the centering on the runway.

When pilots learn to fly, they are taught many techniques to deal with all the above. Eventually it becomes second nature, but it requires constant training to stay sharp. As opposed to parking, the consequences from a mistake are at a different level. And, also, different aircraft behave very differently, affecting all the above.


In my opinion, the "landing" part of getting an airplane on the runway is a developed skill. This skill is acquired and practiced within the framework of certain basic aerodynamic principles that flexibly accommodate various different types of airplanes (C172 to B767, etc).

However, the essential prerequisite for a good "landing" is the "approach," which at its core is a fairly rigid and procedurally dependent profile. If you are expecting to make consistently predictable good landings, but your approaches are not stabilized and on the appropriate profile (speed, glidepath, rate of descent, airplane configuration, etc.) you will be frequently disappointed.

If you begin your flare over your runway aiming point and you have made a "stabilized" approach getting there, a smooth and skill-based touchdown/landing will be much easier to learn and repeat.

Consistently stabilized approaches are a necessary component to master even in adverse and challenging weather conditions. As your skill in making stabilized approaches increases with practice, your ability to adapt to difficult weather conditions, changing or unexpected ATC instructions, etc., will be easier for you to manage.

A good landing starts with a good approach.


A small anecdote here. With a bit of analysis later.

I used to haul gliders up into the air mostly using a Piper Supercub. On a good day I did 20 to 40 starts, and landings of course. After a while it can become a bit boring so I invented small challenges for myself. Such as always leaving the glider in the best updraft around the field at at end of tow (the glider pilot does the disconnect). A bit of interest as the updrafts moved around during the day. Another challenge was to make 10 perfect landings in a row. Now what a perfect landing consists of can be argued about, but suffice to say I never managed to get above 3 in a row.

The old saying is that a good landing is any landing you can walk away from. A perfect landing, in my challenge, would be go well above that. It would include the exactly correct airspeed, flaring at the exactly correct height and touching down at the exactly correct place. Between landings the conditions changes: mostly wind speed and direction would change all the time. Literally one minute it could be mostly still, the next minute a thermal bubble would leave ground and induce a 10 knot tail wind. A perfect landing would cater for these differences.

In a very small and light plane as the Piper Supercub these environmental things, say a sudden tail wind has a great effect on how the plane lands. In a very large plane, 10 knots of wind makes much less of a difference of course. The old, almost antiquated, instrument set in the Supercub did not include any instruments helping in landing above the air speed indicator. And, well, the flying is "by the pants" anyway, after a while you hear and fell airspeed well enough to not really need any instruments in.

Knowing how to land without instruments (covered), was actually part of the curriculum when I learned flying gliders. Of course, they tend to be much easier to land than planes with a heavy iron in front. And I think that as long as we fly around in basically 1950 motor planes it is a good idea to practice - it can happen. Of course, in the far future, computers and instruments will take over this and no-one will be required to land manually.

So in effect, accept that landing is not a "analytical art", but a learned skill. Similar to learning how to ride a bike or to play an instrument, only a bit more involved. A good teacher can help but it is you that has to learn.


"You need to have a known starting height ... when and where to roundout ...

You're overthinking this. Whether you fly straight in from 10 miles away of roll out onto final from your baseleg, the key is proper glide angle to reach the runway. Height is only important if there are obstacles to clear. You job on final is to fly a safe approach speed (trim for 67 knots), and adjust your glide angle with power/flaps as needed.

Because a little extra speed is carried, the "round" can be one or more steps down. Round to 15 feet, ease down to 5 feet, pull power, land. This takes practice. In light aircraft, because of wind, landing cannot be done by rote. When you get good, rounding will get better.

Roundout height will depend on your aircraft, but airspeed is most important during this phase of landing. Rounding out is a distinct and separate phase of landing, properly done at approach speed.

In a Cessna 172, a good round out is done at around 10-15 feet. Remember, you are still flying the plane here, preparing for the flare and then the landing. You now have time for any aileron and rudder adjustments to be in line with the runway.

Next allow speed to bleed off, gradually raise the nose. This is the "flare". Now you're landing. Look at this as "the next step down".

If you have sufficient runway, a little "float" is fine, don't try to be perfect. "Short field landings" come later, but a slower approach carries far greater risk in variable winds. Stick with that 67 knots approach speed for now, and be consistent with it, not too fast either.

Pick the longest runway available at first and practice. It gets better with experience. Work on "setting it down" from your roundout height. As speed decreases, the plane essentially lands itself.

  • $\begingroup$ “around 10-15 feet“. At 500ft/min in the glide slop is less than 1 second margin of error. No? $\endgroup$
    – MandoMando
    Commented Dec 31, 2021 at 12:36
  • $\begingroup$ @MandoMando that would be level to 10-15 feet (as a "step"). When you start to round depends on your glide angle. This is why we all say practice, as a power on approach will have an angle of around 3 degrees, whereas power off full flaps can be around 5-6 degrees. Do not fool around with slower "short field" approaches until you get a good feel of rounding. Your second step down (as you are floating) will be to 3-5 feet as speed bleeds off and your nose comes up. $\endgroup$ Commented Dec 31, 2021 at 14:18

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .