BOAC, perhaps dazzled by the graphics?
Ka8, #2 provides the short answer.
The ‘Slippery Runways’ file graphically illustrates the theory and practice for a range of runway conditions.
Page 25 shows that when AB is used, and the deceleration target is met, then reverse has no further (decelerating) effect because the AB ‘backs off’ the brake component to maintain the required deceleration. N.B. the diagrams only show deceleration, not distance.
The landing distance charts (-700 shown on page 12) consider the component of each decelerating device and the time taken to deploy them, thus pure AB might provide marginally better performance than with AB + Rev, and a lower AB setting might ramp up the deceleration slightly slower than AB MAX. The distances depend very much on the assumptions made in use; technical systems might be closer to the theoretical limits, whereas pilot selection (Rev) might assume greater time delays.
MAX Manual is slightly better than AB MAX (and even better with Rev) as there is no limiting deceleration target, but the ultimate deceleration available and thus distance will still be limited by a range of other brake / runway factors – page 19.
These aspects are summarised, in simple words, on page 31, but the performance stills depend on ‘reasonably attainable assumptions’ (page 17, in the ‘Overrun’ file).