Descent Planning: How to Stop Arriving High, Fast and Embarrassed

There is a species of arrival every sim pilot knows: the airport appears sooner than expected, the descent turns into a dive, the speed won't come off because the nose is down, the flaps can't come out because of the speed, and you arrive at what should be a stable final doing an impression of a meteor. Then comes the float, the deep landing, the firm touchdown - and the pilot logs another "bad landing" that was never a landing problem at all.

Here is the reframe this article exists for: the approach begins at top of descent. A bad arrival is usually decided forty miles out, in cruise, by nothing more than not doing thirty seconds of arithmetic. Learn the arithmetic and the whole bottom of your flights changes character.

The three-times rule

The foundation of all descent planning is one ratio. A comfortable descent gradient - about 3° - covers three nautical miles for every thousand feet of altitude lost. From that, top of descent (TOD) is mental maths:

Distance needed = altitude to lose (in thousands of feet) × 3

Cruising at 11,000 feet into a field at 1,000 feet: 10,000 to lose, so start down 30 nm out. FL350 into sea level: 35 × 3, roughly 105 nm - yes, an airliner descent begins over a hundred miles from the runway, which is the first thing the rule teaches: descents are much bigger than they feel.

Two refinements make the rule honest:

Add slow-down distance. The rule delivers you to the airport at altitude - but you also need to decelerate, and you cannot efficiently descend and slow at the same time (more on why below). Add a buffer: a couple of miles in a piston single, ten or more in a jet coming back from 300 knots. A working jet formula: three times the altitude, plus a mile per ten knots of speed to lose.

Correct for wind. The ratio is over the ground, so a tailwind stretches the descent. Rough correction: add a mile of TOD for every ten knots of tailwind component; subtract for headwind. A forgotten 40-knot tailwind at altitude is the single most common reason a "perfectly planned" descent ends up high.

The descent-rate version

The same geometry, expressed vertically - useful once you are established down: to hold a 3° path, your vertical speed should be about five times your groundspeed. 120 knots over the ground: 600 fpm. 90 knots: 450. A jet at 280: 1,400. This is the crosscheck that catches drift early - if the maths says 600 fpm and holding the profile is taking 900, something is wrong (usually that tailwind), and you have found out with twenty miles to fix it rather than at the final approach fix.

For pilots flying anything with VNAV, this is also the sanity check on the automation - SU5 improved MSFS's VNAV path logic considerably, but "the magenta says so" has never been a descent plan. Know what the box should be doing and you will notice when it isn't.

Energy: why high-and-fast is one problem, not two

Height and speed feel like separate quantities. They are the same quantity - energy - in two forms, and you can trade between them but not delete them at will. Push the nose down to fix being high and the speed rises; pull up to fix being fast and you balloon above profile. This is why the golden rule of descent management is: never try to fix altitude and speed in the same segment. Fix one, then the other.

What actually gets rid of energy is drag. In order of availability: reduced power (the planning tool - which is why TOD matters), then speedbrakes/spoilers if you have them, then gear and flap once below their limit speeds - each a deliberate step down the energy ladder. In a slick piston single with none of those to spare, time is the only tool, which is why starting down early is worth more in a Mooney than in an A320.

When ATC or your own lateness leaves you genuinely high: first slow down early, then descend - counterintuitive, but drag rises steeply with configuration, so getting flaps out early buys descent capability - or add track miles (an extra turn, a wider join). What you never do is accept a dive at the bottom: an approach entered 20 knots fast is statistically an approach that should end in a go-around.

A worked descent, both ends of the fleet

The C172, 6,500 feet, airport at 800 feet, pattern at 1,800. Altitude to lose to pattern: 4,700 - call it five. TOD 15 nm out, plus a couple of miles to slow from cruise to the 90-knot downwind: start down about 17 nm out. Power back to ~2,100 RPM, 500 fpm (at 110 over the ground, the ×5 rule says ~550 - matches), and re-run the check every few miles. Arrive at pattern altitude before joining, checks done - so the circuit starts calm instead of catching up.

The A320, FL340, field near sea level. 34 × 3 ≈ 100 nm, plus slow-down, plus wind. The FMS will compute its own TOD; your job is the crosscheck (does its number roughly match yours?), the monitoring (×5 rule against groundspeed all the way down), and the constraint discipline - crossing altitudes on the arrival are the profile, not suggestions. Managed descent flying is not "watch the automation": it is running the same arithmetic in parallel and voting.

One piston-specific footnote, because sim engine modelling increasingly cares: the lazy alternative to planning - staying high, then closing the throttle and diving - is also bad for the (simulated) machinery. Big piston engines dislike long idle descents: rapid cooling, and in carburetted types a descent at idle is precisely when carb ice forms unnoticed. The real-world habit worth copying is the cruise descent - power reduced modestly, 500 fpm, engine warm and working - which happens to be exactly what the three-times rule gives you when you start down on time. Good planning and good engine handling are the same descent.

Building the habit

Like everything else in this series, this becomes a skill through structure, not intention:

1. Before descent, say the plan out loud: altitude to lose, TOD distance, target rate. Thirty seconds, every flight.
2. Re-check every 3,000 feet: three-times rule against distance remaining. Early corrections are gentle; late ones are the dive.
3. Set an arrival gate: by 10 nm (GA) or the approach fix (airliner) - at initial approach speed, ready to configure. This gate feeds directly into the 500-foot stability gate you already fly.
4. Then grade it. A recorded flight shows the whole descent and approach, stage by stage - where the profile wandered, what speed you actually crossed the gates at, and what it did to the landing at the end. My FS Flights scores the enroute descent, the approach segments and the touchdown on every flight, and the pattern across a month of arrivals is blunt: the flights that started down on time are the flights that land well.

High and fast is not a personality trait. It is a missing habit - and it is thirty seconds of arithmetic away from fixed.