Leading laps: some stats

[Schummy]

The number of laps in the lead is an interesting info. It is a mean to show the level of domination a driver (or team) has in a season. It also gives nuanced info about who can be the WDC at the end of the season.

It has some biases, though. For example, it tends to bonus a front grid position, as first laps counts the same as last laps (but last laps are the ones who gives the real WDC points). On other side, leading laps corrects or filter some random noise (random accident, mechanical retirement, pit stop gaffes, (lol) team orders,... ).

If we take "number of wins" as the purest stat in motorsport (emphasis in "if"), the WDC points is a "socialist" system because it rewards greatly lesser drivers-cars and relatively belittle bigger players. Reversely, lead laps are an "elitist" stat because it over-reward winning drivers and under-reward no winning ones. For example, Alonso tends to do bad in leading laps, worse than its number of wins or amount of points (surely due to his strategic race craft).

Enough discourse already! Let's some numbers.

Leading lap (or laps leaded) in 2013

DRV  LL   LL%    FPG   PT%
==========================
ALO  37  21.8%     0   26%
VET  33  19.4%    -5   22%
WEB  32  18.8%    -7   22%(lol)
RAI  27  15.9%   -15   17%
SUT  11   6.5%   -52    5%
BUT   8   4.7%   -64    3%
HAM   8   4.7%   -64    3%
HUL   8   4.7%   -64    3%
MAS   4   2.4%   -87    1%
ROS   2   1.2%  -107    0%
==========================

• LL=Leading laps
• LL%=Percentage of LL
• FPG=Final points gap (at the end of the season, forecasted acording to LL%)
• PT%=Probability of Title (acording to FPG)

Obviously, it´s very early still, so it is not even a "prediction", but some patterns can be seen.

Two not so subtle details:
- Webber's 22% is unreal because we know he is going to be hindered even if he begins to go faster than Seb (except if Seb becomes a tortoise). Theoretical numbers shows Mark would have a chance to the title, but we knew it already.
- Hamilton is perceived as a potential contender, but his low 8% title probability shows he has not (yet?) a winning car, but just a car that he can place in a high position.
- Sutil and Hulkenberg numbers probably are a bit of random noise, BUT they are doing a good job and, for example, Force India is not an useless car. There could be a surprise.

[bravo]

Love this stuff Schummy, possibly the only useful purpose for statistics?


Please keep it up!

[Armco Bender]

Lead laps only means you stayed out on your knackered tyres longer while the others were in the pits changing thiers.

[Schummy]

And having a fast car... and probably having a nice grid position. All of it having a good wear rate of tyres.

That in the first stint... after the first stint you have to be simply fast to be in first position.

I don't see it a very simple thing to achieve. In fact, I see it as an exigent stat. Not totally correlated with points, but pretty informative. The only spurious thing I see is when a lesser driver gets some leading laps because his pit stop is out of sync. But it will be filtered out as the season passes by.

[Schummy]

bravo: Thanks!

The *other* purpose of using statistics methods is to pass Statistics courses.