Letting numbers speak
A lot of people will tell you that "numbers never lie." This is a half truth. Numbers may not lie, but the data going into creating them can be examined and broken down in multiple ways so that the results generated run contrary to what the numbers are telling you. Coaches do this all the time. We take data that we gather, and we play with it to come up with a satisfactory result.
Here's a quick example.
Two athletes are vying for the last seat in your top 8.
Athlete A pulls a 6:12 2k score, and weighs 171 pounds.
Athlete B pulls a 6:21 2k score, and weighs 159 pounds.
Let's assume that they are of equal height and build, and have comparable skill sets (technique). The reason we assume this is so it eliminates excess variables.
On first glance, I'm going with Athlete A. They're faster. Or are they?
Most of the coaches I've worked with or spoken with use a formula or two to group athletes based on what the formulas tell them. By breaking down a team into specific groups, you can eliminate some of the guess work that goes into building lineups, and you make seat racing less of a headache than it tends to be. By grouping athletes with similar numbers together, I know that if I'm going to seat race, I know who to pick and who not to pick, rather than running through an entire team. The largest seat racing matrix I've run was 16 athletes, and that took several days (thankfully the weather was solid every day and we were on a buoyed lake course).
So back to the example. I use Power:Weight ratios where I try and find athletes who move more of their weight. But for the sake of this, I'm going to use the C2 calculators, specifically, the weight adjusted calculator. The page explains how this calculator works, but the basic idea is that larger athletes typically put out faster times, however, they are not always the faster athlete.
Athlete A has a weight adjusted score of: 5:36.3
Athlete B has a weight adjusted score of: 5:38.7
So while the raw data suggests a difference of 9 seconds, the weight adjustment shows that there's really only roughly a 2 seconds difference. Now is when you can bring in the other variables such as technical differences, overall fitness, etc.
This is a very rudimentary start, but it is sufficient for this example. On paper and even by weight adjustment, you might take Athlete A. Maybe Athlete A is more fit, and matches up better with the rest of the boat. Or maybe you decide that the 2 seconds lost in terms of time are more than made up for because Athlete B has more consistent technique (very important in the last 500 when the boat is tired and sprinting) and performs better under racing pressures, so you figure the 2 seconds difference will be made up through a better overall performance.
That's up to you as the coach. I can tell you that in my experience, the fastest times typically don't yield the fastest boats. My lineups tend to result in mixed speeds which may hurt the top end speed of my A boat, but really boosts my B and C boats.
Anyhow, just some food for thought when looking at data. It's fun, but I think while numbers may not lie, you should take them with a grain of salt until otherwise proven.
Addendum
One thing to keep in mind when running the numbers is to not get too hung up on the numbers. In my limited experience, getting stuck on numbers is a good way to drive yourself crazy. Every year in the fall when I talk with my team, I tell them how I make decisions, and how I build boats. I like to use the numbers and the data available, but there are times when I make what I term, "Coaches prerogative." It's when I decide to make a decision based on a gut feeling, even if the data supports a different conclusion. I will test the theory in practice and in races, but there are times that I believe a coach has to take a swing because their experience is telling them something that the numbers can't or won't.
One other thing to keep in mind when running the numbers is that this is a sport. There's calculations, and formulas, but there's also chance involved. Sometimes things happen that you simply could not have predicted, or more simply, the best team/boat doesn't always win. They may win most of the time, but there are moments when everything goes out the window and the results hint at something else going on that the numbers can't measure or calculate.
I've been running through previous seasons, looking at the test results and lineups and putting the numbers through the formulas I have now that I didn't know about then. And I'm discovering two things. The first is that the boats I've built, by and large, are the best boats that I could have built. And the second is that the numbers I've generated indicate a level of success far below what was actually achieved by the boats in question.
Don't get hung up on the numbers. Like I said, in my experience, you should take them with a grain of salt.
Here's a quick example.
Two athletes are vying for the last seat in your top 8.
Athlete A pulls a 6:12 2k score, and weighs 171 pounds.
Athlete B pulls a 6:21 2k score, and weighs 159 pounds.
Let's assume that they are of equal height and build, and have comparable skill sets (technique). The reason we assume this is so it eliminates excess variables.
On first glance, I'm going with Athlete A. They're faster. Or are they?
Most of the coaches I've worked with or spoken with use a formula or two to group athletes based on what the formulas tell them. By breaking down a team into specific groups, you can eliminate some of the guess work that goes into building lineups, and you make seat racing less of a headache than it tends to be. By grouping athletes with similar numbers together, I know that if I'm going to seat race, I know who to pick and who not to pick, rather than running through an entire team. The largest seat racing matrix I've run was 16 athletes, and that took several days (thankfully the weather was solid every day and we were on a buoyed lake course).
So back to the example. I use Power:Weight ratios where I try and find athletes who move more of their weight. But for the sake of this, I'm going to use the C2 calculators, specifically, the weight adjusted calculator. The page explains how this calculator works, but the basic idea is that larger athletes typically put out faster times, however, they are not always the faster athlete.
Athlete A has a weight adjusted score of: 5:36.3
Athlete B has a weight adjusted score of: 5:38.7
So while the raw data suggests a difference of 9 seconds, the weight adjustment shows that there's really only roughly a 2 seconds difference. Now is when you can bring in the other variables such as technical differences, overall fitness, etc.
This is a very rudimentary start, but it is sufficient for this example. On paper and even by weight adjustment, you might take Athlete A. Maybe Athlete A is more fit, and matches up better with the rest of the boat. Or maybe you decide that the 2 seconds lost in terms of time are more than made up for because Athlete B has more consistent technique (very important in the last 500 when the boat is tired and sprinting) and performs better under racing pressures, so you figure the 2 seconds difference will be made up through a better overall performance.
That's up to you as the coach. I can tell you that in my experience, the fastest times typically don't yield the fastest boats. My lineups tend to result in mixed speeds which may hurt the top end speed of my A boat, but really boosts my B and C boats.
Anyhow, just some food for thought when looking at data. It's fun, but I think while numbers may not lie, you should take them with a grain of salt until otherwise proven.
Addendum
One thing to keep in mind when running the numbers is to not get too hung up on the numbers. In my limited experience, getting stuck on numbers is a good way to drive yourself crazy. Every year in the fall when I talk with my team, I tell them how I make decisions, and how I build boats. I like to use the numbers and the data available, but there are times when I make what I term, "Coaches prerogative." It's when I decide to make a decision based on a gut feeling, even if the data supports a different conclusion. I will test the theory in practice and in races, but there are times that I believe a coach has to take a swing because their experience is telling them something that the numbers can't or won't.
One other thing to keep in mind when running the numbers is that this is a sport. There's calculations, and formulas, but there's also chance involved. Sometimes things happen that you simply could not have predicted, or more simply, the best team/boat doesn't always win. They may win most of the time, but there are moments when everything goes out the window and the results hint at something else going on that the numbers can't measure or calculate.
I've been running through previous seasons, looking at the test results and lineups and putting the numbers through the formulas I have now that I didn't know about then. And I'm discovering two things. The first is that the boats I've built, by and large, are the best boats that I could have built. And the second is that the numbers I've generated indicate a level of success far below what was actually achieved by the boats in question.
Don't get hung up on the numbers. Like I said, in my experience, you should take them with a grain of salt.
Comments