For years, we've measured how fast a car is by two metrics: top speed and 0-60 times. I will now point out that 0-60 times are bullshit, mostly because carmakers (using the Porsche 918 as an example) are gaming the system.This is something that just about everyone knows, since the number of times that regular human beings go from 0-60 is approximately never. Real speed is acceleration from when you are already rolling, like when you're merging onto a highway, or you're powering out of a turn on some backroad.But this realization comes to me from Road and Track's published performance figures for the Porsche 918. The first figure they note is that the Porker (weighing something between 3,715 and 3,858 lbs depending on how you measure) rips from 0-60 as fast a a Bugatti Veyron. That's 2.5 seconds.

R&T go on to give the 918's 5-60 time, a figure that Porsche does not quote. It's 2.8 seconds. That's slower than the car's 0-60 time. How could this be?The answer is simple - carmakers these days are getting very, very good at launch control systems, which perfectly manage the engine's power and the tires' traction to give perfect 0-60 times, every time. Porsche's system, for instance,is so computer-optimized you can launch your car over 50 consecutive times, or while drinking a coke.The joke is that the car has specific programming to set it up to launch from 0-60 (or 0-100kph/62mph) for the sole purpose of grabbing headlines. The 918, when in a real world situation, like from a slow roll to a highway run, is actually slower than it is in 0-60 fantasy land. I'm singling out the 918 here, but as launch control systems get better and better, it seems like launch-control-optimized cars are getting more and more divorced from genuine on-road performance.Carmakers are gaming the system, and they're doing it just so that their cars appear faster in car magazines and Internet arguments. Drive them like a real human being and not some kind of standstill-to-speed-limit robot and you'll see how fast they really are.