F1 Fanatic

I think the statistics should be as clear-cut as possible, with little to none judgement calls and grey areas. So:

1. For the purpose of the statistics first lap passes should not count as overtaking moves, even if they actually are. Genuine overtakes on the first lap will occasionally happen and this can be noted in the comments, but for the purpose of the statistics they shouldn’t count. And “first lap” is just an example. It just as well could be only the first sector, but in my opinion we need this clearly defined point.

2. Pass in the pits does not count as an overtaking move. Similarly passing a car which has stalled or crashed should not be counted as an overtaking move.

3. Pass and re-pass should count as two overtaking moves… In most cases, that is. In general we can talk about overtaking move if it sticks for at least one corner. So, if car A passes car B in turn 4, keeps his position in turn 5 and loses it in turn 6, then I count it as two moves. Now, if car A passes car B in turn 4, and for whatever reason loses position in turn 5, then I count it as a failed overtaking attempt.

4. Different tyre compounds and use of the rear wing do not matter. Overtaking move is an overtaking move if it’s wheel to wheel and results in change of position. Of course lapping backmarkers does not count.

@Guy: it does not have to be complex, even if it’s strict. Overtaking move is passing another car which:
(1) takes place on track (after initial order shuffle in the first lap/sector),
(2) results in change of position between two functional cars (so passing backmarkers or stalled cars does not count),
(3) “sticks” for at least one corner.

So, did Alonso’s overtake Massa, after Massa was given a team order? Yes. It meets the criteria. It was unfair, sure, but Alonso overtook him. I think we should not split hairs too much. Just define some strict, simple rules, and all the exceptions and doubts can be put in comments. It should be an objective data from which people with different opinions could draw their own conclusions.

Luckily this is not as complex as defining a taxonomic class. But even that is possible and sometimes necessary. There are many definitions of birds, but to count all living bird species you would have to pick one. We don’t need it to be perfect, but it must be consistent within given framework. You can’t have two researchers using one definition, and three using another. The result would be meaningless.

To be honest I don’t like the idea of crowdsourcing. I can easily imagine that a “grey area” kind of manoeuvre made by for example Vettel, can be judged differently than a similar manoeuvre by Webber. There is a different level of sympathy toward different drivers – and it’s not bias – it’s natural, just not entirely objective. That’s why I think that “majority rule” shouldn’t be used in this case.

Going back to the ornithologist example: it would be like a group of scientists which decided to count all bird species, and every man in the group worked with a different definition.

We don’t have to define our rules to every minute detail. We just need a consistent framework. I’m actually for establishing this framework by crowd-sourcing, like we do right now. However once it is established we should stick to it.

Your latest example isn’t very good because it deals with prediction. We are not discussing predictions; we are talking about gathering data. When you do that, you need a consistent methodology and a clear definition of what you are looking for. In my opinion simple definition is better than a convoluted one – and definitely better than none at all.

Then we can say: “Within this framework we got these results. Grey-area cases are described in the comments”.

With your method, we can say: “This is the number of overtakes and we voted on every single one”. You still have to describe all the grey-area cases, but additionally you must specify which cases were included in the final tally and which were rejected.

“If a McLaren is on slicks in a downpour and is passed by an HRT on wets…”

It is quite clearly an overtake. Performance difference and different tyres do not matter. We will never know the exact technical aspects of each car. We will never know the exact difference between tyre compounds on every track, in all possible conditions. We could leave cases like that up to vote, but then it would be just guessing.

Of course we will discuss and analyse every questionable manoeuvre afterwards. After all, we are F1 fanatics, that’s what we do ;)

Keith will need some way of determining what counts as an overtaking move. That’s what he’s asking, and I think that my definition is pretty decent. I might be wrong, but I imagine that he wants some consistency and ability to give us the numbers relatively quickly. That’s where a simple, yet strict definition would come in handy. Then we would have room to debate.

If we start with the debate and follow it up with the statistic produced in said debate… well, I don’t know. It seems to me like a backwards way of doing it, but maybe that’s just me :)

Couldn’t agree more peteleeuk!

A fair point Keith! You’ve got me scratching my head now…

But then again can you really take away an over-take from every other driver?

It took 23 other drivers not to make the mistake that Vettel did and I think they should be ‘rewarded’ as such in this system.

I guess it really doesn’t make much difference if you count them or not, it’s all relative at the end of the day. As long as we have consistency there won’t be an issue.

One thing that keeps cropping up in this discussion. Is it an over-take when a driver makes a mistake, NOT just fails to defend?

A mistake being an off, like how Vettel had that horrendous start at Silverstone 2010 as Keith mentioned or perhaps when Hamilton ran off at Korea and Alonso took his position? Personally I would class those as over-takes for all 23 drivers plus Alonso in Korea…but that’s where the debate is.

I’ve got another idea regarding first lap / first sector overtakes. Instead of throwing them out the window, let’s just count them separately. This way we’ll have more data, we won’t discard anything, and in consequence we’ll have a fuller image of the race.

Regarding Andrew’s question: “is it an overtake when a driver makes a mistake, not just fails to defend?”

In my opinion yes, it’s an overtake. In general overtaking moves are possible because the driver in front leaves some sort of opening and the one in the back is skilled enough to use it. Sometimes the mistake is minor, sometimes it’s a royal mess up, and there are many steps between the two. Unless the driver takes himself out of the race, I’d count an error resulting in a pass as an overtake.