As has been mentioned by IAN-UK there are a lot of fudge factors in load and balance calculations.

We know the empty weight of the aircraft and they are weighed on a regular basis, we also know the weight on the catering and all of its equipment. We know the weight of the cargo and the Mail and all of the baggage. The human beings are total guess work but organisations such as the CAA and the likes of IATA (iirc) do regular sample audits of human traffic weights.
As such we use different assumption for the mass of humanity based on regions of the world in longhaul. USA is more mass or weight per head than the Far East. But is is all still a guess and a fudge.

Given so many variables can change right up to the last minute prior to that last hold or cabin door closing, we base all of our take off performance calculations on an assumed weight based off of a ‘provisional loadsheet’. This complies with all of the regulatory performance requirements and our performance when using take off thrust de-rates and assumed temperature method (or FLEX temp in Airbus terms) thrust reductions, we are inherently safe as we are actually producing more thrust than the performance calculation allows for (the air is denser than what we have told the aircraft it is).

Given these assumptions, and due to the fact that the payload tends to go down rather than up due to late arriving bags/cargo etc, we have a tolerance on the fleet in question of +1000 to -4000KGs allowable difference between the ‘provisional’ figures and the actual or ‘final’ figures. We typically receive those just as we commence the taxi after engine start. From doors closed the turn around manager/coordinator has now been able to input any changes to the planned load and communicate that with load control, who in turn send us the final figures through an acars message. If we don’t get it, we call them up.

With regards to the water…..we don’t typically fill the tank full unless going somewhere that the potable water has failed certain tests. So we are taking about less than 1000 kgs of weight, but a bit more than 500Kgs. Given all the other assumptions we really wouldn’t notice, especially when taking off around the 210 to 220-ish tonne mark for a flight of this length.

The only change that might have occurred would be with aircraft trim, and that is merely adjusting the datum point for the horizontal stab whilst taxiing out and not an uncommon occurrence, and one highlighted through a procedure we follow to ensure data is ‘in compliance’ with this or appropriately replan and recalculate if there are ‘changes to’ the provisional figures outside of allowable tolerances.

Given the state of resourcing with various service partners airport wide, I would be sceptical of stories that the water was simply forgotten. The attendant panel is accessed a good few times pre departure by the IFM/L and various inputs are needed, such as the passenger head count figure. Given there is a TRM/C who needs to update load control with actual figures I have a feeling this was probably more to do with no staff available to service the tanks without incurring a significant and sizeable delay….or that there was a problem with the potable water system once under way.