How sharing performance data has made SailGP Season 3 the most competitive yet

Jonathan Turner
Jonathan Turner
March 16, 2023

How do you gain a competitive advantage over your rivals in a professional sport which has been inherently designed to constantly re-level the playing field, event by event?

That is the challenge facing the world’s best sailors who make up the nine international teams competing on the high-performance SailGP global yacht racing circuit. 

Traditionally, the top tier of professional yacht racing has been an intensely secretive environment, where incremental performance gains are both hard-earned and fiercely protected. 

Whether it be government-funded national sailing squads competing for Olympic medal success, or multi-million dollar syndicates challenging for some of yachting’s most prestigious trophies – such as The Ocean Race or the America’s Cup – these organisations have always had a tightly-drawn shroud of secrecy protecting their precious performance data from the ever-prying eyes of rivals. 


Breaking the status quo

This long-established status quo came under attack, however, in 2019 when, out of nowhere, along came SailGP with a disruptive strategy diametrically opposed to the traditional way of doing things.

From its very inception, SailGP’s founders – American software tycoon Larry Ellison and prolific title and championship-winning New Zealand yachtsman Russell Coutts – adopted a game-changing concept making the vast array of performance data that streams off the circuit’s fleet of identical F50 catamarans available to all the teams.

SailGP F50 catamarans racing at around 100 km/h during Saint-Tropez Sail Grand Prix.

It was a bold move that puzzled many observers at the time, but we now know that the reasoning behind this, until then, unprecedented data sharing policy was two-fold. 

Firstly, as part of SailGP’s self-declared bold goal of ‘changing the face of sailing’, allowing open data access meant the league organisers could harness the numbers to help explain the undeniably complex sport of yacht racing to a mainstream audience of sports fans.

Secondly, the ability of competing teams to access and analyse not only their own performance data, but also that of their rival teams, through the Oracle Cloud, made it easier for all crews to quickly move themselves up to the required level on the steep learning curve that enabled them to be competitive out on the water.

Because all the SailGP boats are completely identical, teams could quickly apply the insights they gained by poring over their competitors’ data to modify how they sailed their boat the very next time they raced. 

Matching performance data with time stamped video and audio meant the team’s coaches could better debrief their sailors about especially good or bad manoeuvres. They could also give insight into whether the boat had been correctly ‘moded’ (the overall setup of the foils and sails for the prevailing wind and wave conditions), or even give side-by-side comparisons of how their team’s setup differed from a better performing rival.

One eye on the future

The second reason for SailGP adopting the shared data model is tied to Coutts and Ellison’s long term plans to grow the fleet size well beyond – perhaps even double – the first season’s six-boat lineup. Supplying all the teams with universal data access made it feasible for new syndicates to join the circuit with at least a reasonable expectation of becoming competitive in a realistically short timeframe.

The Denmark SailGP Team presented by ROCKWOOL is the perfect case study for this scenario. 

After joining SailGP at the beginning of the league’s second season the crew, led by Danish driver Nicolai Sehested, knew they had a huge skills gap to bridge in order to match the F50 experience gained by the six teams from the inaugural season. 

As well as careful analysis of their own data, having free reign to plunder the treasure chest of performance information that those teams had accumulated during the first season catapulted the Danish crew up the learning curve faster than they could ever achieve by trial and error.

Denmark SailGP Team - when it all started in 2020

“When we joined SailGP as a new team we had a lot of catching up to do,” Sehested recalls. “A big part of the way we achieved that was through analysing the data from the other teams. Everything all the crews are doing out on the water is recorded and saved, and after sailing we would look at all the data points from the day and try to learn from what we saw.”

At the core of SailGP’s performance data sharing platform is the computing power, data storage, networking, and security capability of Oracle Cloud Infrastructure (OCI), which also incorporates advanced technologies like artificial intelligence and machine learning. 

A state-of-the-art racing machine

Using OCI in conjunction with an Oracle Fast Connect direct high-speed, low-latency, data connection to an Oracle Autonomous Data Warehouse, the SailGP F50 foiling catamarans – each of which is bristling with around 800 onboard sensors that generate more than 240,000 data points per second – effectively become virtual nodes on the Internet of Things (IoT).

Every aspect of the F50s’ performance is captured live, including: 

  • Boat speed
  • Heading (direction), 
  • VMG (velocity made good) – the speed the boat is travelling towards the next turning mark on the racecourse, even when it is zigzagging upwind or downwind  
  • Wind data – the strength and direction of the true (actual) wind, as well as that of the apparent wind (the combination of true wind and the wind generated by the boat’s forward motion)  
  • Wingsail twist and camber – the 3D profile of the wingsail
  • Hydrofoil rake angle – the forward and aft alignment of the foil daggerboards
  • Hydrofoil wing flap angle – the angle of the hinged flaps on the aft edge of the hydrofoil wings
  • Ride height – the height a boat’s hulls are above the surface of the water
  • Pitch angle – the boat’s horizontal angle fore and aft
  • Roll angle – the boat’s horizontal angle side to side
  • Percentage flight time – how much of the race a boat stays airborne 

Gathering this myriad of data – and much more besides –  and making it available in real time requires sophisticated resources.   

“The data we are collecting off the yachts is what we refer to as time series data. We are quantifying things like hydraulic pressure, in the lengths of actuators in the system – and that is all matched up against a timestamp,” explains Edward Hawthorne – one of SailGP’s control systems engineers.

The SailGP organisation utilises this vast stream of real time data a variety of ways:

  • As a feed for the SailGP mobile app, where SailGP fans can access live performance data from all the boats while they are racing
  • To drive the Liveline augmented reality graphics package used to give TV and online spectators a better understanding and appreciation of what is happening out on the racecourse
  • To enable the remotely-located team of expert yacht racing umpires to monitor the racing in real time and, when necessary, to replay and adjudicate on rule infringements during racing – such as early starters, boundary penalties or close quarters boat-on-boat incidents
  • To give the team coaches on the water a dashboard of performance information of the entire fleet, enabling them to feed critical information directly to the crew while they are racing

Although considered controversial at first, it is hard now to view SailGP’s performance data sharing policy as anything but a raging success. 

On-board camera of the Flight Controller of the Denmark SailGP Team Rasmus Køstner

Its key benefits are clear to see. By enabling easier onboarding of new teams the league’s ambitious expansion plans become more feasible, and making the complex sport of yacht racing easier to understand to mainstream sports fans is enabling SailGP to build a highly engaged global audience. 

But, perhaps the biggest benefit of allowing universal free data access is the process of continuous improvement it drives throughout the fleet. In contrast to the first two seasons, when a handful of experienced teams dominated the podium, SailGP’s third season has been the most closely fought yet. 

Although there may still be favourites, all of the now-nine teams have proved themselves capable of winning races. And, with any new advantageous boat handling setplay or superior setup quickly pounced upon and replicated by the rest of the fleet, the racing looks set to become even closer over coming seasons.

Sailing vocabulary

  • Foils - Foils are 'ski-like structures' mounted below the hull of the catamaran. When moved through the water, they generate lift helping the F50 to fly.
  • Upwind - Sailing the boat into the direction of the wind.
  • Downwind - Sailing the boat with the direction of the wind.
  • True wind - This is the wind that you'll feel when stationary and the wind speed you'll see on your local weather forecast.
  • Apparent wind - Apparent wind is a combination of the induced wind and the true wind. By travelling at an angle to the true wind the combined apparent wind becomes bigger than the true wind, enabling boats to travel faster than the true wind speed.
  • Hulls - the main body of a ship or other vessel, including the bottom, sides, and deck.
  • Hydrofoil - a lifting surface, or foil, that operates in water. They are similar in appearance and purpose to aerofoils used by aeroplanes.
  • Rake - the lean of the mast forward and aft. Changes in rake change the balance of the helm.
  • Daggerboard - a board that slides vertically through the keel of a sailboat to reduce sideways movement (balancing the boat).

Want to read more?