I was sick on the weekend, came back at work today...just a few thing about the dyno.
This kind of rolling road doesn't require any imput from the operator, as it gets all the data from climate sensors, OBD port on the car and pressure/temperature sensor on the exhaust (we make sensor ports before and after the cat on the oem exhaust too when we develop a complete system).
The power loss (Shleppleistung) is calculated in the warming up phase and in the decelerating phase after the power run. It's very accurate, if a car has the tyres a bit low on pressure the loss can be higher but the calculated engine power is usually spot on.
Anyway these graphs I posted are the best of 3/4 tries, both with the oem system and with ours, the higher outputs for both.
I don't know how many brands get (or I should say claim) gains in the 10 bhp range on small engines like the twingo's with just the catback, especially on cars with a system already decent, with straight pipes and the (somewhat) right diameter.
Especially on recent small turbo cars (the last Polo 1.2 tsi comes to my mind) you just don't get any increase, and the cause in the ECU that limits the boost even when there's lower counterpressure after the turbo that should increase it and the already nice design of the oem exhaust.
Obviously the most gain on NA cars comes from the headers and the cats, for example we got very good increase on an already very high specific output engine on the 997 GT3 mk2 3.8. We used "step-design" headers (where the step stands for the dual diameter primaries, that allow maximum speed out of the exhaust ports but reduced counterpressure and larger manifolds after a few inches).
On the Evora 3.5l we got 27 bhp of gain with a beefier torque curve with the complete exhaust.
On a bit older cars, like my clio or the saxò vts/106 rallye, there's a larger margin of tune as the exhaust has a restrictive cat and the exhaust is designed most for noise reduction and small costs (and you see that on the quality, VERY poor).