Baku City Circuit: The Compromise With No Clean Answer

The first thing I hear at Baku is not the engine. It is the tyre screaming against asphalt that has been baking in the sun for hours, and right behind that, a silence stretching more than two kilometres. That silence is the longest straight in Formula 1 history, where the car runs flat out from Turn 16 all the way to Turn 1, and where every aerodynamic decision is wagered in under thirty seconds.

I sit in Munich, headphones on, rewinding the braking zone into Turn 1. At 54, I have learned that emotion is also a rare form of data. And the emotional data at Baku always takes the same shape: a very long scream, then a dry crack as cold brakes are stamped at more than 340 km/h.

Almost every preview before the Azerbaijan Grand Prix tells you something very simple: Baku is a low-downforce circuit, like Monza, so chop the rear wing flat and go straight. Randy Singh, McLaren's Racing Director, does not say that. He says his team will assess the overall impact when they arrive. It sounds bureaucratic, but it breaks a consensus that has stood for too long, and I believe that consensus has been wrong for years without anyone bothering to recheck it.

Context: two circuits stitched into one lap

Baku City Circuit is 6.003 km with 20 corners. The main straight is around 2.2 km, with top speeds above 360 km/h. Read only those two facts and you will immediately conclude this is Monza with walls, and you will be wrong at exactly the point where thousands of analyses were wrong before you.

The lap does not end at the start of the straight. It turns into the old city, where stone fortress walls squeeze the road down to as little as 7.6 metres, and where Sector 2 is a chain of slow corners that gives tyres and brakes no moment of rest. The Maiden Tower stands close to the kerb in that section, and at Baku the distance between a perfect lap and an afternoon in the barriers is measured in centimetres.

That is the whole problem. Baku is two circuits stitched into one lap, and the car is configured only once.

Based on my experience covering races since 2026, I have never seen a circuit where the two halves of the lap demand two such contradictory personalities. Monza has one problem. Baku has two, and they fight each other in every possible way.

2026 makes everything harder. The new regulations split power close to 50/50 between the combustion engine and the electrical system, the cars are smaller and lighter, and most importantly for Baku: active aerodynamics. Both front and rear wings move, letting the car switch between a high-downforce and a low-drag configuration. Boost mode, which replaces DRS, gives the driver a short burst of extra electrical deployment.

There are silences on a racetrack that say more than any blockbuster contract. At Baku, that silence lives in Sector 2. And in 2026, it will speak louder than ever.

Downforce: betting on Saturday or betting on Sunday

Baku is not an ultra-low-downforce circuit. The reason sits in the old city. More downforce lets the car rotate better through Sector 2, and Sector 2 decides grid position, because that is where gaps between cars are created. A tenth in the old city is worth more than three tenths on the straight, because on the straight you only need top speed, while in the old city you need front grip, rear grip and the driver's confidence.

But that same downforce becomes drag on the longest straight on the calendar, where you get eaten alive during the race. This is the classic street-circuit compromise: bet on qualifying, or bet on the race. You cannot have both at maximum.

What I want to stress, and what McLaren tacitly admits, is that no setup is right for both halves of the lap. There is only a setup that is less wrong.

Active aero makes the problem less binary, but does not erase it

Moving wings make the downforce choice less binary. In theory, a team can run more downforce for qualifying and still open the wing to protect top speed on the straight. That is why I believe a higher-downforce setup becomes more feasible at Baku than in previous years, and this is the biggest difference between Baku 2026 and every earlier version of Baku.

But there is still a choice to make. Active aero changes the shape of the compromise, it does not remove the compromise. Downforce still affects car balance in every corner where the wing is closed, still determines tyre wear, still determines surface temperatures through the old city. And Baku has plenty of those corners.

One hypothesis I put forward and want to verify with data: if the low-drag mode is more efficient than expected, the fight at Baku 2026 could shift from setup strategy to energy management. In that case, performance engineers become as important as aerodynamicists.

Brakes: the problem with no clean answer

This is the most overlooked part of any Baku discussion, and also my favourite, because it is purely technical.

The lap swings between two extremes. After the 2.2 km straight, the brakes into Turn 1 are almost completely cold. You stamp on them at low temperature, at exactly the moment you need the biggest braking force of the entire lap. That is the perfect recipe for a lock-up, and at Baku a lock-up at Turn 1 usually means losing the whole rest of the lap.

The rest of the lap is the opposite: rapid successive braking zones demanding heavy cooling. If you get stuck in traffic, cooling demand rises further. You cannot simultaneously cool at maximum and keep the brakes ideally hot. Race engineers must pick the compromise that does the least harm, and at Baku the phrase "least harm" is always an admission that there is no clean solution.

Here I want to say something I only dare to frame as speculation. If a team opens its brake ducts wider for the slow corners, drag rises on the straight, and that could carry consequences for tyre temperatures in traffic. My confidence in this hypothesis is low to medium. I raise it because it deserves checking against tyre temperature data in practice, not because I am certain.

Energy deployment is the new DRS

Max Verstappen once said that anyone saving Boost to use out of the final corner will be a very tricky proposition to pass. That line is underrated, and I think it describes an entirely new defensive game at Baku.

With old DRS, the attacking right belonged to the car behind. With Boost mode, that right can belong to the car ahead if he manages his energy well. On a 2.2 km straight, the gap that opens or closes depends on when you press the button, and at 360 km/h that moment is measured in tenths of a second.

But there is an unresolved paradox. A stronger electrical system creates extra thrust, yet deployment is limited per lap and per corner. At Baku, if you spend many consecutive seconds flat out, energy drains fast. That means the fight on the straight could shift from aerodynamics to energy management, and whoever understands his deployment map best holds the biggest advantage.

This is the point I think no team has fully nailed down. Baku is the first race of the season with a straight long enough to expose every weakness in an energy strategy.

A one-stop race is not vanilla

At Baku, without a safety car, the race is almost certainly a one-stopper. That sounds simple, but the history of this event says otherwise.

In 2026, Lando Norris was eliminated in Q1 because of yellow flags and because he took too much risk with tyres. He started 15th. McLaren put him on the hard compound, intending a long run to pass cars while others pitted. He passed Daniel Ricciardo, Lance Stroll, Nico Hülkenberg, Yuki Tsunoda and Oliver Bearman in the opening phase. By the time the medium starters stopped, he was P10, then rose to P5 behind Alex Albon. Albon boxed on Lap 31; Norris stayed out. Carlos Sainz and George Russell, on fresher hards, passed him, yet he kept going until he had cleared most of his pit window. He stopped on Lap 37 and lost exactly one place, to Verstappen.

That is a reverse one-stop: start on the harder tyre, run long, then switch to a softer tyre with a big pace delta for a short final stint. It worked without a safety car. It also shows something strategists often forget: at Baku, running in clean air on old tyres is sometimes worth more than track position itself.

I still remember Oscar Piastri winning this race in 2026 with a drive in which he controlled the rhythm by never pushing the tyres into the danger zone. It was not a flashy performance. It was an informed one.

Under the 2026 rules, the pit window is wide, tyre degradation is not yet fully established, and overtaking remains an unknown. A one-stop race at Baku is not vanilla. It only looks vanilla when you read it through the results sheet instead of the sequence of decisions.

Yellow and red flags as a strategic variable

Baku has an above-average frequency of yellow flags, red flags and safety cars. In 2026, Verstappen's tyre exploded on the straight and the race was stopped. In other years, crashes in the old city turned the final laps into a lottery.

That changes how teams bet. On one hand, it encourages extending the first stint and hoping for a race intervention. On the other, waiting too long is a lethal risk, because if a red flag comes after you have already pitted, you have paid for something others got for free.

This is why I always tell younger people in this industry that Baku strategy is a probability problem, not an optimisation problem. And those who forget it usually lose the race in the engineering room rather than on the track.

Where I could be wrong

Now the part where I could be wrong, and I always reserve this part for myself rather than for others.

The drag coefficient of the 2026 low-drag configuration has not been validated on the Baku track. If low-drag mode is not as effective as expected, the advantage of a high-downforce setup disappears, and the old binary problem returns intact. That is a scenario I rate as having a real probability.

My hypothesis about hot airflow and tyre temperatures has no numbers behind it. If tyre temperature sensor data shows a negligible effect, then that is a speculation I should withdraw rather than defend.

And I want to say something about the source material I am using. The original analysis was produced in partnership with a risk management company. That does not make it wrong, but it tends to emphasise risk and unpredictability more than necessary, because that is the language of that industry. Readers should filter out the technical part and leave the rhetoric behind.

I have been wrong in exactly this way before. In 2026, I wrote that Erling Haaland would break Pep Guardiola's pressing structure because a classic centre-forward slows the circulation of the ball. He scored 36 goals in 35 Premier League games. I was wrong, and I delighted in dissecting that error instead of defending it. The sweetest mistake is the one that shows me I am still listening.

But one thing I hold firm: any technical prediction about Baku before GPS, top speed and sector data arrive from practice is provisional. Randy Singh says McLaren will assess on site. So will the others. And so should I.

Strategy is not a mummy, so do not seal it behind museum glass.

What I will be watching

If you want a verifiable prediction, here are three things I will watch at Baku.

Grid position will be decided in Sector 2, not on the straight. Whoever is fastest through the old city takes pole, regardless of their top speed. If a team chooses a low-downforce setup and still takes pole, I will have to rewrite this entire article.

The race will be decided by who stays in clean air longest. The reverse one-stop will appear again, and it may appear at a team nobody expects.

And Boost mode will create a new kind of defence out of Turn 16. The driver who saves energy until he sees the white flag will control the straight rather than the driver behind.

Fans do not remember the numbers board, they remember the breathing of the race. At Baku, that breathing is the tyre screaming for 2.2 km, and the brake being stamped while the temperature needle has not yet moved. I will listen to it, and I will write it down. The teams, meanwhile, will have to choose, as they always have at this place.