Importance of Height and Weight on 2K Times

[Gammmmo]

Relative speaking our Danish wr lightweight pulling 5.56 is proberly among the very best, if not the best, but against the open record he still needs 20 seconds. Don,t know the exact weight numbers, weigh in for lightweights can vary, sometimes lightweights can row above 75kg, after rehydrating after weight in, but roughly speaking wr lightweight at 75kg, heavyweights 96kg. And that weight was due to height.

Plug those times and weights into our weight adjustment calculator http://www.concept2.com/indoor-rowers/t ... calculator and the little guy comes in within a second. And as soon as you put a hill into the equation (running or cycling) the big guy is out of the running. C2JonW

Certainly, heavyweight rowers are also notorius crap runners. Often ok cyclers on the flat.

As regards cycling on the flat - with time trials there is a tendency for larger riders to be better but it's all about power/cda so smaller guys with great positions can quite easily beat bigger, more powerful guys. It just depends hugely on morphology/fleixbility and to some extent how deep your pockets are because cda can be improved through the use of wind tunnels and velodrome/field testing.

[Tim K.]

Why are we talking about other sports and how a rower would do in that sport when the OP is clearly stating the importance of height and weight on 2K (rowing) times?

As I have stated before, height, in and of itself, all else being equal (mass, fitness, age, body composition, etc) is an advantage.

Why it is an advantage:

We all know it takes more energy to go faster. The relationship is square. Power is a cube but the energy expended over the distance is a square. Double the speed and you quadruple your energy expenditure over the same distance. Once you know that, you know that the most efficient way to cover a distance is at a steady pace. If you travel slower for the first half of the distance and faster for the second half, your average speed will be the same and you will cross the line at the same time but when you cross the line your energy expenditure will be greater than if you had traveled at a steady state.

Just like heart rate, we average speed over time. The speed is not constant however. Graphed in 1/10 second it would be a sign wave.

There are several different possible scenarios, this is just one of them but they all end up in favor of the longer stroke rate (within reason of course):

2 rowers, one 6'er and the other 6'6". Stroke rate is the same, energy put into the boat/fan the same. Problem for the 6' guy is that in order for him to put the same amount of energy into the boat/fan, because his stroke is shorter, and he is doing the same amount of work in a shorter stroke, the boat will have a higher peak velocity for every stroke where the longer stroke will have a lower peak velocity for every stroke. Averaging out the speed of the boat/fan, because of the higher peak speeds and knowing the same amount of energy went into each stroke, we know definitively that the average speed must be lower.

Same amount of energy went in, average speed is lower. hjs, there is your 'free energy"

OTW this would apply to all aspects of motion, the oars traveling through the water, the boat, the rowers mass. I doubt anything but the boats velocity sign wave was included in the erg simulation and I wouldnt bet the house that C2 included that in its equations, but this is not the only aspect in which a taller rower has an advantage, again, all else being equal.

The inherent increase in stroke length due to an increase in height is a distinct efficiency advantage, independent of all other aspects.

[jackarabit]

Cry havoc and let slip the dogs of fizzix! Ain't gonna be pretty.

[Tim K.]

[PaulG]

Cry havoc and let slip the dogs of fizzix! Ain't gonna be pretty.

I think the second law of thermodynamics is due for repeal, personally.

[Ralph Earle]

Meanwhile, back in the real world, here are the results for 6K tests in 2002-2003. The year was included in the regressions to account for training effects over the season.

Women: n = 192; Inches = -6.61 ± 1.17; Pounds = -1.09 ± 0.16; Rsq = 0.53.
Men: n = 311; Inches = -4.49 ± 0.87; Pounds = -0.96 ± 0.11; Rsq = 0.60.

It appears that a longer drive is more important over 6K than over 2K.

[Tim K.]

I have been trying to translate your data into dummy so I can understand it and I think I have it figured out, but would you please explain in lay persons terms what it all means. Thanks!

I wouldn't have thought about it before but it doesn't surprise me at all that as distance increases stroke length becomes an even bigger factor. Efficiency over time/distance pays dividends

[left coaster]

Wow... that's quite the advantage.

The variance accounted for is a bit less than with the 2K though, 40% of the time is still "something else". Still, knowing that 60% of your performance is basically coming down to height and weight is sobering. I might be a bit over 6' but I have arms like an alligator and rather short legs. Need to really get that hip pivot working for me lol!

Tim -- it means that for the entire sample Ralph examined, on average for each inch taller a person is (for men) their 6K time is 4.49 seconds faster and for each pound heavier they are they are .96 seconds faster. For the men these 2 variables account for 60% of the variation in 6k time, meaning that 40% of the time remains unaccounted for. Interestingly, the standard deviation for men is .87 seconds which is quite small. In translation this means that the advantage for height is quite consistent across the sample.

[G-dub]

I knew there was a reason why my 5k blew!

[jackarabit]

I want to hear more on the subject of those p values which I'm told represent the probability that what is going on can make itself look exactly like what one thinks is going on.

http://fivethirtyeight.com/features/not ... -p-values/

[Tim K.]

I have arms like an alligator and rather short legs.

We should start a support group

Tim -- it means that for the entire sample Ralph examined, on average for each inch taller a person is (for men) their 6K time is 4.49 seconds faster and for each pound heavier they are they are .96 seconds faster. For the men these 2 variables account for 60% of the variation in 6k time, meaning that 40% of the time remains unaccounted for. Interestingly, the standard deviation for men is .87 seconds which is quite small. In translation this means that the advantage for height is quite consistent across the sample.

I knew what N= meant

[left coaster]

technically it's a lower case "n" Tim An upper case N denotes the entire sample or population while n denotes a subgroup. I believe 311 is not the entire sample Ralf has access to (chuckle). It's actually true, but I am playin' with you a bit Timbo.

[Tim K.]

Fricken awesome, 0 for 4

[Tim K.]

Ralph, I wanted to thank you for taking the time and applying you skills to provide us with this info. I think information like this is very important for a lot of people as many who come want to "measure up" and when they fall short end up feeling like failures and like there is no point in continuing this wonderful sport. As Carl Watts stated, a much more comprehensive ranking system rather than light weight/heavy weight, would be a huge asset in growing the sport and maintaining interest. (you wont find it in part 1 as while you were working on the stats, it was cleared out by a moderator along with the rest of my interaction with another member)

This is the reason I refused to allow one outspoken member bully me into submission and the reason things got as "ugly" as they did.

[Ralph Earle]

I want to hear more on the subject of those p values which I'm told represent the probability that what is going on can make itself look exactly like what one thinks is going on.

http://fivethirtyeight.com/features/not ... -p-values/

I don't quite understand your statement, but the citation gets it right near the bottom. A p-value is the probability that an assumption about a population is true, given the evidence obtained from a random sample of that population. In our case, the assumption is, "In the population of highly trained and fast ergers, height and weight have zero effect on the total time taken to erg a specific distance (e. g., 2K, 6K)."

But the data from these particular samples of this population show height and weight having correlations with total time that is very unlikely to have occurred by chance, if the assumption is true.

Other random samples of the same size from the same population might well show different results, however. So imagine that one were to take draw very many samples of the same size and compare the results. The p-value is an estimate of the proportion of those many samples which would have results at least as different from zero as the samples actually drawn.

Commonly, a p-value less than 0.05 (one chance in twenty) is deemed "statistically significant at the p<.05 level." This simply means that if I drew, say, 100 samples from a population in which height and weight had no effect on time, I should expect about 5 of them to show a correlation with time at least as different from zero -- positive or negative -- as the one I actually chose.

Now, to preserve my stellar reputation as a statistician, i need to add a couple of points. First, the data from the 2002-2003 team testing results are NOT a "random" sample of the population of team tests. In fact, they are the entire populations of men and women who submitted height, weight and time to US Rowing that season. Technically, then, "statistical significance" does not apply -- the data are what they are for these male and female populations of ergers.

If I wanted to illustrate the role of p-values, I would download all the results from 2002 through 2016 and chose many random samples from the complete data set and compare the results. But, of course, that would be foolish, since, if I have ALL the data, I should just analyze ALL the data! (Since I'm about to be sidelined for most of the summer by major surgery, I just might do that.)

Second, "statistical significance" is a malleable mathematical concept. It can mean one chance in twenty, one in a hundred, one in a thousand. It all depends on how confident a researcher wants to be in the outcome (and whether they can afford to collect enough data to have that confidence). "Statistical significance" is not the same as "meaningful." For example, for the men's 6K, a pound of weight was worth -0.96 seconds. If it had been as small as -0.25 seconds, it would still have been "statistically significant," but trivial -- you'd have to gain four lean pounds to knock just one second off your 6K time.

To sum up, the 2002-2003 data indicate that height and weight matter, just not very much. But on the odd chance that there was something out of the ordinary with that data set, I will analyze the 2008-2009 and 2015-2016 results, so we'll have the beginning, middle and end of the time span.)