Sundaram Ramaswami
Statistics are an essential part of Formula 1, a sport that’s all about numbers. Apart from teams who use the data to make their cars faster, stats are important for a viewer as well, as they elevate the overall race experience.
We sat down with the man behind most of the sport’s statistics over the past 18 years – Sean Kelly. He’s also known as the ‘Virtual Statman’ on social media.
Here’s how the interview unfolded –
Sundaram Ramaswami: What exactly do you do in the world of motorsport, especially in Formula 1?
Sean Kelly: When you hear a stat or historical information on air, usually I’m the source of that. That entails writing a lot of stuff in advance, prepping each broadcast, things that we know we’re going to need. For instance, what record is Lewis Hamilton likely to break this weekend? More nuanced stuff about the history of the race track, what’s happened on this date in history or things like that.
When the cars are not racing and in a normal non-coronavirus world, I’d be at the race track entertaining guests with Formula 1 Experiences and with the Formula 1 Paddock Club, and also some bits and pieces with Shell lately, where they entertain the guests for Ferrari. So it’s quite a full quota of things.
If you buy an official F1 race program, you’ll see my numbers in there as well. So, yeah, you’re stuck with me. Basically, even if you’re trying to avoid me, you can’t get away from it. So I’m going to bombard you with numbers one way or another.
SR: How were your early days as an F1 fan, and how did this interest in statistics and patterns come about?
SK: It was, in fact, an accident, and the first race I watched was the 1987 Brazilian Grand Prix. I had never seen a race in my life before, and I thought, “Oh look! All these wonderful, colourful cars and they make lots of noise. It’s very interesting. I’d like to watch another one.”
After that, I just kept watching the races and from that day forward, I’ve only missed two races. I didn’t make a conscious decision as such.
Like many people, I wanted to be a driver. But what I was really good at was absorbing the information from these races, and I thought everybody else was taking it in the same nerdy level that I was. It was only when I got to be a teenager that I realised, “No, not really. People are not studying this in the detail that I am, or at least they’re not retaining the information in the way that I am.”
My mum said that I should do something with that skill. And I used to go, “Yeah, right. Like anybody ever makes any money doing that.” Eventually, I had the sort of push factor to explore if it can be pursued. So at the end of 2002, I cold-called the American broadcaster at the time, which was the ‘Speed’ channel. I left them a voicemail message along the lines of, “It can be a much better show if you hired me.” To my astonishment, they actually called me back three days later and said, “I’m sorry, what was it that you wanted?” So I explained to them rather nervously and they asked me to send them an example.
I did that, and also said to them, ”Could you send me what you’ve been using up to now?”. They faxed it to me and I went through it with a sharpie and put corrections on it and sent it back to them an hour later. It was like a teacher marking a student’s paper. After that, they decided they’d take me and paid me $200 a race. There were 16 races in 2003, and they paid me $200 a race. So that’s how it started. I didn’t really have any idea that was going to be a career at the time.
Stat#1
The 1975 & 1982 Swiss GPs were held at Dijon-Prenois in France, while the 1997 & 1998 Luxembourg GPs were held at the Nürburgring in Germany.
SR: In my opinion, statistics or patterns are complex while it’s also very simple. We have a lot of numbers, a lot of different patterns, but you have to identify them manually. A machine cannot do it for you. But then it’s also simple as you have to present it in at most one or two sentences so that people can consume it easily. So what do you think about all the numbers that you find?
SK: What’s important with a TV broadcast than with a nuanced Formula 1 oriented website where you’ve got a market just of the really hardcore fans, you’ve got to think to yourself it must appeal to somebody who doesn’t usually watch Formula 1. They’ve got to understand this too.
It’s imperative that no matter how far we drill down into the numbers, it must be something you can understand in a sentence. For example – This is only the fourth time that this car has done this with this driver from this country since 1973. You’ve lost everybody, no one cares.
It has to be something more straightforward. He’s got the most wins in Formula 1 history. Anyone can understand that. Most pole positions in history, the 1000th world championship event. Everyone gets it. So it’s those things in a TV broadcast. You’ve got to boil it down to simplicity.
So that’s probably why I’ve become quite in-demand because I kind of err in the side of simplicity and you’ve got to keep it simple. It could be PhD level mathematics, but only people with a PhD degree can understand it and most people do not have that.
SR: How chaotic was the 2019 German GP and the 2020 Italian GP?
SK: Those races weren’t too bad, they are actually fun races to do because you have unexpected developments. When Lewis Hamilton is just going round and round in the lead, you can telegraph what’s happening. You know he’s going to win this race, and then he’s going to have the all-time wins record and you can see it from an hour away.
What’s more fun is when you have these unexpected races, like Hockenheim 2019 that you mentioned, like Monza in 2020, when suddenly there’s an unexpected event, where you think, “Oh wait a minute! That hasn’t happened since when?”
That’s really exciting. It wasn’t like this at the start of my career, but nowadays my job is to beat Twitter. That’s basically what my job is. I need to be quicker and more accurate than Twitter because that’s a free information service. But when you have these unexpected races, that’s where you earn your money.
Stat#2
Since 2015, Fernando Alonso has had the most retirements in F1 (30), followed by Romain Grosjean (28), Nico Hulkenberg (27) and Carlos Sainz Jr. (27)
SR: What are your favourite stats? Are there any particular ones that didn’t receive the sort of acclaim they should have?
SK: My personal favourite stat in my career comes from Monza 2003, the fastest Grand Prix in F1 history. A record that still stands. I was working for ‘Speed’ back then and this was before the streamlined service we have these days where everything is easily accessible. Back then, I’d call them from my house as soon as the race was ending and would tell them all the important stats from the race, which they would write down and pass to the commentators.
The 2003 Italian GP broke the average race speed record (247.585 km/h) that stood since 1971. In advance of the race, I had calculated the exact time, to the thousands of a second, for it to be a record. Eventually, the record was broken. So I called them up and told them and they went straight out with it even before Michael Schumacher reached parc fermé. Surprisingly no one else ran it that day. So on a personal level, that’s always been my favourite.
One other is Sauber completed their 500th world championship event at last year’s Turkish GP. Their team manager Beat Zehnder has only missed one session in 27 years. Not one race, not one day, one session. He missed FP1 at Monza in 2008 because he had to go back to Switzerland due to a family emergency. Monza is quite close to the Swiss border, so he drove there and got back again for FP2. That is the only session that he has missed out on in the 27 years that Sauber have had a team in Formula 1, which I think is an astonishing statistic.
Regarding all-time data, Jim Clark had participated in 72 Grands Prix, won 25, retired from 28 of them, and finished second just once. Ever. If memory serves me right it was the 1963 German Grand Prix. If anything sums up how brilliant Jim Clark was, it’s that invariably the car would win or he would retire. He didn’t have much in between. I mean even Lewis Hamilton doesn’t have that.
SR: A lot of us want to know how you maintain your database and how much of effort does it take for a particular race?
SK: As you mentioned, it’s quite painstaking to find all this stuff. People think that I have these algorithms and complex databases to do this for me. I don’t. I actually sit for maybe 20 hours or so before each race. That’s how many hours of work it is to get a race weekend together in advance (not including the race weekend itself). So you sit there and painstakingly go through it and you find patterns and go, “That’s interesting.” You get these things that an algorithm or a program won’t spit out because it doesn’t understand the context of things properly.
There is an argument to say you could do this automatically and yes, it would save me a lot of time. That’s definitely true. What it would also do is rob me of the need to study every single aspect of the numbers, and one way or another, you’re going to have to sit there and study it. If you want to get the most out of it, you have to study it. So it doesn’t really matter if you have an algorithm that updated everything automatically for you, you’d still have to sit there for just the same amount of time and figure it all out. It can be quite annoying, especially when we have back-to-back race weekends, as it can get quite hectic.
There’s no other way. You’re supposed to be the best in your job if you work in Formula 1, whether you’re the person who sweeps the floor, whether you’re the person who drives the car. I tend to remind myself fairly regularly in these times, that they can always find someone else to do this. If you really don’t want to do this, I’m sure there’s plenty of other people who’d be more than happy to stand in for me.
Stat#3
Two drivers from the 2020 Formula 1 season featured in the ‘A1 Grand Prix series’, a one-make championship for nations rather than teams, that ran between 2005 and 2009. Nico Hulkenberg representing Germany was crowned champion in the 2006-07 season, whereas Perez took part in one round for A1 Team Mexico that year.
SR: Do you have any tips for people who would like to pursue Formula 1 with respect to statistics or analytics?
SK: If you’re doing it like I’m doing it, for the entertainment, you don’t need the level of qualification that you would need for a team for instance. If you’re doing data analytics with a team, that’s a much more scientific approach. I don’t want to associate myself with that because that’s not what I do. That’s a far higher level of engineering and mathematical understanding than I have.
The three questions I always get. How did you get your job? Can I have your job? Do you need an assistant?
With regard to getting a job in F1, there are two things that you really need. The first is tenacity. You may not be the most qualified person, but you have to be able to roll with the idea that you won’t get in at the first go. It might look very easy in my case, but it still took me four months to convince them to hire me. The deal wasn’t done and it took a lot of persuasion over a period of time. On building my career, I got a lot of rejections, and that is normal in F1. You’ve got to be very thick-skinned, you’ve got to be prepared for the idea that some people will say no, it’s part of the game.
I’ve been rejected by people multiple times, most famously by the BBC. When the BBC got the rights in 2009, I actually had a phone call with the producer for the BBC F1 coverage, and it seemed like the deal was done and at the last moment, they decided not to do it. It was really upsetting because I really wanted to work for them. It’s the BBC, the network I watched in the 80s and it didn’t happen.
Then a year and a half later in Melbourne at Albert Park, my phone rings and I see it’s a Monaco number. Who’s calling me from a Monaco number? “Hello. Is that Sean Kelly? It’s David Coulthard.” He called me up to the booth and said, “I’ve heard about your work, could I try it? Because, you know, Martin Brundle and I are commentating.” Brundle goes, “You’re the ‘Virtual Statman’ are you?” All of a sudden I’m like, “What’s going on?”
So eventually I did get the BBC gig. It took me almost 2 years to finally get on board with BBC. And that will happen, you have to be ready to deal with the disappointment. It’s not personal. It doesn’t mean that you won’t eventually get there.
The other thing I would mention is to have the ability to fit the job that needs to be done at the moment. In my case, I used to be in the TV studio for years with the ‘Speed’ channel. I then did it at NBC but they, unfortunately, ended their F1 project altogether. So in 2018, I wasn’t in the studio anymore. I thought that this gives me all this spare time that I didn’t have before.
I thought I could go to the racetrack and entertain guests, so I then spoke with F1 Experiences and now instead of being in the studio, I’m at the racetrack talking to guests, as well as doing stats for the other broadcasters. So you have to change with time. Don’t assume that the job you have now will exist in perpetuity. You have to have that ability to survive and be tenacious. I don’t think this gets discussed enough. At least half a dozen times we’ve thought our career was over.
Brilliant insights and stories from the Statman, weren’t they? You can follow Sean Kelly on social media (@virtualstatman), and experience his magic around a race weekend.
If you love statistics, do check out our stats section where we post some mind-boggling stats and facts after each Grand Prix.
