2019 Australian GP: Is Ferrari’s power unit broken?

Time for Ferrari again. I was planning on doing an analysis of the race leaders, but I thought that this would be a good time to get some data from Ferrari and see if it indicates that something could have been wrong with the Ferrari power unit.

If you are an experienced F1 fan, you can skip this small introduction about modern power units and go to the fourth paragraph. If you are fairly new to the sport, then it is worth mentioning that modern F1 engines are hybrid, that means that they have an internal combustion engine (ICE) that works with fuel, just like normal cars, and an energy recovery system (ERS) which consists of an motor generator unit – heat (MGU-H ) and a motor generator unit – kinetic (MGU-K). While most of the power of the unit comes from the ICE, the ERS allows the cars to harvest power from their brakes and from the heat lost from the engine. Unlike the ICE, the ERS is charged and recharged every lap, and teams have to be smart about how to deploy the energy obtained from this system.

Durin the quali for the race, teams usually run highly aggressive engine and ERS modes, allowing the cars to reach top speed. In those cases, cars drain all the energy obtained from the ERS in just a single lap, but that is ok since they only care about doing the fastest possible lap. During the race however, they have to modify the way they deploy that energy, since if they use all the power in one lap, and another team does not, they can be in a disadvantage during the next lap.

One of the theories about why Ferrari did so poorly during the race, and not necessarily that bad during quali (it was pretty bad though) is that maybe Ferrari ran a very aggressive ERS mode during quali to compensate for a problem with the power unit. That means, that maybe during the race Ferrari could not run such an aggressive ERS mode, making them even slower. Fans on Reddit, and even Mark Hughes, a famous F1 journalist, speculated that this theory may have some substance behind.

Let’s take a look at a quick graph that I made to analyze the data and see if the numbers confirm this theory.

Delta between quali and race by team

My idea here was that if Ferrari, or any team with a Ferrari power unit for that matter, ran a very aggressive ERS mode during the quali, and not during the race, then the delta between the laps ran during quali and during the race would be higher than for other teams that use a non-Ferrari power unit.

I decided to run this analysis only for drivers that advanced to Q3, since it would not be fair to compare times done during Q2 with lap times done during the race. In this case, I got a mean time for first 5 and 10 laps of the race, and subtracted the fastest time done during Q3. As previously stated, a larger delta would mean that a car perhaps ran an aggressive and unsustainable ERS mode during quali.

The results show that the largest delta for the first 5 laps was seen with Lando Norris, the only driver to qualify to Q3 with a Renault power unit. Grosjean, running with one of the cars that has a Ferrari power unit, had the second largest delta, followed by Hamilton and Pérez, both running with a Mercedes power unit.

For the delta of the first 10 laps compared to the fastest quali time, Grosjean, Pérez and Leclerc took the top 3 spots. It is important to note that Leclerc had a terrible lap 9 (1:32.630), which made his average time increase by quite a big margin. If we remove that lap from the data, then his delta goes from 8.75 seconds to 8.48 seconds, a number that is more in line with the data obtained from the other cars with Ferrari power units.

If we look only at Ferrari as a team, there is no clear indication that they were running a more aggressive ERS mode than other teams were during Q3. Sebastian Vettel had the second lowest delta over both 5 and 10 laps, just trailing Max Verstappen (a Honda analysis may be required pretty soon huh?). Leclerc also had a low delta if we just consider the first 5 laps, but what about when taking the 10 lap average? Then his delta is the second highest after the one done by Grosjean. Having said that, if we remove his atrocious lap 9 and take his 8.48 seconds delta, he was still among the best, only trailing Verstappen, Vettel and Bottas.

Delta between quali and race by PU

If we analyze the data by combining every team with a certain PU, we get a clear picture of our results. Honda had the lowest delta (good for them!), while Renault had the worst delta. It is important to understand that we only have one data point for each team, so the results are not as robust as the ones of Ferrari and Mercedes.

A fairer comparison is the one of the Ferrari vs Mercedes PU. In this case, the Ferrari data set has 5 different drivers, while the Mercedes one has 3. Since we have more than 1 driver in each PU for these 2 team, I decided to add some standard error bars. In this case, we can see that the numbers are indeed quite even. Ferrari actually had a lower delta than Mercedes over the 5 lap distance, while Mercedes had better numbers over the 10 lap distance. Standard error shows no difference between the Ferrari PU delta and Mercedes PU delta over 5 or 10 laps.


Just from the data that we have, which is obviously a lot less detailed than what the teams get, I see no reason to believe that the Ferrari PU was running more aggressive ERS mode during quali that any other team that advanced to Q3. The delta between the Ferrari PU when comparing quali to race pace was very close to the one of the Mercedes PU.

Clearly there are many factors involved, with one of the most important ones being car setup. A car can be fast during quali, and slow during the race, or the other way around. Sometimes the car is slow during both quali and race due to setup problems, and that is something that only teams can analyze with their own data.

My personal conclusion from this data is that there was nothing really wrong with the Ferrari PU, at least not during the first 10 laps of the race. Cooling problems could diminish the performance by the PU at the middle/end of the race, however an analysis to determine that is a lot harder since you have to take the compound being used at that time, fuel load, as well as tire degradation.

I hope this helps you a bit to decide what do you think happened to Ferrari during the 2019 Australian Grand Prix.


Submit a Comment

Your email address will not be published. Required fields are marked *