You mean this data?

SHELL 95

Max power 205bhp @ 5600rpm
Max torque 234lb It @ 2650rpm

BP 95
Max power 204bhp @ 5600rpm
Max torque 242lb ft @ 2300-3100rpm

Shell optimax
Max power 209bhp @ 5500rpm
Max torque 242lb It @ 2500-2900rpm

Tesco 99

Max power 212bhp @ 5500rpm
Max torque 242lb ft @ 2400-3500rpm

BP Ultimate

Max power 212bhp @ 5300rpm
Max torque 252lb ft @ 2400-2500rpm

Sunoco 'Race' fuel

Max power 218bhp @ 5200rpm Max torque 248lb It @ 2500-2800rpm

The spread in torque from worst to best is about 7.5%, not 10%, and if you pick Ultimate over BP 95 the difference is only 4%.

Not trivial by any means, but if you factor in the cost difference, not really cost effective. Interestingly, they found that the supermarket fuels matched the branded fuels-which perhaps isn't that surprising given that the fuels are sourced from the major refineries.

The mickey mouse Which? report was far more comprehensive, running no fewer than 10 dyno runs to establish the average power base for each condition. They also established the actual octane rating of the as-received fuel, which EVO did not.

BTW, the supermarket 95 fuels were excluded because the supply was inconsistent.

This logic seems bizarre. It assumes that 'branded' fuels remain consistent, with only supermarket fuels showing variability due to supply. That seems highly unlikely, given that vapour prssure limits change three times/year, not to mention the different sources of crude received by refineries.

What is common to both reports is that the link between octane rating and power output is fairly weak. Neither report could demonstrate an improvement for n/a engines running on 'premium'.

No quantative report has ever shown an improvement for part throttle fuel economy.

I also have had to replace the lamda sensor on the SWMBO Mini, which I've been reliably(?) informed was down to use of std petrol with lots of town driving.

So your informant is saying it's unfit for prupose... Pity all the other Minis driving around town happily are not all sitting broekn down at teh roadside.

BS is my view.

In the Golf TSI, an increase of nearly 8bhp for an extra £3 or so per tank isn't bad (Shell 95 versus the high-octane brew)

It would be interesting to see the full dyno plots.

8 BHP at 6000 rpm is all but useless, how often do you drive at those revs. As craig-pd130 suggests a dyno plot is required to see if there are power improvements everywhere.