Is Perez too far away from Max?
- Like
- Digg
- Del
- Tumblr
- VKontakte
- Buffer
- Love This
- Odnoklassniki
- Meneame
- Blogger
- Amazon
- Yahoo Mail
- Gmail
- AOL
- Newsvine
- HackerNews
- Evernote
- MySpace
- Mail.ru
- Viadeo
- Line
- Comments
- Yummly
- SMS
- Viber
- Telegram
- Subscribe
- Skype
- Facebook Messenger
- Kakao
- LiveJournal
- Yammer
- Edgar
- Fintel
- Mix
- Instapaper
- Copy Link
Methodology
I decided to analyze the qualifying performance of the Red Bull drivers since 2017 when Max completed his first Formula 1 season. To start, I removed the sessions where both drivers were not participants. An example of this is the first race of the 2021 season. At the Bahrain GP, Sergio was eliminated in Q2, therefore only the Q1 and Q2 sessions were used to compare his performance against Max’s.
There are some challenges when doing an analysis like this one. It is hard to compare the performance of different drivers at different tracks in different years. You can’t compare their lap times in their original units (seconds) because the first four races of the calendar have not been the same in every season. So, what’s the best way to get a representative comparison?
I came up with two main options. First, using some sort of percentage change to quantify the speed of each driver. The units of percent delta are fairly easy to interpret (Driver 1 was 1.04% faster than Driver 2 at Bahrain for example). The percentage change normalization comes with a couple of drawbacks, however. First, it doesn’t control the spread of the data. For example, some tracks will naturally have more drivers having more or less the same lap time. Think about short tracks where it’s hard to get a massive advantage over the competitors in a single lap. Second, the percentage change of driver a vs driver b doesn’t contain the data of how each driver did against the rest of the grid, not only within the team.
I concluded that these two drawbacks were not acceptable, and decided to standardize the data. The process is quite simple. First, I obtained the average time done by all the drivers in every quali session analyzed. Eg, for the 2021 Emilia Romagna GP, I obtained the average lap time done in Q1, Q2 and Q3. This includes the lap times done by every driver who participated in each session, not only Red Bull drivers. Second, I subtracted this average time from all the lap times done in that session. Finally, I divided that resulting number by the standard deviation to end up with the standardized lap time.
An example of the standardization process is shown in the table below.
2021 Emilia Romagna GP | |||||
---|---|---|---|---|---|
Quali summary - Red Bull | |||||
Driver | Lap time | Session lap time | Lap time - average session lap time | Standardized lap time | |
Average | Standard deviation | ||||
Q1 | |||||
Perez | 01:15.394 | 01:15.573 | 0.508 | -0.178 | -0.35 |
Verstappen | 01:15.108 | 01:15.573 | 0.508 | -0.464 | -0.913 |
Q2 | |||||
Perez | 01:14.716 | 01:15.073 | 0.289 | -0.357 | -1.232 |
Verstappen | 01:14.884 | 01:15.073 | 0.289 | -0.189 | -0.653 |
Q3 | |||||
Perez | 01:14.446 | 01:14.743 | 0.256 | -0.297 | -1.16 |
Verstappen | 01:14.497 | 01:14.743 | 0.256 | -0.245 | -0.957 |
The results
Interpretation
Remember the process we just did? The results are shown in the chart above. A standardized lap time of 0 means that it was exactly the average time done during that particular quali session (Q1, Q2 or Q3). Negative times, which means dots shown to the left of 0, represent faster lap times than the mean (this is good!), while positive times represent slower lap times than the average.
The gradient bands shown stand for high or low-density areas. Darker areas show where most of the laps were recorded, while lighter areas show the opposite.
The horizontal colored lines represent the 66% and 95% quartiles. The thick line around the middle contains the middle 66% of the laps, while the thinner line spans 95% of the lap times done by each driver. Laps outside any of these two intervals could be considered as anomalies, that is laps that were perhaps a bit faster or slower than you would normally expect.
As usual, click on the image to view the “image” in high resolution. Since it is actually a vector, not a rasterized image, you can zoom in as much as you want. To get a hi-res png images click here.
Analysis
The final results are quite interesting. The numbers show that Checo has been the fastest teammate Max has had in the first four races of a season since the 2018 season when Danny Ric was still at Red Bull. Sergio has only been slower than average on once which was on the last race at the Spanish GP. On that occasion, the Mexican driver finished in 8th place, with a standardized lap time of 0.488. His standardized delta against Max currently stands at 0.46.
In contrast, by the fourth race of the season, Albon had been slower than average on 6 out of the 10 quali sessions in which he participated, having his worst showing at the Hungarian GP with a standardized lap time of 0.733 done in the Q2 session. The standardized delta between him and Max after 4 races was 0.97.
Pierre Gasly had an even worse start than both Perez and Albon when compared to the average standardized lap time. However, Max had a worse start in those four races too, reducing the delta between both drivers. Note that for Gasly’s analysis, I had to remove the 4th race of the 2019 session (Azerbaijan) since he was disqualified for exceeding the 100 kg/h fuel flow limit during qualifying. In the three races that were analyzed, Gasly never managed to be faster than the average lap time of the session, while Max was faster than the average on every single quali session in which both drivers raced. The standardized delta between the Frenchman and Max after 4 races was at 0.72.
Finally, the 2018 session was the last time Max was challenged by a teammate. During the first four races of the season, Ricciardo was slower than the average only twice, once at the Q1 session of the Chinese GP, and once at the Q2 session of the Azerbaijan GP. Max on the other hand was slower than the average once, at the Q2 session of the Chinese GP. In that period, Daniel managed to beat Max on 2 occasions. First, at the Q2 session of the Chinese GP, and second at the Q3 session of the Azerbaijan GP. The final delta between both drivers during that period was 0.33.
I won’t go into detail about the 2017. At that time, Max was still a rookie, while Ricciardo had been racing in F1 since 2011. The fight between both drivers was quite even, with the standardized lap time delta being of just 0.04.
Conclusion
Max has always been a superb qualifier and has been beat just a few times during his already long career in F1. It is hard to envision any driver jumping into the second Red Bull seat and start beating him consistently. Having said that, Sergio has been showing decent one-lap pace during his first four races at Red Bull. His numbers may not be superb, but he has been consistently stronger than both of Max’s previous teammates, Pierre Gasly and Alex Albon.
If you guys liked this analysis, let me know so that I can keep doing it throughout the season. I am currently working on a similar analysis for the rest of the grid as well, so hopefully this article has been informative.
As usual, if you enjoy visiting my site, please consider sharing my posts on social media or with friends and family. If you want to support me, I would appreciate any donations that you want to make.
Great analysis
Excelent!
I like the concept that you have adopted but could you redo it and just use Q3 session info. Quite often top teams (Merc, RBR and in hte past ferrari) do not push themselves to the limit in Q1 and Q2 which will influance the analysis. Using data from just Q3 might give a better representation?
Hello Danny
While normally I do agree with your opinion, I feel like this season the competition is tighter than in the past few years. Just take a look at Daniel Ricciardo. He has a strong car, but he has been unable to get into Q3 lately. I think that this wouldn’t have happened one or two seasons ago.
Also, the issue with just using the Q3 data is that we lose all the information of all the teams that rarely, if ever, get into Q3 (see Haas for example).
Having said that, I added an additional table in my latest post which should give you a good idea of how the battles have developed in each quali session. The table is under the section “Standardized lap times: Session by session” of my latest post 2021 Season quali battles: Rounds 1 to 6.
Let me know what do you think about this solution.