No planet has ever actually moved backwards. Retrograde motion is a line-of-sight effect: Earth and the planet travel the same direction on different orbits at different angular speeds, and when Earth overtakes on the inside, the planet appears to slip backwards against the background stars. Drag the timeline below and draw that loop yourself.
The dashed line is the sightline from Earth. Wherever it meets the outer ring is where the planet appears in the sky.
Now the right. Wherever that sightline meets the background stars is where we see the planet. When Earth overtakes a superior planet on the inside — or an inferior planet overtakes Earth — the sightline starts swinging the other way, so the track doubles back and draws a loop. The two points where the colour changes are the stations: the moments when the rate of change of longitude passes through zero.
Ecliptic longitude across, ecliptic latitude up, both at the same scale. Where the line changes colour is a station — the instant the direction flips.
The same positions, with the origin moved from the Sun to Earth. The dashed circle is the deferent, the orange circle the epicycle. The red curve is the path the planet really traces as seen from Earth.
When you switch to Earth-centred, not a single planet is recomputed. The only thing that happens is that the whole picture is translated so Earth sits at the middle. Ptolemy's epicycle is not a device from a rival theory — it is the shape of that subtraction: planet minus Earth.
This is why the epicycle model can reproduce retrograde motion perfectly well: it is the same geometry as the heliocentric one, with a different origin. Heliocentrism did not win because epicycles gave wrong answers. It won because once you move the origin, every epicycle disappears and all that is left is each planet on its own ellipse.
What is drawn here is an idealised epicycle model, geometrically equivalent to the heliocentric one. It is not what Ptolemy actually used — the Almagest adds eccentrics, equants and a good deal more that is not implemented on this page.
Each band is one retrograde episode. Click any of them and the three panels above jump to that loop; click the empty space to move the date only.
Across the 250 years from 1800 to 2050, the share of the time each planet spends appearing to move backwards:
| Scrub through time | Longitude rate | The apparent track across the stars |
|---|---|---|
| t₀ | > 0 °/day | Direct |
| t₁ | 0 °/day | Direct → Retrograde |
| t₂ | < 0 °/day | Retrograde |
| t₃ | 0 °/day | Retrograde → Direct |
| t₄ | > 0 °/day | Direct |
Start on the left. Both the planet and Earth only ever travel one way around the Sun; neither stops and neither reverses at any point. The only thing that changes is the dashed line — the direction from Earth to the planet.
That is why a superior planet is always retrograde around opposition, and an inferior planet around inferior conjunction. Those are exactly the moments when the two planets are closest and the sightline swings fastest. You can check it on the left: during retrograde, the solid line between them is at its shortest.
This is a statement about geometry, not about influence. Retrograde motion, stations and direct motion are ordinary astronomical terms for an effect of perspective. They do not act on anything that happens on Earth.
Related tools: Opposition and conjunction · Ptolemaic and heliocentric · Solar System Orrery
Orbital elements from NASA/JPL, "Keplerian Elements for Approximate Positions of the Major Planets" (Standish & Williams, 1992), valid 1800–2050. Positions are geometric — no light-time or aberration — because this is a diagram of the ecliptic plane, not a simulation of a telescope view.
Interactive retrograde teaching tools have a long history: Nebraska (NAAP), Foothill AstroSims, the University of New Mexico, SimuFísica, NoA at the University of Fukui, Marble Cafe, and jsOrrery. None of their code is used here; the calculation and drawing are our own.