The pace of development did not let up during the second season of Formula 1's higher-downforce regulations, and there was a very intense battle at the front - with Mercedes and Ferrari going tit for tat in terms of car performance for much of the season.

In the end, Mercedes pulled away and won both world championships, and this reflects the fact that it wasn't just about car development but also about team development. It wasn't just about making the car quicker, it was about the key decisions they made, how they understood the tyres and how to optimise the car.

To better understand the ups and downs of the battle throughout the field, I take the fastest single lap set by each car on a race weekend expressed as a percentage of the outright fastest.

This can then be averaged out over a season to get an overall percentage performance with each of the 21 circuits equally weighted.

The 'supertime' results unsurprisingly put Mercedes at the top, but in some areas the order differs from that of the constructors' championship.

With such a big gap between the top three teams - Mercedes, Ferrari and Red Bull - and the rest, there were really two different championships going on.

First, let's look at the big three teams.

1 Mercedes

Average performance pace: 100.119%

Mercedes decided to stick with what it knew this season with a car that built on its 2017 machine, which proved to be the right choice.

But it still wasn't easy, and Mercedes' great strength was how it learned from its struggles in the first half of the season to leave Ferrari behind in the crucial second half of 2018. The depth of understanding was what made the difference.

Just as it found last season, tyres were a challenge early on. It's all about the compromise between getting them working for a single qualifying lap and over a race stint.

There are pressure minimums so you have to get the tyres warm before you go out and the driver can then go slowly to let the rear temperatures come down.

But you can't fit slow laps into a 20-lap race stint and Mercedes had some difficulties achieving this compromise at times. But despite that seemingly being harder for it than Ferrari, Mercedes got on top of it and won the championship.

There was lots of talk about the team's wheelrims, which appeared at Spa. These were an attempt to avoid too much temperature build-up in parts of the wheel, which would transfer to the tyre. Mercedes did stop running the design for a while, even though the race stewards ruled it legal.

It's clearly not a moveable aerodynamic device because a wheel has to move, and it was an innovative way to control the build-up of temperature. I'm all for this kind of innovation, it was great stuff.

On the power unit side, Mercedes made good steps with the upgrades it introduced at the French and Belgian Grands Prix, but this season proved that the advantage it had in the early years of these engine rules has been eroded by Ferrari.

2 Ferrari

Average performance pace: 100.237%

Everyone at Ferrari has good reason to be kicking themselves, because the car was good enough to have at least given Mercedes a tougher run for its money - and with that extra pressure on Mercedes, Ferrari might just have beaten it to the championship.

Things always do go wrong during a season, but Ferrari didn't respond well to them and the same mistakes were repeated. That's not just for Vettel, who threw away too many points, but also for the team.

Ferrari made bigger changes to its car concept this year, switching to a higher-rake design and lengthening the wheelbase. The longer wheelbase meant it lost a little of the advantage it had at slower tracks last year, and there were times when the high rake seemed to cause confusion.

Having lost its advantage at the slower tracks, the high rake was supposed to make up for this. The aim is to get the front wing lower, particularly for the low- and medium-speed corners, to counteract the tendency to understeer.

Then, at higher speed, you want the downforce centre of pressure to move backwards giving more rear stability. While Mercedes stuck with what it knew, Ferrari changed and it took a while to get on top of it.

Many times early in the season we saw a struggle to get the car working at its best on Friday, although big steps were often taken for Saturday. But later in the season, Ferrari lost its way with the floor changes it experimented with in the run of races starting with Sochi.

This is where understanding the concept is so important, as with a high-rake car changes to the floor can have a big impact on the way you are sealing the underfloor. This is crucial to the diffuser performance and in turn overall car performance. The fact that Ferrari had to roll back these changes says it all.

There were inevitable rumours about what Ferrari might be doing with its power units, as the performance was strong on the straights in qualifying in particular. But I don't believe Ferrari was doing anything illegal - and neither did the FIA.

3 Red Bull

Average performance pace: 100.797%

Red Bull's RB14 was definitely a good car, but whether it was the best if you allowed for the power deficit of the Renault engine is very difficult to be sure of.

As McLaren has realised this year, it's not simply a case of putting a more powerful engine into a car. There are always compromises when fitting a different engine.

When Red Bull was dominating F1 with Renault, it didn't have the power of the Mercedes or Ferrari but the Renault required less cooling and the car was optimised for downforce, never really performing that strongly on the straights.

The same seems to be the case now. So much cooling is required with these engines that even another 30bhp could mean some of the airflow you use for downforce has to go to cooling, which means the compromise changes.

So putting a Mercedes or Ferrari engine into a Red Bull isn't necessarily that simple. The Red Bull was a well-balanced car and, as a result, very good on its tyres. This shows a good understanding of how to control the aerodynamic characteristics to be strong in low, medium and high-speed corners.

The car also rides the kerbs very well, and that combined with good corner speed also means it was slightly less aggressive on the tyres when putting the power down out of corners.

We did see a lot of reliability problems and Red Bull was very good at throwing Renault under the bus when this happened. But Red Bull had its own problems in this area too.

Overall, we saw how good the Red Bull was when it dominated in Monaco and Mexico. The key question now is how the Honda relationship works, because that's a company that won't accept the same treatment Renault has had.

4 Haas

Average performance pace: 101.970%

Haas made a very good step this year because it's still a relatively new team on the block. It has stabilised a reasonable amount and is going in the right direction.

The car resembled an updated version of last year's car rather than a new car, but this was a conscious decision that helped prevent the team from getting lost.

There were occasions when Haas was right up there at the front of the midfield, but more consistency is required, and there were too many points thrown away. The team will keep tidying these things up, so although there is a little overall disappointment within the squad, its performance level was impressive.

Now there needs to be more understanding of the strong and weak points to level out those performances. But overall, for a team that's only in its third season, Haas did a good, solid job.

5 Renault

Average performance pace: 102.137%

If I was technical director at Renault, I wouldn't be that pleased with the season despite finishing fourth in the constructors' championship.

The car was too erratic over the course of the year and there needs to be real understanding of why that happens for Renault to take the next step.

Renault has added a lot of good people as it expands, but that in itself is a big challenge. Making all that work as well as it does at Mercedes is what Renault needs to do to reduce the gap to the top three teams, which after all has to be the objective as a manufacturer team.

The power unit deficit is also a problem and it has been since these hybrid regulations were introduced some five years ago, but it's not just the engine that needs to improve. It's also the team and the car development, because Renault is being measured against the big works teams.

Overall, Renault didn't close gap to the top three this season, and needs to do so next year.

6 Force India

Average performance pace: 102.207%

Force India's season can only be judged in the context of the financial problems that held the team back before it was saved in the August break.

Technical director Andrew Green, who I've worked with in the past, is very good at making sure things are well researched and understood before getting to the track.

That means these upgrades worked, but not being able to produce the parts that were designed for financial issues will have held the team back more than most as a result.

The early-season car was a basic package just to get going, then the developments slowed down because of the lack of funds. It took time to get on top of the car, and the team did that well.

If you look at the performance, Force India still managed to be one of the most consistent teams. This was a good effort in difficult circumstances.

7 Sauber

Average performance pace: 102.603%

Sauber was very impressive. After a poor start to the season, I expected the team to take longer to understand its relationship with Ferrari and to benefit from the improved resources from Alfa Romeo, and use them wisely.

But by the end of the season, it was consistently challenging at the front of the midfield, which was a wake up call of what can be done to the other teams around it.

Problems with cooling in the early races did hold the team back, which once solved played a part in the rapid improvement. We also saw some good detail work on the car, suggesting the aero department is working well again.

Sauber has the chance to be a real force at the front of the midfield next season after the dramatic improvement this year.

If that momentum can be carried over, with Kimi Raikkonen returning next year, 2019 could be a big season for Sauber.

8 Toro Rosso

Average performance pace: 102.805%

Toro Rosso continued to be an inconsistent performer this year. While you could suggest that's down to Honda, the team has been like this for some years now.

It makes you wonder if Toro Rosso has really understood its underlying problems. The new relationship with Honda seemed to work very well. Yes, there were too many penalties, but at the beginning of the season Toro Rosso said it was investing in 2019.

But the chassis performance level was too inconsistent - it had some strong performances at the front of the midfield but unfortunately others that were way down the order. Usually that kind of thing is down to the team.

Grid penalties didn't help, but the car performance need to be more consistent. The loss of technical director James Key won't have helped as he fitted into the team very well, so it will be interesting to see how things go under a new technical leadership.

9 McLaren

Average performance pace: 102.923%

McLaren came into the season with high hopes for its Renault-powered car, and it was certainly the time for the team to stand up and be counted.

Unfortunately, what happened was that McLaren's weaknesses were exposed. You have to measure McLaren's performance compared to Red Bull, and the gap between the two shows how much progress needs to be made.

The McLaren factory is a big old set-up and it has the potential within if used correctly. The biggest problem is that the McLaren got worse over the season as a development ceiling was hit.

Eventually, it became clear that this was down to not understanding the wheel wake and downforce being lost through the corners.

But needing downforce in corners is hardly a new thing, and this does raise some concerns about the way the aerodynamic specification and development is being done. But you always learn more when things aren't working.

So if McLaren can learn from these lessons, a step forward for next year is on the cards. If not, heads will roll.

10 Williams

Average performance pace: 103.606%

At the start of the season, the Williams looked like a big step forward because the car was much more complex and closer to the kind of thing we see from the top teams.

But anyone can copy the concepts, it's about the understanding of them and it's clear that Williams was lacking in that area. The progress made over the season was disappointing.

Although some gains were made on the stability of the car under braking and it added a few small developments, there was never much of a breakthrough.

Let's hope the lessons of these problems have been understood and learned and next year's car will be an improvement.

To get a more balanced picture of the development war, we have divided the season into five four-race blocks, with Australia acting as a standalone starting point. This shows the third of those blocks - races 14 to 17 -was where Mercedes really pressed on and Ferrari struggled.

At the back, McLaren faded as the season progressed, while Sauber moved from the back to the front of the midfield rapidly.

Sauber was the big winner this year compared to last year, with a 1.5% gain turning it from a backmarker last year into midfield leader by the end of 2018.

The only other team to make a significant relative gain was Haas, which also moved to the front of that battle. At the bottom, the big losers were McLaren and Williams.

The switch from Honda to Renault actually led to McLaren being 0.413% slower this year, while Williams was 1.081% back from last year. As for Mercedes, it was actually fractionally worse off.

But when you are at the front, as it was last year, there's nowhere to go except to stay the same.

And while Ferrari posed a bigger challenge in 2018, Mercedes remained F1's strongest operation on and off track.