What’s so different about the 2026 cars is that the state of charge of the battery will influence how much braking happens from energy recovery. Assume that the car is going into Turn 1 at the same speed each lap and the braking point needed to slow around the corner is the same. Go into the corner with low battery charge and the car will want to grab a lot of electricity = less brake force required by the driver. Go into the corner with an almost full battery and the car will top up then stop needing to generate electricity via the rear brakes = more brake force required by the driver.

But in actual racing, how much electrical power the driver used in a lap is variable. The charge state of the battery approaching a corner is variable. The speed carried into a corner is variable. The tyre deg and therefore amount of grip is variable. The car’s computer is supposed to ‘learn’ how much braking/energy recovery is needed based on previous laps but drivers are still going to caught out applying brake pressure just bc it’s so complex.

the variable braking point problem you described doesn’t exist in isolation. it interacts with super-clipping, where the MGU-K starts harvesting energy while the driver is still flat on the throttle. so the ICE is pushing the car forward while the hybrid system is simultaneously creating resistance in the drivetrain to recharge the battery. the driver is literally flooring it while the car is fighting itself to harvest energy.

that decision to harvest mid-straight then changes the battery state going into the next braking zone. which changes the braking point. which the driver now has to recalculate in real time. every input cascades into the next one. it isn’t a sequence of separate systems anymore, it’s one continuous interdependent loop pretending to be a car.

and the whole thing resets at every circuit. so the muscle-memory issue isn’t just “learn new brake points” it’s that there are no fixed brake points anymore. the correct braking point becomes a moving target that shifts lap by lap depending on battery state.

lewis has nearly two decades of braking instincts stored in his body, not in a notebook. that’s not something you overwrite in a couple of test sessions. antonelli and lindblad are quick partly bc they don’t have years of old instincts to unlearn.

which brings us to what drivers are actually juggling simultaneously at 300 km/h:

current battery state

projected battery state at the next braking zone

whether the brake point is 30 meters earlier or 10 meters later this lap

current aero mode (corner / straight / overtake)

whether the car is harvesting or deploying energy right now

tyre degradation affecting grip and therefore braking distance

the position of the car ahead and what their battery state is likely to be

oh and the track and the cars behind them

that’s a massive cognitive load to manage at 300 km/h. lando didn’t see the cooling fan he hit in qualifying because he was already juggling all of the above. lewis even said he sat in a team meeting and felt like you needed a PhD just to follow the discussion. these are the best drivers in the world and the complexity is overwhelming them.

and the audience sees nearly none of this. battery state isn’t shown on the broadcast because teams don’t want competitors reading their energy strategy. so from the broadcast it just looks like pierre suddenly loses 60 km/h on a straight for no obvious reason. super clipping looks like a mechanical problem. variable braking points look like driver mistakes

the sport has basically created a car that requires layers of hidden telemetry to understand and then hidden that telemetry for competitive reasons. to a casual viewer it just looks random and incomprehensible.

Alysa Liu leaving the ice after her free skate!