F1 2022: Carlos Sainz Jr, Charles Leclerc, and the battle for the role of ‘lead driver’.
Ferrari’s Charles Leclerc and Carlos Sainz Jr topped the recent WTF1 poll of driver pairing quality¹, but who will be the number one driver when rubber hits tarmac in 2022?
The narrative says that these two vary in approach — Leclerc producing moments of magic let down by some bad luck or poor decision making; Carlos having a more quietly consistent season. Ultimately, this consistency resulted in The Smooth Operator pipping his teammate over 22 rounds with less than 12-months experience in the seat.
What’s the full story? Let’s take an in depth look at the data from 2021, through Saturdays and Sundays, to see if there’s any clues to how it’ll play out in 2022 before the opening race in Bahrain this weekend.
Data provided by Ergast API² and FastF1³
Before we begin, this is my first post of this kind. I hope to write more pieces — not necessarily F1 centric either. If you want to know more about who I am and what I’m trying to do here, check out my intro post below.
Qualifying Stats
As we do with any race weekend, let’s start with a bit of Qualifying before we take a look at the Race stats. One thing you’ll hear a lot about Leclerc is his outstanding one lap pace, with surprising back to back poles last season in Monaco and Baku. Do the overall stats match the narrative? Yes. But there’s more than meets the eye.
- When collecting the data, I noticed Carlos’s listed Q2 time at Saudi Arabia (1:53.652) was 26 seconds slower than Charles’s. This was down to mechanical issues caused by a minor accident. This is obviously a massive anomaly and not representative of the season, and results in a 1second+ average gap per weekend. I have elected to use the Sainz’s Q1 time of 1:28.236 to give a corrected average.
In past seasons, Leclerc has been proven to consistently outperform teammates on Saturdays and continued this trend in 2021. Over the course of the season outperformed Carlos in qualifying, beating the Spaniard in almost all considered metrics above. However, it’s actually astoundingly close.
This is not to say Carlos is any sort of slouch, his record against Charles is not nearly as one-sided as other teams on the grid — in fact they were the second most evenly matched in qualifying (13:9) after Alpine’s Ocon and Alonso (11:11). At 0.247s, the pairing had the closest average gap of any pairing. Not bad, when you narrative considers Charles to be a quali monster.
It could have reasonably been closer at 12:10, revisiting the events of Monaco. Fans will remember that after his provisional pole lap, Leclerc proceeded to boot his SF21 into the wall around the second chicane of La Piscine, red flagging the session. Carlos, on an absolute flyer and about to challenge for pole, was forced to abort his final lap. This handed Leclerc the front row start (if he ever made it) and left Carlos in a wholly unsatisfactory P4. Without getting into the potential cynicism of Charles’s shunt, F1 fans everywhere were left wondering: should there have been a different Ferrari driver in P1 on Sunday afternoon?
That being said, Leclerc found himself on the front two rows more often (P1-P4), therefore sitting in positions to challenge for the lead by grabbing an early tow from cars in P1 or P2 as well as displacing at least one RBR or Mercedes driver piloting a statistically faster car.
All this being said, Sundays are the days that matter in F1 and if you can’t convert a high quali finish into a hefty points haul, then What Are We Even Doing Here?
Race Stats
On a race weekend, one could say that the Race is the chess to Quali’s checkers.⁴⁵
- Includes Belgium, where technically there was no opportunity to improve qualifying position. LEC: P8, SAI: P10.
As such, Sunday’s tell a different story, with Sainz showing some outrageous consistency, not facing a single DNF. There’s only 22 opportunities to score points, so avoiding retirements greatly improves your chances of finishing higher in the standings. Leclerc had 2 retirements (one DNS in Monaco and one DNF in Hungary) which cost him dearly. Leclerc failed to score in 4 races in total. Another remarkable fact, Sainz scored points in all races but 2, where he twice finished one place outside the points in P11 (Portugal, France).
The above figures are probably the most important in this entire piece when it comes to showing the key difference between Carlos Sainz Jr. and Charles Leclerc.
As you can see, 3 of Leclerc’s non point scoring drives happened in races where Sainz achieved podiums, resulting in large point swings (Rounds 5, 11, and 15). The 4th big swing, and the most decisive, again came from a Sainz P3 in Abu Dhabi to net 15pts, where Charles finished in P10 with just one point after a botched strategy call to try and take advantage of a VSC — not strictly his fault.
Another strength of Sainz is his ability to convert starting positions into high point scoring races. On average, Sainz improved from grid to finish by +1.41 places. Leclerc’s average position change was -1.27 — skewed largely by his retirement at Monaco of course.
There is an argument to be made that, DNF/DNSs can be considered anomalous on the basis the uncertainty of the reason (on paper, retirements due factors outside the drivers control look the same as retirements due to driver error). Certainly Charles could do little to prevent getting caught up in the Bottas/Stroll bowling session into turn 1 at Hungaroring. Looking at the average points per non-DNF finishes in the table above, Leclerc does come out on top. More often than not, if both drivers finished in the points, Leclerc was taking more of them.
Taking it back to chess language if you’ll indulge me, Leclerc had a tendency to aggregate small advantages on Sainz round by round before occasionally blundering away his lead (through his own fault or otherwise).
A Quick Word on Shakedowns and Testing:
I’m of the general opinion that preseason testing tells the general public very little about the season will pan out — only the teams, who collect great swathes of data, will know have any clue about how their winter development has gone. As well as this, drivers within a team don’t perform identical tasks across the sessions, so it’s not easy to compare stints.
With that being said, there are some takeaways from Bahrain.
Leclerc did produce the fastest laps for Ferrari and the 3rd fastest across the whole test event. This was more than a second faster than Sainz’s best attempt.
Fastest Laps:
LEC: 1:32.415 (Tyre = C4), SAI: 1:33.532 (Tyre = C4)
The Scuderia, as many predicted, used their development time well and do look like potential title challengers — which will only fuel a rivalry further.
Final Thoughts and Predictions for 2022:
I’ve tried to make my predictions purely on the the data but it’s hard not to take into account the ‘intangibles’ — in particular, team politics. A factor to consider is how quickly Ferrari will adopt a “Lead Driver” if they find themselves in a position where one or both of their drivers are fighting for the championships. Due to their closeness last season, I don’t believe Ferrari should show any sort of favouritism in the early rounds until a clear leader rises to the top.
But if I put my political/business hat on, I think Ferrari will be more willing to adopt Leclerc as the lead driver, due to his enormous contract and golden-boy image.
If Leclerc can put in an early lead over Sainz in 4 races similar to what we saw last year, we may see a shift in resources and strategy to serve a potential title challenge — at the expense of Carlos. Conversely, I think Sainz would need to prove himself to be consistently better than Charles over many more races to see the same switch-up in resources.
I’m not a betting man but I do think it’s a helpful mental model to put percentage odds on outcomes.
Due to all the factors, and despite falling behind his teammate last year, I’m going to go with 60% odds the Leclerc comes out on top.
His ceiling certainly seems to be higher than Sainz’s right now and if he can maintain some consistency and avoid costly errors will probably come out on top. I’m certain that Carlos will not back down easy, so it’ll be another closer season.
Have any thoughts? Got any questions? Any topics you want me to have a look at? Get in touch with me on Twitter. This is the first of hopefully many pieces so feedback is always welcome and appreciated.
Here are some really useful tutorials I used as a starting point:
- This last one, while not F1 specific, really helped me with not getting unnecessarily bogged down with coding if there was a simpler alternative in Excel/Powerpoint.
References and Footnotes:
- Ergast F1 API — Digitised Season Statistics for all F1 Championships
- FastF1 — A python package for requesting and plotting F1 telemetry.
- WTF1 Podcast — Driver Quality
- It’s fitting to bring up chess in a piece about these two drivers. The pair could regularly be found battling over the board as well as on track. On chess.com, Carlos is currently beating Charles in rapid (10 minute games), with an Elo rating⁷ of 904 vs 769, respectively. Foreshadowing?
- “‘We got quite crazy with it’: Inside Formula 1’s driver chess club” — https://www.autosport.com/f1/news/we-got-quite-crazy-with-it-inside-formula-1s-driver-chess-club/6996077/
- All this talk of Lead and Support drivers has got me thinking — who is the best and most willing 2nd Driver of All Time? I’m talking about the driver doing the most to help their team, sometimes at the expense of their own race. Let me know if that’s something you’d like to read about.
- Elo Ratings can be applied to any Zero-Sum-Game (F1 included) and are also incredibly useful for establishing relative ability. It’s something I’m looking to apply to F1, so look out for that piece.
