Saved on Wheelbase: McLaren Paid with Floor Area

Why the short-wheelbase MCL40 initially gained a weight advantage - and then ran head-first into its own geometry. Shut the windows. It is going to get stuffy in here.

The short version is this: McLaren shortened the car to make it lighter. Once it learned how to save the same weight elsewhere, those missing centimetres suddenly became very valuable to the aerodynamicists. That is essentially where McLaren finds itself with the MCL40. You may stop reading now. While we are going forward.

Neil Houldey, McLaren’s Technical Director, Engineering, explained to The Race - with FormulaPassion picking up the story afterwards - that Woking deliberately chose a comparatively short wheelbase for its 2026 car.

The reasoning was perfectly rational. The new regulations made mass an even more sensitive design problem. A more compact car allowed McLaren to start with some margin and then spend the saved kilograms wherever they could buy performance.

But as the MCL40 developed, the team found other ways of taking weight out of the car. And then discovered that it was no longer short of kilograms. It was short of square centimetres of floor.

For 2027, McLaren is therefore considering stretching the car slightly. Physics thought about the problem for a couple of milliseconds and sent the invoice.

First, McLaren threw away some kilograms

The 2026 regulations already reduced the maximum Formula 1 wheelbase to 3400 mm. The previous generation could stretch another 200 mm beyond that. At the same time, minimum mass was reduced and the cars became narrower.

In other words, after years of Formula 1 cars steadily growing heavier, engineers were suddenly asked to build something both smaller and lighter. So McLaren’s decision made perfect sense.

A shorter car can potentially mean less material in the primary structure, less floor and bodywork surface, more compact packaging in certain areas and, generally, less structure that has to be made sufficiently stiff to survive the aerodynamic and mechanical loads acting upon it.

And the benefit of removing even one kilogram is a little more interesting than the usual: “Less mass equals more speed.”

If the car is genuinely above the permitted minimum mass, reducing m reduces the force the tyres have to generate in almost every dynamic condition.

In a corner:

Fᵧ = m·v² / R

To travel at the same speed v around the same radius R, the lighter car requires less lateral force. The tyres say thank you.

Under acceleration:

aₓ = Fₓ / m

For the same tractive force, reducing mass increases acceleration.

Under braking, the kinetic energy that ultimately has to be dissipated is:

Eₖ = m·v² / 2

Less mass means less energy for the brakes to turn into heat.

But there is another detail: racing tyres are load-sensitive. Doubling vertical load does not normally double the lateral force available from the tyre. So unnecessary kilograms hurt not only through schoolbook F = m·a. They also force the tyres to operate in a less favourable part of their load-versus-grip characteristic.

Of course, once the car is already at minimum weight, simply making another component lighter no longer makes the complete car lighter. But it still buys freedom. The engineers can put the mass back as ballast where it helps the centre of gravity, or spend that allowance on a heavier component whose performance gain is worth the penalty.

This is exactly the sort of benefit Houldey describes. The short car gave McLaren weight margin that could be spent on performance elsewhere. So the short wheelbase was not necessarily a mistake.

It was an exchange of floor area for mass at the prevailing exchange rate.

The problem is that the exchange rate changed.

Then the aerodynamicists asked for centimetres back

At the most basic level, floor downforce is the result of integrating the pressure difference over the working surface:

Fz = ∬ Δp(x,y) dS

Here Δp is the local pressure difference and dS is an infinitesimal element of surface area.

There are two obvious ways of making the resulting force larger.

You can increase Δp: accelerate the flow harder, produce a more useful pressure distribution, improve the way air enters the underfloor, control the floor-edge structures more effectively and manage expansion towards the rear of the car.

Or you can have more S - more surface over which you have the opportunity to create that pressure difference in the first place.

A longer floor is not automatically a better floor. You can add ten centimetres of completely useless surface, make the boundary layer more difficult to manage and receive nothing except additional drag for your trouble.

But an aerodynamicist would rather be allowed to decide whether those ten centimetres are useful. A short wheelbase removes that choice.

As development progressed, McLaren’s engineers realised that they had become better at extracting performance from additional floor area. Early in the design process, sacrificing some length appeared to carry little lap-time penalty.

Now that potential surface area is worth more. This distinction matters: the short floor did not suddenly become worse. McLaren became better at making floor area productive - and now it does not have enough of it.

One hundred millimetres

Let’s use a completely hypothetical example. These are not the actual dimensions of the MCL40. Suppose one design has a 3300 mm wheelbase and another 3400 mm. The difference is only 100 mm. Almost nothing for a naked eye, but for an aerodynamicist, however, that is an additional longitudinal region in which it may be possible to manipulate: channel geometry; pressure distribution; boundary-layer development; the interaction of longitudinal vortical structures; the flow delivered towards the rear of the floor; and the longitudinal position of the aerodynamic centre of pressure.

And it is not only the amount of additional force that matters. It also matters where that force acts. If a small part of the floor produces an additional force ΔFz, its contribution to pitch moment about the centre of gravity is:

ΔMᵧ = ΔFz · l

where l is the longitudinal distance from the centre of gravity.

The same newton of downforce is therefore not aerodynamically equivalent everywhere on the car.

The car does not live merely by the total vertical aerodynamic force Fz. It also lives by the position of the resultant aerodynamic force - in practical terms here, the longitudinal centre of pressure (we know that you can’t calculate the point of so called Center of Aero Pressure – you can only find the plane where it lays).

Its migration under braking, acceleration, changes in ride height, yaw and steering input helps determine whether the driver gets a predictable car or an argumentative device that changes its personality somewhere between corner entry and mid-corner.

So a few extra centimetres of wheelbase do not necessarily mean simply “more downforce”. They mean more space in which to construct the required downforce map.

But doesn’t a short-wheelbase car turn better?

It is extremely tempting to write: “Short wheelbase. More agile car.” Then close the textbook and make tea.

Life, unfortunately, suggests that we are not yet allowed to relax.

In a purely geometric bicycle model with no tyre slip:

tan δ = L / R

where L is wheelbase and R is corner radius.

For small steering angles:

δ ≈ L / R

So yes: for the same radius, a shorter car geometrically requires less steering angle. But a racing car does not negotiate a fast corner like a supermarket trolley.

The front and rear tyres operate at slip angles. Yaw moment is generated. The response depends on vertical load distribution, cornering stiffness, aerodynamic balance, differential behaviour and a considerable collection of other variables.

So let us isolate just one of them.

The car’s moment of inertia about the vertical axis - effectively a measure of how reluctant it is to be rotated into yaw - can be written as:

I_z = ∫ r² dm

Mass located well ahead of or behind the centre of gravity is particularly expensive because the distance r appears squared.

The angular acceleration produced by a given yawing moment is:

α_z = M_z / I_z

Or, if we express it through yaw rate:

dω_z / dt = M_z / I_z

A more compact mass distribution can therefore potentially reduce I_z and make the car respond more rapidly to a yawing moment. Potentially.

Move one heavy component towards the end of the car and the wonderfully convincing argument about short wheelbases can be thrown straight into the bin. The geometric length of a car and the distribution of its mass are not the same thing.

The unpleasant matter of load transfer

There is another cost associated with a shorter wheelbase, and this one is almost embarrassingly schoolbook-ish.

For longitudinal acceleration, the change in axle load in a simple quasi-static model is:

ΔF_z = m·aₓ·h / L

where:

m - vehicle mass;

aₓ - longitudinal acceleration;

h - height of the centre of gravity;

L - wheelbase.

And there sits our wheelbase. In the denominator. Everything else being equal, the shorter the car, the greater the longitudinal load transfer. If we again use our entirely hypothetical 3400 mm and 3300 mm wheelbases, the ratio is:

3400 / 3300 = 1.0303

So shortening the wheelbase from 3400 to 3300 mm increases quasi-static longitudinal load transfer by roughly 3.0%.

A small number? Perhaps. But braking in a modern Formula 1 car is exactly the sort of situation in which a few percentage points get attached to extremely large forces. The front axle gains vertical load while the rear is unloaded more heavily. Under acceleration the process reverses.

Then tyre load sensitivity joins the conversation. Shifting load between the axles does not merely alter the mechanical balance - it changes how effectively the four contact patches are being used.

And at this point it becomes obvious why wheelbase cannot be chosen independently of suspension, why tyres cannot be considered independently of their load characteristics, and why the floor cannot be designed as though the rest of the car were somebody else’s problem.

Their physics has already simmered together into one wonderfully thick engineering stew.

And aerodynamics joins the load-transfer problem

The real MCL40 is, naturally, an order of magnitude more complicated than our little equation. Under braking, suspension loads change. Ride heights change, the car pitches, the floor changes position relative to the road, the mass flow underneath the car changes, the pressure field changes.

Mechanical load transfer therefore alters aerodynamic forces, which in turn alter wheel loads again.

We get a coupled loop:

aₓ → ΔFz → suspension travel → floor position → Δp → aerodynamic load → wheel load

And all of it evolves continuously through corner entry.

This is why the apparently simple statement that “McLaren’s wheelbase is too short for the floor it now wants” can, if one is sufficiently determined, turn into an excellent argument about whether the original trade-off was worth making at all. McLaren did not simply choose a vehicle length. It selected one geometric parameter shared by several coupled physical systems.

What have the wings got to do with it?

The 2026 cars add another complication. Their active front and rear wings switch between corner and straight configurations. The objective is therefore no longer merely to maximise downforce-to-drag performance in one aerodynamic state, but to preserve useful aerodynamic behaviour across substantially different configurations.

McLaren has acknowledged that rivals can currently be more efficient in both corner mode and straight-line mode at certain points. The floor, meanwhile, does not fold away at the press of a button. It remains the foundation of the aerodynamic platform.

Even though the 2026 regulations moved away from the extreme ground-effect philosophy of the preceding generation, the floor has again become one of the main development areas of the MCL40. And that explains why additional centimetres have suddenly become valuable.

If McLaren can produce more useful underfloor load without paying an unacceptable drag penalty, it gains freedom to unload other aerodynamic devices, manipulate balance and chase not merely maximum F_z, but the relationship:

F_z / Fₓ

In other words, to buy each newton of downforce at a better exchange rate in newtons of drag. Formula 1 has not yet decided to constrain the overall aerodynamic characteristics of the car inside a prescribed performance window in the way FIA WEC does with its top-class regulations. So aerodynamic efficiency remains one of the central currencies available to the designer.

Did McLaren get it wrong?

That would be too strong. At the beginning of the project, one optimisation problem was particularly urgent: minimise mass.

McLaren chose the shorter car. A few months later, the situation changed. Weight was removed elsewhere, while floor development revealed opportunities that the original design work had not valued as highly.

Now the optimisation problem looks different: maximise aerodynamic efficiency with mass already under better control. And the optimum geometry has moved somewhere the current short-wheelbase car cannot reach.

That is how real racing-car design works. There is no universally “correct” wheelbase, or universally “correct” suspension stiffness, or “correct” centre of gravity.

There is only a collection of quantities tied together tightly enough that moving one causes several others to arrive immediately with an invoice.

McLaren saved several kilograms - exactly how many, we do not know - at the beginning of the season and gained the freedom to spend them on performance. Now the car has matured enough for its aerodynamicists to ask for several centimetres back.

They will have to wait until next year. Which means the MCL40 has not run out of ideas.

It has quite literally run into the finite limit of its own floor-area integral.