How do you know when your improvement curve is slowing, and it's time to either switch your training methods, or switch your focus? After two years of looking for a tool that would create accurate maneuver loss metrics, and save me hours in debrief prep during boat handling training blocks, I gave up and built my own.
Let's rewind to 2017 when the mission to better assess sailor development, and the road to creating Racing Alpha really began...
After the 2016 Olympics, I had seen the top level in Rio, and I knew it was within reach. The key to success: boat handling mastery. With new talent entering the class and veterans slated to return, the process began to build the best tacks and gybes possible. My belief was that by focusing on these two, super simple fundamentals, we could create a modest advantage against even the top teams in the fleet, and that this boat handling edge would allow us to take advantage of smaller shifts than the competition, play more conservative downwind strategies around the lay lines in the corners of the course, and be more aggressive tactically near the marks. Translation: mastery of the fundamentals = high end performance opportunities down the road.
Using the foundation we had laid in 2015 and 2016, I began building a series of Role Sheets for maneuvers, refining the details that went into top tier maneuvers, but pretty quickly realized that to make the final 10% of gains - those last few refinements which would allow us to go from top-10 to best-in-the-world, we needed to be able to test some of these details in a quantitative way.
49erFX Overpowered Tacking Role Sheet
I set out on a mission to find software that would allow me to measure maneuver losses, only to find that nothing really existed that could do what I needed. So I rolled up my sleeves using a tool called GPS Action Replay (GPSAR), and started working on a process for evaluating maneuvers. The end result after a month of daily testing and tweaks was this spreadsheet, which required me to use GPSAR to manually set a wind direction, select the maneuver entry, record a number of data points, select the tack exit, record more data points, select the whole tack, record more data points, and then input those data points into formulas to calculate losses and potential calculation error. Each day, after hitting the dock, I would collect the GPS tracker, download the tracks, put the tracker on the charger, and dig in for 1-2 hours of processing to get data for ~10 tacks per day. Not efficient time, but ultimately a very effective process!
We designed experiments, first trying to figure out what the optimal tacking and gybing techniques looked like, and then focusing on how to recreate those techniques as consistently as possible. We realized that while the individual data points for each maneuver were sometimes unreliable, if we used many data points together over a long period of time, we could see trends. We could compare one set of 10 tacks to another set of 10 to compare technique. We could look at improvement over time to see if we were moving the needle in the right direction. A few times, we hit plateaus, but using data from the rest of the fleet, we knew that we were not quite there yet, so we switched our training methodologies, and kept hammering away.
Finally, we made the breakthrough. We narrowed in on 4 factors that moved the needle, and once we found them, we focused relentlessly on being able to repeat them. By the Europeans in May of 2019, we knew we had a weapon.
Racing Alpha has been the evolution of this process. The manual tagging tool allows for this exact same calculation that I used in my original spreadsheet with only the need to select the entry point and end point of the maneuver. Using this process, you can tag 10 tacks in a minute or two, and compress my 2 hour process into 2 minutes!
The next evolution is in making the automated tagging process more accurate. Every time you add a manual maneuver tag, Racing Alpha gets a little bit better at figuring out how to identify your maneuvers next time. Right now, the automated maneuver identification is good enough to show long term trends in improvement in a meaningful way. Using the Reports tab, Lucy Wilmot and Erika Reineke have been tracking their improvements over time. The negative percentages in the bottom right corner of the graph legends mean that they are moving the needle in the right direction - minimizing their maneuver losses over time.
October training progress
November training progress
As they progress, these improvements will get smaller and smaller until their technique plateaus. At this point, the trick will be using the Expected Tacking Loss metric, to see if we’ve achieved those world class target that we’re after. If we have - hooray! If not, we’ll switch up the training technique, until we figure out how to bust out of the plateau!
There are a lot of factors outside of your control on the race course, but in the words of Hamish Wilcox, coach for Pete Burling and Blair Tuke:
“Boat handling definitely sits in the mental skills area as a ‘tick-off’. You know, it's something that you can just tackle and achieve. And you don't really have to have any special skills to do it. It’s amazing how many sailors aren't that great at boat handling. And yet it is one of the controllables one of the things that you can actually just achieve through determination and perseverance.”