Hit the apex

“Hit the apex” is the first thing anyone tells you and it’s often not as sage as it sounds. Not least because hitting the apex in real life doesn’t always map to what you’re used to in a video game. It treats the apex like a spot painted on the tarmac, the same for a spec Miata and a GT3. It isn’t. The apex is an output, not a landmark. The corner sets the geometry; your car decides where inside that geometry the fast line actually clips.

Get the ratio wrong and you’ll brake in the right place, turn in at the right place, kiss the same kerb as the quick lad two garages down and still hand back two tenths on the way out. This is the second thing worth understanding about a racing line, after “get the car straight before you feed the power in.” It’s a bit more physics than you strictly need to drive well; plenty of fast drivers feel all of this without an equation in sight. But if you want to know why the line moves, read on.

On my first real track day in my 911, the racing instructor had great advice, get slow on entry so you can be quick on exit. Pay attention to where the apex is relative to where your car needs to be to maximise grip. Not all cars have the same capabilities.

Early apex, late apex, and the swap you can’t dodge

An apex has three properties, and they’re welded together: where it sits, how much angle the car still has to unwind there, and how fast it’s travelling. You don’t get to choose them separately. Turn in earlier and wider and you clip the inside sooner, at a shallower angle, carrying more speed. Turn in later and tighter and you meet the inside further round, at a sharper angle, going slower. Move one, the other two move with it.

Key principle — For a given car, an earlier apex buys entry speed and costs you exit; a later apex costs entry speed and buys you exit. There’s no free apex.

Fig. 01 — Angle and speed never move apart. Slide toward a later apex and you sharpen the angle but shed apex speed. The faded lines mark the extremes; the sweet spot lives between them, and where exactly depends on the car.

The full-throttle test

You can find a workable apex with none of the physics below. Pick one, and from that apex bury the throttle and hold it. If the car runs out of road and you’re mowing the exit grass, the apex was too early and too fast, take a later, slower one. If you reach full throttle and there’s tarmac to spare, you left time on the table. Take an earlier, faster one. When full throttle from the apex uses up the very last of the track at exit, you’re in the window.

It falls apart the moment you’re traction-limited and lighting up the rears, and in corners that shouldn’t be run to the edge at all. It also can’t tell a good line from a scruffy one that happened to use all the road anyway. For that last tenth, you need to know what the tyres are actually doing. That’s why I initially started running Outlap as an AI engineer assistant, to give drivers like me (not pros with a team!) a fighting chance on a track day.

Corner exit is a drag race

Here’s the reframe that makes everything else fall into place. From the apex onward, stop picturing a curve. Picture a drag strip that starts at the apex and runs dead straight down the following straight. Your only job is to build speed in that direction as hard as the tyres will allow.

At the apex, all your speed is pointed the wrong way: across the track, toward the exit kerb, not down the strip. Split it into two: cross-track speed (sideways, useless to the clock) and in-track speed (down the strip, the only thing the stopwatch sees). Through the exit the tyres hand one to the other while cross-track bleeds away as in-track builds. When cross-track hits zero you’re straight, at the edge, pointed down the road, at full chat. That’s a lovely exit.

Key principle — Only the car’s speed in the ideal direction lowers your lap time. Cross-track speed isn’t wasted — it’s the raw material — but it only counts once the tyres have turned it into drive down the road.

Fig. 02 — Same grip the whole way out; the amber arrow never changes length, it just rotates. At the apex it’s spent entirely turning the car (red). By the exit it’s all drive down the road (green). The corner is just the handover between the two.

It’s tempting to think more apex speed is free lap time. It isn’t. At the right apex the tyres are already at 100%, every bit of grip spent holding the arc. One extra km/h of cross-track speed and there’s nothing left to hold it & you run wide. Extra apex speed only helps if you also had more angle to point it down the road, and we’ve already seen those two pull against each other. So the real question that sets your apex is: how much cross-track speed does this car want to arrive with, so it can convert the lot into drive before the road runs out? That answer depends entirely on how hard the thing accelerates.

More power wants a later apex

Take two cars with identical grip. The first can’t out-pull drag on the way out and it holds a constant speed, so its quick line is a single constant-radius arc from apex to edge, and its apex is simply whatever makes that arc fit the road. Hand it the first car’s apex and it’s no quicker: it’s already using all its grip to hold the arc, so the extra power has nowhere to go but the scenery. Give it room instead with a later, slower, sharper apex so that as it accelerates and its line naturally opens out, it arrives at the edge exactly as it runs out of tarmac.

Key principle — The more a car accelerates relative to how it grips, the later its apex. Grip alone doesn’t move the apex; the ratio does.

Fig. 03 — Same grip, same corner. The powerful car (red) sacrifices apex speed for a later apex, then lets the line open as it accelerates — its force stays aimed down the road. The low-power car (blue) can only hold a constant-radius arc, so its force keeps pointing across the track. Red is quicker through the corner and down the straight.

Your car’s actual shape: the G-G plot

There’s one picture that holds all of this at once: the G-G plot. Put longitudinal g up and down (acceleration up, braking down) against lateral g left and right (cornering), then trace the most the car can do in every direction. A bare tyre traces a circle, roughly the same grip whichever way you load it. A whole car doesn’t.

The bottom half is usually round: four big brakes mean you can pull maximum g braking, cornering, or any trail-braked blend of the two. The top half is where cars part ways. Most can’t put enough power down to reach the grip limit under acceleration — especially in the faster corners — and most only drive two wheels. So the top gets sliced off. How much is missing is your acceleration-to-grip ratio, and it’s exactly what sets your apex.

Fig. 04 — Left: no drive to spare — the top is gone, the exit is a constant-radius arc, the apex is earliest. Middle: most road-and-track cars — the top gets clipped. Right: enough power and all four wheels driven — a full circle, the latest apex. Almost every car you’ll ever drive lives somewhere along this line.

Entry is the same trick, backwards

Everything so far has been about exit but entry is the mirror image. That tidy circular approach we used to picture the early/late trade is really the entry a car would want if it couldn’t brake and turn at the same time: pure cornering, riding the side of the G-G plot the whole way in. Almost every car can do better than that. Trail the brakes off as you wind the steering on and you travel from the bottom of the plot round to the side, keeping the tyres pinned at the limit the whole way. The line that draws isn’t a circle, it’s a spiral that tightens into the apex.

Fig. 05 — Trail off the brake as you add lock; the tyre force travels from the bottom of the circle (braking) round to the side (pure cornering), staying at the limit the whole way. Need an earlier apex? Open the entry spiral. Later apex? Tighten it.

What to do with this at the circuit

Don’t try to compute any of this from the driver’s seat. Nobody does, and the numbers change corner to corner and lap to lap. Use it as a mental model, not a calculator:

• Run the full-throttle test to get into the ballpark, then stop fiddling and go faster.
• Losing exit but clipping the same apex as a quicker driver? You probably want it later — you’re carrying apex speed the car can’t use.
• Feel for the handover. A good exit isn’t the one where you hit the kerb; it’s the one where the car stops fighting sideways and just drives.

When you generate a setup, Ross AI model gives you a circuit-specific starting point for the grip half of the equation (pressures, bias, bars, the lot) so the car’s at its limit when you go hunting for the apex. Where that apex actually sits is still yours to find. That’s the fun part.

A physics primer, not coaching. Outlap’s setup recommendations are advisory only — always verify with your team and drive within your limits.