Panning is usually taught as instinct: put the guitar a bit to the left, the piano a bit to the right, see how it feels. I find it easier to think about as a coordinate system. Every sound gets an X-position, and the job of a mix is to distribute those positions, deliberately, not by feel, so that no zone is overcrowded and the energy on both sides stays balanced.

The Axis

Think of panning as a number line from −100 to +100.

Position−100−500+50+100
Hard LLeftCenterRightHard R

The Rules as Axioms

Axiom 1, Low frequencies anchor at X = 0

Bass and kick live at X = 0. Frequencies below ~80–100 Hz are effectively omnidirectional, their wavelengths are too long for your ears to use interaural time difference (ITD), the main spatial-localisation mechanism, to determine direction. Panning them off-centre doesn't create a spatial effect; it creates uneven headroom and a phase problem.

Bass at X ≠ 0 is a physics constraint, not a stylistic choice.

Axiom 2, The centre is load-bearing

X = 0 is a structural column. Overload it and you lose punch and separation. The only elements that must anchor there are the kick drum, bass or 808, lead vocal, and snare. Everything else should be displaced outward to create space around them.

Axiom 3, Symmetry ≠ balance

Panning a guitar to −60 and mirroring it at +60 is geometrically symmetric but not automatically balanced. True balance means equal perceived energy on both sides, not equal positions. Three quieter elements on the left and one louder element on the right can balance perfectly if their RMS levels compensate.

Position and loudness are two independent variables.

Axiom 4, Width is a function of frequency

The auditory system localises high frequencies more precisely than low ones, via interaural level difference (ILD). High-frequency elements can be panned wide without destabilising the mix, while low-mids need to stay closer to centre. The usable panning range expands as pitch rises, like a triangle.

Frequency rangeApprox. HzPan zone
Low20 – 250 Hz±0 – 10
Mid250 Hz – 4 kHz±20 – 50
High4 – 20 kHz±60 – 100

Axiom 5, Duplicate elements must diverge

Two identical signals summed at X = 0 add in phase, you get +6 dB of the same thing. Panning them to ±60 instead creates a Haas-effect spread: the signals decorrelate spatially, producing perceived width and fullness without a volume increase. This is the acoustic basis of double-tracking.

Two diverged copies > one centred copy.

The Placement Map

ElementX positionNotes
Kick drum0Never move this
Bass guitar / 8080Never move this
Lead vocal0Anchor of the song
Snare0 (or ±5)Needs to punch centre
Lead synth / piano0 to ±15Slight width OK
Claps±20 to ±40Spread for width
Hi-hats±30 to ±60Alternate L/R if multiple
Rhythm guitar±50 to ±70Classic rock: hard pan ±100
Backing vocals±30 to ±60Mirror pairs symmetrically
Harmony vocals±50 to ±80Wider than backing
Piano (non-lead)±20 to ±50
Pads / atmosphere±60 to ±100Fill the edges
Sound FX / one-shotsAnywhereUse movement / automation
Percussion layers±40 to ±80Scatter asymmetrically
Strings / orchestra±30 to ±90Spread like a real stage

The Frequency-Space Matrix

Put the two axes together and the mix becomes a 2D grid, pan position horizontal, frequency vertical. The goal is even distribution across both: no cell empty, no cell overcrowded.

Hard LLeftCenterRightHard R
HighPadsHarmoniesFXHarmoniesPads
MidGuitarBVoxLead vocalBVoxPiano
LowPercClapKick / Bass / SnareClapPerc

That's what it looks like at the end of a session:

Logic Pro mixer channel strips showing pan and level automation across a full session
Logic Pro mixer, pan positions from an actual session.

Everything above is theory until it's a fader you've actually pulled. This is what distributing the axioms looks like on a real track, bass and kick locked to 0, guitars split wide left and right, vocals stacked and spread outward from the centre. No cell empty, no cell overcrowded.

← Back to Writing