A few more numbers...

[CDN-fan]

I also wanted to check which drivers were consistent...
These would be the drivers with low variability (low calculated variance) in their results.
This again has some limitations, * especially as Grosjean and Schumacher both have 6 dnf's out of 13 races so far.
For all drivers, the dnf's typically "happened", and will average out. However at Monza, Webber improved his stats slightly by choosing to dnf rather than replace his tires and finish the race. So remember that these numbers are more for illumination and entetainment than for "proof" of anything important.

For final starting grid position variation (lower is better):
12.00 Vettel
14.56 Alonso
16.14 Raikkonen
19.42 Massa
21.24 Grosjean
21.42 Button
26.27 Webber
35.14 Perez
38.81 Schumacher
39.74 Hamilton
44.90 Rosberg


For final position race results (excluding dnf):
5.62 Schumacher*
6.61 Alonso
7.22 Vettel
8.70 Webber
12.36 Raikkonen
17.58 Rosberg
18.23 Perez
19.42 Massa
28.07 Hamilton
30.33 Grosjean*
39.90 Button


For position change +/- variation:
6.14 Grosjean*
13.52 Massa
15.48 Alonso
18.49 Vettel
21.64 Webber
25.59 Raikkonen
29.09 Rosberg
38.15 Button
41.38 Perez
58.14 Schumacher*
60.42 Hamilton


So if you're managing a team, these variance numbers become useful for determining the stability, reliability, and preparation of both car and driver.
So, who would you want as your drivers?

Alonso, Vettel, Massa, and Raikkonen look the most "stable".
Rosberg, Perez, and Button, are more inconsistant,
Hamilton is definitely in a class of his own...

[SS454]

can you explain these numbers? Doesn't make any sense to me.

[CDN-fan]

These numbers were calculated from the results of their final grid position, where they finished, and how many positions were gained or lost during the race.
These are based on the same numbers I used to calculate the average starting position, etc.
So instead of using the spreadsheet's AVERAGE function on the column of numbers, I used the VARIANCE function, which is an indication of inconstancy.

For example, for this set of 9 numbers: 5, 6, 5, 4, 5, 4, 6, 6, 4, the AVERAGE = 5.00, while the VARIANCE is 0.75 in this sample. Since the numbers are all quite similar to each other, there is little variation between all the numbers, resulting in a low VARIANCE value.

However, for this set of 9 numbers: 1, 2, 3, 4, 5, 6, 7, 8, 9, the AVERAGE is also = 5.00, but now the VARIANCE is a much larger 7.50 for this sample. Since the VARIANCE looks at the difference between each number in a group and the group AVERAGE, it is a very useful way to indicate the consistency, or variability, just using a single calculated number.

In VARIANCE, smaller is tighter = good, and larger is scattered = bad.

So looking at the driver's results above, those drivers with the lowest VARIANCE values are getting results that are consistently tighter, and more stable.

And don't forget that since HRT's driver's are usually last, their results variance will be low, but this time it's a bad thing as it means they are consistently near the back of the pack, and unlikely to do otherwise...