Increasing Use of Data in Competitive Sailing Racing

College Sailing Practice
Crew (Not Sailing)

A lot of people in my life hear that I sail and for some reason they think of crew….. so the above photos break it down for those people…

When I was a kid the only way to test if your boat was setup properly was to find another boat and sail next to them and if they were significantly faster then you, then you probably had to change your rig setup. As I got older I became a coach and we had a few more tools at our disposal; however, the tools were mostly for boat setup and weather updates for lightening (lightening cancels sailboat racing as each boat has a big metal pole pointing to the sky). The boat setup tools really just measured things like tension on a wire, and weren't that helpful once you got on the water. Finally when it came to technology as a coach, the most “techy-savy” coaches had speakers on their boat. The speakers didn’t help the sailors in any way but they provided entertainment when the wind died.

In the last decade the sport of sailing has seen a massive influx of technological advances that has had drastic changes on the space and large part of that is due to big data analytics. In this blog post I am going to highlight some of the changes and the new tools that are supplying massive amounts of data that are being used and changing the sport. These changes vary from real-time decision making during a race all the way to the actual construction of boats.

The AMERICAS CUP…. has been changing the game

If a person has seen one sailing race in their life, it is most likely the America’s cup races. That has been the point in recent years, to get bigger boats going faster to attract new spectators to the sport. This makes a lot of sense as most races are held pretty far off the coast where the wind is and you need a boat to watch the races. To make the boats go faster there were teams of scientists and engineers that came up with the massive AC-72s seen above. However, in order to make these boats go so fast (AC-72s go 44.15 knots ( 50 mph) in 15.8 knots of wind (2.79 times the wind speed), there was a lot of data analysis that needed to occur. These new boats broke so many barriers that so many new factors came into play requiring massive amounts of research and data to configure. At such high speeds, even the materials of the boat need to be altered because the 3 tons of weight combined with the forces propelling that boat even made the foils extremely hot. Water actually boils around the foils above 20 knots. (Foils are like plane wings under the water that look like they are holding the boats up on stilts).

America’s Cup Sensor Data: Team Oracle USA has stepped up their data game dramatically in recent years and have a solid amount of details on their team home-page(Data analytics obviously being provided by Oracle). For the recent America’s cup, the Oracle AC 72 had over 1,000 sensors to collect information on the weather, the boats integrity and the sailors health. Every-time the boat went out for a sail to race or practice the team generated between 250gb–500gb of raw data.

Weather Models: The Oracle team also recently started developing their own models. These models taking into account years of local sensor and cross-validation data to understand what a particular mix of conditions (water and air temperature, barometric pressure, tide, ocean swell, water depth around the course, etc.) might do to wind and current gradients over the course of the day.

SAP Sailing Wind Model Forecast

The Oracle Team started in recent years implementing data science into almost every application of developing, training and sailing these AC-72s.The team used the data for several applications including “Real-time feedback to sailors”,”Course planning and race playbook”, “Yacht-Design” and “Deep analysis and research”

SAP Sailing Analytics Platform

SAP Sailing offers one of the most advanced data analytics platforms currently on the market. For every race on the platform, there is a massive amount of information for each competitor. The metrics include every maneuver (turn basically) executed during a race, mark (buoys the boats sail around) passing time, time, Speed in [kts], Speed out [kts], Speed change [kts], Lowest speed [kts], Max. turning rate [deg/s], Avg. turning rate [deg/s], Maneuver loss [m], Direction change [deg].

Below is an example of a race from the SAP Sailing from a race in San Diego for the Extreme Sailing Series. You can get to these analysis by going to the SAP Sailing link below, clicking on any race, clicking on the race tab, and clicking race details on the right hand side. The tool is very interactive and pretty fun to play with. You can watch the movements of all of the boats on a map with the buoys and course as well as the speed and other metrics of each boat.

SAP Sailing Demo

SAP Data Mining Tool

I was given access to the SAP Data Tool and was able to fumble around with some of the data and was able to graph one race on Plotly. (Still figuring the tool out as it is very involved and my computer keeps crashing trying to run queries)

First I chose the race I wanted (World Cup Series 2018 — Gamagori, Japan — Laser Radial) and selected the two competitors I wanted to analyze (Erika and Paige, the two US Sailing Team Laser Radial Sailors). For this analysis I chose to get all of the data for each competitors tack/gybe details i.e. entry speed, exit speed, degrees turned. This data is below in the text editor Atom.

Next I opened the data in Plotly and chose just to graph and compare the exit speed of each maneuver and wanted to break it down by tacks/gybes per person. In Plotly, this is pretty simple, I just uploaded the data and added two traces using the maneuver exit speed value column and had each trace assigned to either tack details or gybe details. From that (and fumbling around with the settings a bit) I was able to make the below graphs to compare the data.

Each dot point represents one maneuver’s exit speed while the Box Chart next to the dots represent the max, min, median and average speeds. What I found the most interesting is that this is just one data point from one race between two people. From this data you could make several conclusions as a coach and it is just a fraction of the data accumulated for one race and a fraction of a fraction of the data accumulated during the regatta.

Improving Decision Making in Ocean Race Sailing Using Sensor Data- Volvo Ocean Race

2013 Volvo Ocean Race Boat

In 2017, Jos van Hillegersberg, Mark Vroling and Floris Smit from the University of Twente wrote a paper called “Improving Decision Making in Ocean Race Sailing Using Sensor Data.” The paper outlined the implications from the 2017 Volvo Ocean Race. The race is the “world’s pre-eminent round-the-world yacht race” and for the first time all of the boats had the exact same design (called one-design class) which included dozens of sensors that would relay information of each competitor during the race.

  • “Volvo Ocean 65 has over 160 sensors on board which generate data for around 40 variables”
  • “The data that is generated by the Volvo Ocean 65 fall one of 3 categories: wind data, navigation data and boat data”

The data table from the paper is below:

Data-collection system:“For the data storage platform, the university’s MySQL server was used. For creating the visualizations and evaluation dashboard the data analytics tool Tableau was selected based on availability, skills and resources and requirements to quickly build various visualizations on in-memory datasets. Tableau is a software package that lets users create a wealth of visualizations. It is well known to be user friendly and relatively easy to use. This makes Tableau suitable for doing exploratory data visualizations in this project. The data is transmitted from the on-board sensor data collection software to the SRAA in csv format. Using Python, the csv files are converted to several SQL files. The SQL files are imported using MySQL Workbench. Then Tableau is configured to extract the data to store it in a more efficient in-memory format before being ready for visualizing (Figure 4).”

The majority of these cases the teams are competing at the highest levels with advertisers supplying a lot of money that allow the data analytics to be provided. But SAP for example is expanding to more and more teams and allowing access to teams that couldn’t afford to do it on their own. It is a really interesting time to be in the data science world and sailing is a great example of data analytics proving how helpful it can be.

Sources: