Overtaking - 2013 debate

[Saribro]

except you fail to understand what evolution is.

http://dictionary.reference.com/browse/ ... on?&path=/
Evolution doesn't have to be a random, "natural" process.

[Richard]

your theory doesnt make any sense because if you apply it to other categories where they have the SAME cars, you still see plenty of racing. So it doesn't hold!

Indeed, spec series have more restrictions to prevent aero dominance, even though we all know that aero dominance gives us faster cars. So those spec series are sub-optimal when it comes to speed.

However the premise of F1 is to have the fastest cars and allow optimisation to achieve that aim. That results in aero dominated designs which make overtaking very hard. Fewer restriction to allow even faster cars would result in even less overtaking.

We have proof of this when we look at series with more restrictions and see that they have more overtaking. If you want more overtaking in F1 then we need to ban more things so we remove the arms race based on speed. I gather this has already been tried, some people call it Indy while others call it GP2.

It's your choice, sub-optimal cars with wide margins of error that can overtake (GP2 or Indy) or cars optimised in an arms race to be very fast (F1).

[Richard]

Evolution actually encourages the competitors to find different strategies so they can take advantage of what the opponents leave behind.....

Lotus ... found something else, which turns out to be a weakness at Red Bull.

That's why different teams on different race strategies is a sign of good health in F1. We've had a few years of monoculture with RB dominance and no apparent viable mutations. Lotus showed that being brave and trying mutations can lead to an alternative, even though that does mean trying a few non-viable mutations on the way eg front exit exhaust.

There will be a tussle between the two. The competitive nature of F1 tends to favour convergence towards an optimum solution with the side effect of “boring” racing until the next disruptive solution arrives to change the order. Then they’ll converge towards that solution and we’ll go through the same cycle again.

[fiohaa]

your theory doesnt make any sense because if you apply it to other categories where they have the SAME cars, you still see plenty of racing. So it doesn't hold!

Indeed, spec series have more restrictions to prevent aero dominance, even though we all know that aero dominance gives us faster cars. So those spec series are sub-optimal when it comes to speed.

However the premise of F1 is to have the fastest cars and allow optimisation to achieve that aim. That results in aero dominated designs which make overtaking very hard. Fewer restriction to allow even faster cars would result in even less overtaking.

We have proof of this when we look at series with more restrictions and see that they have more overtaking. If you want more overtaking in F1 then we need to ban more things so we remove the arms race based on speed. I gather this has already been tried, some people call it Indy while others call it GP2.

It's your choice, sub-optimal cars with wide margins of error that can overtake (GP2 or Indy) or cars optimised in an arms race to be very fast (F1).

GP2 cars are only 5-6 seconds slower than F1 now, are they not?
F1 cars used to have 850-950hp and durable grippy but nervous grooved tyres which made the cars difficult to drive.
You saw a ton more mistakes by drivers then than now - and that was with cars which had electronic aids, optimal ECU's and traction control. Partly because they had to drive them fast (unlike now) and partly because they were more difficult to drive on the limit, one of the reasons why Villeneuve struggled when grooves came in 98.

So theres your solution.........make the cars like how they were. Done. That redresses your curve nicely.
Oh not to mention the car park tracks you have now. Theres your other solution, stick some sand traps back in.
Done.
Limit budgets actively through accounting disclosures, just like we have corporate governance for PLC's, preventing a team from dominating through money (another point you just dont seem to include in your analysis).
And probably a whole host of other pointless rule changes which have led to this point (nothing to do with evolution)

[Stradivarius]

Evolution actually encourages the competitors to find different strategies so they can take advantage of what the opponents leave behind.....

Lotus ... found something else, which turns out to be a weakness at Red Bull.

That's why different teams on different race strategies is a sign of good health in F1. We've had a few years of monoculture with RB dominance and no apparent viable mutations. Lotus showed that being brave and trying mutations can lead to an alternative, even though that does mean trying a few non-viable mutations on the way eg front exit exhaust.

There will be a tussle between the two. The competitive nature of F1 tends to favour convergence towards an optimum solution with the side effect of “boring” racing until the next disruptive solution arrives to change the order. Then they’ll converge towards that solution and we’ll go through the same cycle again.

I agree that teams will converge towards and optimum, but I am not sure if this will always result in "boring" racing. We often see teams dominate for some years, like Red Bull have done, and like Ferrari did 10 years ago. But even without disruptive solutions, the domination will usually end by itself. The more Ferrari dominated, the less the opposition had to loose by exploring new mutations. That is why we saw all the strongest teams except Ferrari move away from Bridgestone to Michelin after Ferrari's domination started. It wasn't necessarily because Michelin were any better than Bridgestone, it was simply due to the fact that Ferrari had the best car and in order to beat them, the other teams needed to try something different. At the same time, Ferrari, who were in the leading position, started to become conservative. They worked close with Bridgestone in order to develop tyres that were consistent and reliable, because they would normally win if nothing unexpected happened. They had the best car and the best driver and the best team and wanted consistency and predictability. McLaren and Williams and Renaualt and BAR, however, wanted the oposite. Now we see that Red Bull are very keen on changing the tyres, as their speciality is not tyre management, while Lotus wants to keep the tyres as they are.

[fiohaa]

except you fail to understand what evolution is.

http://dictionary.reference.com/browse/ ... on?&path=/
Evolution doesn't have to be a random, "natural" process.

i didnt say it was random. and i specifically said 'evolution by natural selection'.
and no it doesnt have to be a natural process, hence breeding.

but then thats not evolution by natural selection. thats....evolution by 'human' selection.
you could argue ultimately that everything is natural because we are products of the natural world, and therefore theres no such thing as 'man made'.
which is fair enough.

[Stradivarius]

except you fail to understand what evolution is.

http://dictionary.reference.com/browse/ ... on?&path=/
Evolution doesn't have to be a random, "natural" process.

i didnt say it was random. and i specifically said 'evolution by natural selection'.
and no it doesnt have to be a natural process, hence breeding.

but then thats not evolution by natural selection. thats....evolution by 'human' selection.
you could argue ultimately that everything is natural because we are products of the natural world, and therefore theres no such thing as 'man made'.
which is fair enough.

I don't really see the point of distinguishing between natural selection and human selection in this case. The governing mechanism of evolution is the same in any case: What brings success is likely to continue and will be the basis for the next generation/iteration, and what doesn't bring success is more likely to change in the next generation/iteration. The teams use their knowledge to improve their car and if the improvement works they are likely to keep it, but some times it turns out the improvement doesn't work as expected, or it may work, but make further improvement difficult and then this "branch" will die out, when the team finds a better solution trying a different direction. Actually, it won't necessarily die out, because the team will keep the knowledge of things that didn't work well and try it later on if things change. For example, the KERS didn't really die out in 2009 although it was abandoned. When the teams started using KERS in 2011 they could benefit from lessons learned in 2009. This is exactly what happens through natural selection, although generally much slower.

[Richard]

I thought random mutations was one of the key principles of evolution. That's how we get disruptive changes rather incremental changes.

[fiohaa]

I thought random mutations was one of the key principles of evolution. That's how we get disruptive changes rather incremental changes.

yes, indeed that is true,
http://evolution.berkeley.edu/evosite/e ... ndom.shtml

but unfortunately people often get confused with that and evolution itself being random process
i.e. christians going 'ooooh how could we possibly have magically turned from fish into humans by chance'
point being it wasn't random, we were selected by nature.
as for changes, i think you'll find in biology they are hugely incremental...over...hundreds of millions of years. Yes, some steps have occurred I'm sure, but generally its incremental.

i was just making that correction to Stradivarius.

[fiohaa]

I don't really see the point of distinguishing between natural selection and human selection in this case. The governing mechanism of evolution is the same in any case: What brings success is likely to continue and will be the basis for the next generation/iteration, and what doesn't bring success is more likely to change in the next generation/iteration. The teams use their knowledge to improve their car and if the improvement works they are likely to keep it, but some times it turns out the improvement doesn't work as expected, or it may work, but make further improvement difficult and then this "branch" will die out, when the team finds a better solution trying a different direction. Actually, it won't necessarily die out, because the team will keep the knowledge of things that didn't work well and try it later on if things change. For example, the KERS didn't really die out in 2009 although it was abandoned. When the teams started using KERS in 2011 they could benefit from lessons learned in 2009. This is exactly what happens through natural selection, although generally much slower.

im sorry, but you just cannot make the same comparisons between what is happening in a sport, and the natural process of evolution. its absurd.

you have to distinguish between human and natural......its one of the fundamental topics with evolution by natural selection - Hence the theory is 'Evolution by natural selection'. The clue is in the title. Its not just termed 'evolution' which on its own can mean a lot of things.

there are so many human motivations, influences, and not to mention MONEY that shape F1 that deeming it to be a natural process is incredibly naiive.
Okay, teams learnt from lessons and improved on their creations - what on earth is your point? That in itself does not mean that eventually all the teams reach some performance wall.

The OP is trying to make a point that F1 is reaching some kind of performance plateau, and that its cyclical.

I'm arguing that this cannot be the case, due to points I've already made in previous posts. If his theory holds true, all the teams would eventually be competing for wins, and All be similar pace.
That is clearly not the case, why?

1. Massively different budgets (hence why teams should open up their accounts and we should do a Points/£ index)
2. because of this, its not a fair competition. That is a fact. None of the teams are operating in exactly the same environment.

This analogy would absolutely fine, if all the teams were operating in the same environment, each year, and the teams that 'survived' (did the best) were the ones that adapted. Okay......but the environment is not the same for each team, is it?
They each have completely different resources, in terms of labour, materials, All dictated by their budget.

If they all had the same budget - i would 100% agree with you. But they don't.

[Richard]

The term "evolution" isn't exclusive to the natural world, it can apply to other fields. Wiki calls it "Universal Darwinism" http://en.wikipedia.org/wiki/Universal_Darwinism

Here are examples in the context of design development and micro biology.

http://www.cam.ac.uk/research/news/desi ... -evolution
http://www.ncbi.nlm.nih.gov/pubmed/2129 ... t=Abstract

There's also a whole field of genetic algorithms that use evolutionary principles.

[fiohaa]

The term "evolution" isn't exclusive to the natural world, it can apply to other fields. .

which is why i said in my last post:

Its not just termed 'evolution' which on its own can mean a lot of things.

Okay - so what was the original context of evolution we were talking about in? well......you brought it up, with this:

Unfortunately what we are seeing is natural consequence of evolution, the clock can't be turned back.

A journal extract from the abstract of an article entitled: Universal Darwinism, its scope and limits....

My argument is based on hierarchical functional descriptions of natural selection. I suggest that natural selection ought not to be thought of as a single process but rather as a series of processes which can be analysed in terms of a hierarchy of functional descriptions (in much the same way as many people think of cognition). This, in turn, casts doubt on the idea that it is possible in principle to settle debates about whether particular phenomena count as instances of natural selection.

source: http://link.springer.com/chapter/10.100 ... 7-3983-3_5

i can think of 1 other instance where they have tried to apply the darwinian process of evolution, and thats in economics and free market capitalism, with disastrous results (so far)
http://thedailybanter.com/2012/06/free- ... es-darwin/
http://adamsmithslostlegacy.blogspot.co ... ation.html

as i say, my main argument is simply that because all 11 teams have very different resources, they are not competing in the same environment, and so any kind of improvements made cannot be pinned down to an evolutionary process - of any kind.

F1 today is more akin to a mixed economy - a central body (FIA/Government) that works with the teams (companies) in drawing up the rules (regulations), and just like the state of world economies today, none of it is down to a process of evolution, of humans, or systems, or interactions, or mutations. It is 100% man made, based on decisions and ideology (mainly starting in the 70's with Reagan/Thatcher, but I digress)

example. Ferrari that gets a special cut, just for being Ferrari.
Redbull getting round the RRA by having a parent and a subsidiary.
The regulations being fiddled continuously by the FIA, to try and appease all the various conflicting demands of all the teams.

in this instance, its just not possible to apply any kind of evolutionary theory to a sport like F1, where there are central decisions being made constantly. I'm sure with running or swimming or cycling you could, as everyone works harder, trains harder, reaching the human bodies physical limits, the gaps become tighter and tighter to the point where it would take an external force to change the game. Yes all these sports have governing bodies but they hardly manipulate the sport in the same way as the FIA does with F1.

but please, feel free to point out 1 period in F1 where you can genuinely say 'X team/teams have reached their absolute limit' and a rule change caused a major change in competition?

The biggest rule changes came in 2009 for a long time - and look what happened? Brawn won - Ex Honda - Had the biggest budget and resources.
Became Mercedes, had to slash its workforce, probably half as much funding......look where they are now.
Who won in 2010, 11, 12? Red bull. Who has the most money? Red bull.

You could cite the 2005 season as an example, rule change came in, put Ferrari/Bridgestone on the back foot (so....somehow in your world that would be analogous to a plateau being reached in 2004, and the environment is changed to mix things up again). But then I could argue that

1. by that point a lot more teams were on Michelins and they had a lot more testing data to prepare for 2005, so its purely circumstantial evidence.
2. the teams with the biggest resources still won. Its not like Jordan suddenly made a leap in performance, is it.

end.

[Stradivarius]

im sorry, but you just cannot make the same comparisons between what is happening in a sport, and the natural process of evolution. its absurd.

The way I see it, whether or not such a comparison is absurd, depends entirely on what I am trying to demonstrate through the comparison. I don't think it is absurd to compare different competitions and looking at similarities in how the competitors change in order to improve their competitiveness.

you have to distinguish between human and natural......its one of the fundamental topics with evolution by natural selection - Hence the theory is 'Evolution by natural selection'. The clue is in the title. Its not just termed 'evolution' which on its own can mean a lot of things.

there are so many human motivations, influences, and not to mention MONEY that shape F1 that deeming it to be a natural process is incredibly naiive.
Okay, teams learnt from lessons and improved on their creations - what on earth is your point? That in itself does not mean that eventually all the teams reach some performance wall.

The OP is trying to make a point that F1 is reaching some kind of performance plateau, and that its cyclical.

I'm arguing that this cannot be the case, due to points I've already made in previous posts. If his theory holds true, all the teams would eventually be competing for wins, and All be similar pace.
That is clearly not the case, why?

1. Massively different budgets (hence why teams should open up their accounts and we should do a Points/£ index)
2. because of this, its not a fair competition. That is a fact. None of the teams are operating in exactly the same environment.

This analogy would absolutely fine, if all the teams were operating in the same environment, each year, and the teams that 'survived' (did the best) were the ones that adapted. Okay......but the environment is not the same for each team, is it?
They each have completely different resources, in terms of labour, materials, All dictated by their budget.

If they all had the same budget - i would 100% agree with you. But they don't.

Maybe I have not interpreted you correctly, but I fail to see why this analogy is dependent on the competitors having the same environment, the same resources and the same oportunities. Does a bear have the same resources as a deer? Does a short giraffe have the same oportunities as a tall giraffe? Does a tree growing on a peak have the same oportunities as a tree growing in a dip in the terrain?

Whether or not you would like to call it evolution, I think it is fair to say that the teams and their cars do evolve. Of course, there are some aspects to this which are not identical to the evolution of species. For example an animal (except for humans) can't observe a different animal and copy features which it thinks will be beneficial, like most teams copied McLaren's F-duct or Red Bull's exhaust. But here we are discussing what the two have in common. There is a need to perform well in a competition in order to achieve success and this will drive the evolution of both plants, animals and cars. But of course there are differences. In biology "success" is the same as getting reproductive offspring so the blood line goes on, while in f1 success is the same as scoring good results and we may see competitors retire despite being the most successful (like Alain Prost after winning the 1993 title).

The variations may be random (stochastic) both in nature and in the wind-tunnel. But the process of evolution is not random, as it systematically tends to reward the most successful variation. So even if the variations that are "tried" are stochastic variables, the long term result may be deterministic.

If we look at a sufficiently large time scale (with unchanged rules), richard_leeds is right about hitting the wall. Even the differences of the team's budgets won't change this, as even Caterham and Marussia will sooner or later have spent enough money and obtained enough knowledge to utilize 99.9% of the potential of the formula. Even if Red Bull and Ferrari then are at 99.99%, the difference is next to nothing. Now, the rules are not likely to remain unchanged for so long, but that doesn't mean the logic doesn't apply.

[bhall]

[...]

as i say, my main argument is simply that because all 11 teams have very different resources, they are not competing in the same environment, and so any kind of improvements made cannot be pinned down to an evolutionary process - of any kind.

[...]

I think you've missed the forest for the trees here. The evolutionary changes to F1 aren't exemplified within the teams; they're seen in the sport as a whole. Look at any rule change, a variable that affects all teams, and you'll see very clear environmental factors driving that change in one direction or another. I'll leave it to someone else to point those out.

The reason the sport is currently on an evolutionary plateau is because it's been decided, for a variety of reasons, that F1 is only viable within an environment where a maximum of X dollars can be spent to produce cars that go no faster than Y MPH and have a safety factor of Z, and the rules have been formulated within those environmental constraints. Right or wrong, this was all done for the "survival" of the sport.

Change the variables, and the sport will evolve. Cement them, and the sport will stagnate. This is very simple stuff.

[Richard]

I wish I’d never used that e---- word! I was using it to refer to optimisation of engineering design. All I know is that my colleagues and I work with professors and researchers who refer to evolutionary optimisation and genetic algorithms.

http://scholar.google.co.uk/scholar?q=e ... algorithms

I also see that I used the word “natural” in the context of “inevitable” or “logical”, i.e. the natural consequence.

So let’s put it down to something lost in translation. Now lets step away from the e---- and n---- words.

The main point is that the engineering in F1 has matured so we are in the plateau at the top of the S. The FIA try to change the competitive environment to push things back down the S but the teams rapidly optimise to climb back up to the plateau. We can see this when teams bring out new cars every February that have been developed in secret yet they end up with near identical laptimes. They have so many boffins and terrabytes of computer simulation that they converge on near identical performance for a given level of investment.

Essentially we have a tussle between engineers’ convergence on the optimum performance versus the desire of administrators for divergence in the quest for more exciting competition.

This level of optimisation means we don’t have the speed differentials required for overtaking. It is compounded by optimised aero being so sensitive to dirty air. That’s why spec series have more overtaking, their aero design is sub-optimal so less sensitive to dirty air.