This data is incomplete, therefore the conclusions are also wrong.
I watched the FP2 race simulations on Multiviewer, and the top 4 teams split their drivers up looking for information on how the race would go if you have a 2-stop strategy on various tire configurations indicated by Pirelli as the top two fastest for the race.
So NOR, VER and LEC chose to simulate the 2-stop S-M-S race stint, but based on the fact that VER was on average about 4-tenths slower than NOR and even 2-tenths slower than LEC, I think he ran with more fuel on board than the other two, given the speeds in turns 6 and 15 betraying fuel quantity or PU mode.
The other drivers PIA, RUS, HAM, HAM, and SAI chose to simulate on soft tires for the first or last stint of the race. Hence the times were higher or lower depending on the amount of fuel they had on board. However, the times on soft were slower than on medium tires because they all had more fuel than those simulating stint 2 on medium tires, even if VER had slightly more fuel ... or maybe a setting anticipating a rain race?
The other pilots PIA, RUS, HAM, and SAI chose to simulate the first or last stint of the race on soft tires. Hence the times that were higher or lower, depending on the amount of fuel they had on board. However, the times on soft were slower than on medium tires because they all had more fuel than those simulating stint 2 on medium tires, even if VER had slightly more fuel...or maybe a setting anticipating a race in the rain?
In FP3 many drivers tested the solutions found overnight in their simulator, but what intrigued me was the fact that HAM was always testing something and always making changes on the wheel. In his last simulation lap, he was 2-tenths slower on S1 and 3.5-tenths slower on S2 than his fastest lap (L14) even though he was on intermediate tires having completed 24 laps! Then he went straight to the pits. So many pilots still do not show their hands yet ...
wonder what is the current probabilities.
disguised as Ham fan 
