A grade is just a number until you can defend it. So this is the full picture of how FLARE turns a touchdown into A+ through F: what we measure, how the weights work, what we do when MSFS doesn’t give us enough data, and why the scoring is calibrated against real airline standards instead of “did the wheels stay on the ground.”
The composite
Every landing is graded as a weighted sum of six criteria. Each criterion produces a 0–100 score; the composite is the weighted average; the grade is bucketed off the composite.
| Criterion | Weight | What it measures |
|---|---|---|
| Descent rate | 30% | Touchdown vertical speed (fpm) |
| G-force | 20% | Peak G-load around touchdown |
| Centerline | 15% | Lateral offset from the runway centerline at touchdown |
| Touchdown zone | 15% | Distance from the runway threshold to the touchdown point |
| Approach stability | 10% | Stability checks at 500 ft and 300 ft AGL |
| Sideslip + bank | 10% | Asymmetric landing-gear loading at touchdown |
Two penalties come off the composite: 10 points for every bounce, and 25 points if your touchdown pitch reached the aircraft’s tail-strike limit (9° for the 737-800, 10° for the A320, just 8° for the long 777-300ER). A crosswind above 10 kt then earns a small multiplier, up to ×1.15, and the result is capped at 100.
The composite maps to a letter grade as follows:
| Composite | Grade | Read |
|---|---|---|
| 95–100 | A+ | Airliner-soft, on centerline, in zone. |
| 85–94 | A | Smooth and well flown, small things to clean up. |
| 70–84 | B | Safe and professional. Probably one criterion dragged the average. |
| 55–69 | C | Acceptable but visibly rough. |
| 40–54 | D | Firm: likely a hard touchdown or a missed centerline. |
| < 40 | F | A hard landing or way off target. Worth a debrief. |
Two things the weights aren’t doing
They aren’t averaging “how it looked”
The weights are calibrated against where the actual risk lives in a landing. Descent rate is 30% because vertical speed is the single best predictor of structural damage: transport aircraft are certified for touchdowns of 600 fpm at maximum landing weight and only 360 fpm at maximum takeoff weight. G-force is 20% because peak G is what bends gear and bulkheads. Centerline and touchdown zone are 15% each because lateral and longitudinal position together decide how much runway you have left, and runway excursions are one of the most common landing accidents.
Approach stability and sideslip are 10% each because they’re upstream of the touchdown itself. An unstable approach produces a bad touchdown more often than not, but a stable approach doesn’t guarantee a good one. You can fly a textbook stabilized approach and still flare too late. So we credit both, but the touchdown moment matters more.
They aren’t trying to flatter you
A 737-800 hand-flown to a 250 fpm touchdown on centerline at the 1,000-ft markers is a good landing in real operations, and FLARE grades it an A. A 50-fpm “butter” landing 3,000 ft past the threshold is not what it feels like: it’s a long landing that ate stopping distance. FLARE marks the touchdown zone down and the Coach calls out the long landing, however soft the wheels felt. The grade is meant to be a useful debrief, not a participation trophy.
Per-aircraft thresholds
A Cessna 172 isn’t a Boeing 737. The 172 crosses the fence at around 65 KIAS and aims for the first 200 m of runway; the 737 comes in at around 145 KIAS and aims for 240–600 m. So FLARE scores every landing against the profile for the aircraft you’re flying.
FLARE recognizes hundreds of airframes, grouped into categories from light singles and helicopters to wide-bodies and fighters, and each category sets its own:
- Descent-rate curve: the narrow-body airliner curve keeps full marks up to 100 fpm; the light-single curve only up to 60 fpm, because a trainer is meant to be flared to a whisper.
- G-force curve: light singles are held to a stricter G curve than airliners, and some aircraft (the Airbus airliners among them) carry their own.
- Touchdown-zone window: a Cessna 152’s ideal is 50–200 m, a 737’s is 240–600 m, and an A380’s is 340–750 m.
- Approach-speed tolerance: how far your speed may wander at the stability gates, as a share of Vref.
Every aircraft also carries its own Vref, which FLARE adjusts to the weight you actually land at, and, where one is published, its own tail-strike limit.
When MSFS reports an aircraft we don’t recognize from its title (it happens with some freeware and livery-only packages), FLARE asks the sim which aircraft type it is. If that doesn’t help either, the landing is scored against the Uncategorized profile, a middle-ground set of curves, and flagged as the default profile on the landing page. You’re still graded, just against a less specific yardstick.
What happens when MSFS doesn’t tell us everything
This is the part most flight-sim landing trackers get wrong, and the part FLARE specifically fixed in v1.2.
For centerline and touchdown-zone scoring to work, FLARE needs the runway: where its threshold is, which way it points and how long it is. It looks in this order:
- The simulator’s own runway data: the runway exactly as MSFS renders it, matched to your touchdown.
- MSFS ATC: the runway ATC has active, but only if you actually flew along it.
- GPS lookup against bundled navdata: your touchdown coordinates against a bundled OurAirports database (over 72,000 airports and 31,000 runways, CC0 public-domain data shipped with the app). It works offline, and it gives you the parallel-runway suffix (15L vs 15R) as a side benefit.
- Not detected: when everything misses (fictional sceneries, genuine off-airport landings), centerline and touchdown zone are excluded from the weighted total and the remaining criteria are re-weighted to a score out of 100. You’re graded only on what was actually measurable.
Before v1.2, the scoring engine awarded a default 100 on both centerline and touchdown zone when the runway was missing. The rationale was “don’t penalize the pilot for missing data”, which sounds reasonable until you realize it silently added up to 30 points to every landing without runway data. A hard landing way off-centerline in third-party scenery would grade out at C or D when the real grade was F.
The re-weighting is simple arithmetic: with centerline and touchdown zone missing, the remaining 70% (descent, G, stability, sideslip) is rescaled to count for 100%. The same rule covers anything else FLARE couldn’t measure. If the sim never reported G, or FLARE never saw the approach, that criterion drops out rather than scoring zero, and your landing in the portal shows it as N/A.
The net result: an honest grade beats an inflated one. A grade you can argue with on the debrief table is more useful than a grade you can’t trust.
The stability gates
Stability is the only criterion judged at two specific points rather than continuously: once as you descend through 500 ft AGL and again at 300 ft AGL. Each gate checks three things:
- Speed holding steady: your speed within a tolerance of what you flew over the previous 300 ft. The tolerance is a share of the aircraft’s Vref set by its category (7% for most airliners), never less than 10 kt. Being fast or slow overall is reported separately.
- Bank under control: bank angle under 5°.
- On the right path: following an ILS glideslope, or an RNAV approach in its final approach mode, within ±1.5° of the standard 3°. On a visual approach, a steady path within ±2° of the one you established, with a 1,500 fpm limit.
All three must be true for a gate to count as stable. Both gates stable scores 100; stable at 500 ft but not at 300 ft scores 70; stable only at 300 ft scores 50; neither scores 30. If FLARE never saw a gate (you started on short final, say), you’re scored from the gate it did see, and if it saw neither, stability drops out of the grade.
Why 500 and 300? Most airline SOPs require stabilization by 1,000 ft AGL in IMC and 500 ft AGL in VMC. We use 500 as the upper gate (the practical airline minimum) and 300 as the inner gate, the last point where a go-around is the obvious call.
Why this is harder than it sounds
Building a scoring engine isn’t really an engineering problem; it’s a calibration problem. The hard part is deciding where to put each threshold. Too lenient and an A+ becomes meaningless. Too strict and even good landings grade out as B. You have to:
- Anchor thresholds to real standards (FAA Practical Test Standards, airline hard-landing thresholds, FCOM Vref tolerances).
- Account for the sim’s measurement limits. FLARE takes the touchdown rate from the sim’s own contact reading and samples 50 times a second through the approach and landing, but every number is only as good as the aircraft’s flight model.
- Differentiate aircraft thoughtfully (a 757 isn’t a 737, an A330 isn’t an A320).
- Test against real landings across every category, not just the airliners.
The result is a grade you can trust to tell good flying from bad, that respects what the simulator can actually measure, and that doesn’t reward you for what FLARE can’t see.
If you want the full per-component reference (the score curves, the per-aircraft tables, the exact formulas), it’s all on the scoring methodology page in the help docs. And if you want FLARE on your PC, the pricing is exactly one screen.