First of all, if you scroll down you'll notice that this article is ridiculously long! I apologise for that, I knew I was undertaking a massive project but I didn't expect it to get quite so big when I started writing it up. It seemed a shame to get rid of large chunks of what I've written though so I've left it, maybe consider this as seven articles instead of just one!
Because of this, once you've got the general idea of what I'm doing from this introductory post I'm going to suggest you ignore everything else and skip to the summary post (using the contents page below), which is long by itself but contains what I really wanted to show. If you're still interested after that feel free to flick back and have a look at the other posts, you may want to start with 2012 since that is fresh in the mind and maybe also the drivers post.
Before we get going, massive thanks to Brogan for providing me with the data I needed. Also to Galahad and jez101 , who discouraged me from another idea I had to extend this further by suggesting I might need money to do it!
CONTENTS
1) Introduction (this post)
2) 2008 Analysis
3) 2009 Analysis
4) 2010 Analysis
5) 2011 Analysis
6) 2012 Analysis
7) Summary Part 1
8) Summary Part 2
9) BONUS! Ferrari, Red Bull and McLaren partnerships from 2010-2012.
INTRODUCTION
As we all know, car development is a massive part of F1. I think it was DC who said F1 is 80% car 20% driver or something along those lines. Put Alonso in a HRT and he's not going to be getting points but stick Karthikeyan in a Red Bull and he probably would... at least once. Every tenth counts and the engineers should really get as much attention as the drivers. With that in mind, this article is all about looking at who is winning the "development race", i.e. not only who has the fastest car, but also who is finding the most time during the season and which teams have been working their way up or down the grid in the last few years.
Starting with 2008, the last year before the current regulations came in to place in 2009, I'm going to try and answer questions like... Do the big teams really out develop everyone else? Are the new teams catching up to the midfield? Do particular teams tend to be strong at different stages of the season? Who gained/lost the most from certain rule changes? etc. I think questions like these are things we all have a general idea about but maybe the stats will give a different picture... In most cases we will find the "common sense" result, but even then it's interesting to extract real numbers from the statistics.
As a little bonus, I've also had a brief look at Alonso vs. Massa, Vettel vs. Webber and Hamilton vs. Button from 2010-2012. This isn't really the focus of the article, but I suspect people will be interested to look at (and argue about) the results!
WHAT I'VE DONE
Whenever you look at something like this the most difficult thing is to try and remove the aspects you're not interested in (e.g. driver errors) to leave what you really want. In practice this is impossible but I've tried my best to give the whole picture by looking at both qualifying and race performance in different ways. In essence, I'm using qualifying to measure the ultimate, best possible, performance of a team and the races to measure the 'true' performance (e.g. taking in to account that a car breaks down all the time or that a team can only afford a driver who crashes all the time).
Qualifying:
For qualifying I look only at the fastest time set by a team (i.e. the best driver only) in any of the three qualifying sessions. In 2008/9, when the teams had to run fuel in Q3, this is typically the fastest times from Q2 which does mean you have to be a bit careful about how you interpret these years (fast teams may not have shown their true pace in Q2). I completely exclude all qualifying days in which rain affects any session (we don't want our 'ultimate pace' to be influenced by topsy turvy results).
Races:
We're interested in everything here, reliability, adapting to conditions... everything. I use the sum of a team's points and include wet races. Also, in 2008/9 the 10-8 points system was used. To be able to compare the results in these years to 2010-2012 I recalculate the points in all the races using the current 25-18 points system.
SOME OTHER BITS AND PIECES...
5 Race Average:
Originally one way in which I was planning on presenting results was using a 5 race average. In the end i found that looking at quarters (below) showed what I wanted so I only kept this for the brief look at drivers at the end as it was quicker to do. In these plots an entry for race 3 is the average of race 3 along with the two preceding and following races (i.e. race 1 to race 5). Entry 10 is the average of race 8 to race 12, and so on.
Quarters:
The other way I'll present results is in quarters of a season, just to generalise things further and make it easier to draw conclusions. In 2012, with a 20 race season, this is straight forward, each quarter contains 5 races. For other season lengths it requires a bit more thought. For a 19 race season, for example, the first quarter ends at 4.75 races. This means I associate 75% of the fifth race with the first quarter and 25% of the fifth race to the second quarter and so on.
Different Circuit Lengths:
I haven't adjusted for this. I think it makes the results more understandable and it's not sure it's as easy as saying shorter circuits have smaller differences between qualifying times.
Team Names:
In the posts for each year I use the team name that was actually used in that year, so Honda and Brawn appear, for example. In the summary post at the end I just use the names used in 2012, so Mercedes means Mercedes from 2010-2012, Brawn in 2009 and Honda in 2008.
Because of this, once you've got the general idea of what I'm doing from this introductory post I'm going to suggest you ignore everything else and skip to the summary post (using the contents page below), which is long by itself but contains what I really wanted to show. If you're still interested after that feel free to flick back and have a look at the other posts, you may want to start with 2012 since that is fresh in the mind and maybe also the drivers post.
Before we get going, massive thanks to Brogan for providing me with the data I needed. Also to Galahad and jez101 , who discouraged me from another idea I had to extend this further by suggesting I might need money to do it!
CONTENTS
1) Introduction (this post)
2) 2008 Analysis
3) 2009 Analysis
4) 2010 Analysis
5) 2011 Analysis
6) 2012 Analysis
7) Summary Part 1
8) Summary Part 2
9) BONUS! Ferrari, Red Bull and McLaren partnerships from 2010-2012.
INTRODUCTION
As we all know, car development is a massive part of F1. I think it was DC who said F1 is 80% car 20% driver or something along those lines. Put Alonso in a HRT and he's not going to be getting points but stick Karthikeyan in a Red Bull and he probably would... at least once. Every tenth counts and the engineers should really get as much attention as the drivers. With that in mind, this article is all about looking at who is winning the "development race", i.e. not only who has the fastest car, but also who is finding the most time during the season and which teams have been working their way up or down the grid in the last few years.
Starting with 2008, the last year before the current regulations came in to place in 2009, I'm going to try and answer questions like... Do the big teams really out develop everyone else? Are the new teams catching up to the midfield? Do particular teams tend to be strong at different stages of the season? Who gained/lost the most from certain rule changes? etc. I think questions like these are things we all have a general idea about but maybe the stats will give a different picture... In most cases we will find the "common sense" result, but even then it's interesting to extract real numbers from the statistics.
As a little bonus, I've also had a brief look at Alonso vs. Massa, Vettel vs. Webber and Hamilton vs. Button from 2010-2012. This isn't really the focus of the article, but I suspect people will be interested to look at (and argue about) the results!
WHAT I'VE DONE
Whenever you look at something like this the most difficult thing is to try and remove the aspects you're not interested in (e.g. driver errors) to leave what you really want. In practice this is impossible but I've tried my best to give the whole picture by looking at both qualifying and race performance in different ways. In essence, I'm using qualifying to measure the ultimate, best possible, performance of a team and the races to measure the 'true' performance (e.g. taking in to account that a car breaks down all the time or that a team can only afford a driver who crashes all the time).
Qualifying:
For qualifying I look only at the fastest time set by a team (i.e. the best driver only) in any of the three qualifying sessions. In 2008/9, when the teams had to run fuel in Q3, this is typically the fastest times from Q2 which does mean you have to be a bit careful about how you interpret these years (fast teams may not have shown their true pace in Q2). I completely exclude all qualifying days in which rain affects any session (we don't want our 'ultimate pace' to be influenced by topsy turvy results).
Races:
We're interested in everything here, reliability, adapting to conditions... everything. I use the sum of a team's points and include wet races. Also, in 2008/9 the 10-8 points system was used. To be able to compare the results in these years to 2010-2012 I recalculate the points in all the races using the current 25-18 points system.
SOME OTHER BITS AND PIECES...
5 Race Average:
Originally one way in which I was planning on presenting results was using a 5 race average. In the end i found that looking at quarters (below) showed what I wanted so I only kept this for the brief look at drivers at the end as it was quicker to do. In these plots an entry for race 3 is the average of race 3 along with the two preceding and following races (i.e. race 1 to race 5). Entry 10 is the average of race 8 to race 12, and so on.
Quarters:
The other way I'll present results is in quarters of a season, just to generalise things further and make it easier to draw conclusions. In 2012, with a 20 race season, this is straight forward, each quarter contains 5 races. For other season lengths it requires a bit more thought. For a 19 race season, for example, the first quarter ends at 4.75 races. This means I associate 75% of the fifth race with the first quarter and 25% of the fifth race to the second quarter and so on.
Different Circuit Lengths:
I haven't adjusted for this. I think it makes the results more understandable and it's not sure it's as easy as saying shorter circuits have smaller differences between qualifying times.
Team Names:
In the posts for each year I use the team name that was actually used in that year, so Honda and Brawn appear, for example. In the summary post at the end I just use the names used in 2012, so Mercedes means Mercedes from 2010-2012, Brawn in 2009 and Honda in 2008.