In threads where training philosophy comes up, discussions of TRIMPS and TSS and other training models occasionally intrude. These models are not very well known, and even more poorly understood, so probably SolarEnergy, qbrain and I are just talking to each other and killing threads in those conversations. In any case, I figured I would present a brief overview of what it is that we're talking about when this terminology starts showing up.
Best case, this will introduce these models to the subset of swimmers (or coaches) who would be interested enough to use them, but didn't previously know enough to do so.
Plus, even if you're not the type to be interested in quantifying your training, it can be useful to think about workouts in this general framework.
And, at the very least, this might serve as a place to discuss some of the details without worrying about driving those other threads too far off-topic.
Good God, I have enough trouble keeping track of how many minutes/yards I'm swimming every day...I couldn't imagine having to keep track of all this, too. Holy poop! ;)
The normal assumption is that these time constants are fairly transferable. You can (some have) fit the time constants, but it requires a lot of data and work.
It's fairly reasonable to think that they should be constant. They have a fairly biological interpretation, and are related to the rates at which your body can build new mitochondria, or hemoglobin, etc. It seems reasonable to think that these biological rates would be pretty similar across individuals.
There is a common belief among swimmers and swim coaches that more muscular people need longer tapers. For a similar reason, women are often supposed to need a shorter taper than men.
The belief might not be correct, of course, but if it is then it implies that the time constants are related to muscle mass.
Q. Would the energy expenditure for swimming a 400 back be the same as for swimming a 400 fly?
The energy that was expended going forward would be the same. Waste would be higher with the fly, and there is no credit given for waste.
An example of waste is pushing water down toward the bottom of the pool.
I don't believe the difference between back and fly is linear if you were to try to model the total energy.
I don't believe the difference between back and fly is linear if you were to try to model the total energy.
I have been keeping up with Q points for a week or so and the big problem for me comes with fly. The amount of energy needed for me to do more fly seems to increase geometrically. In the example that Chris gave, I would be very hard pressed to do a 400 fly, but I could do a 400 back. To me, the problem comes from how you would increase the multiplier based on fly distance. It is much easier for me to do 10 X 50 fly than 5 X 100 fly. Also, fly seems to have a cumulative effect on me during a workout. For me, energy would increase based on the distance of the repeats and on the total distance of fly in a workout.
Q-points are not all that transferable between people of differing abilities. 1000 points for me will probably not represent the same amount of training as 1000 points for someone else. That might be okay if the time constants associated with the impulse response are a little more uniform. (They probably are, but I'm sure not everyone has the same time constants.)
The normal assumption is that these time constants are fairly transferable. You can (some have) fit the time constants, but it requires a lot of data and work.
It's fairly reasonable to think that they should be constant. They have a fairly biological interpretation, and are related to the rates at which your body can build new mitochondria, or hemoglobin, etc. It seems reasonable to think that these biological rates would be pretty similar across individuals.
This is why qbrain's discarding of the swimmer-specific constants makes so much sense, since they're all just linear coefficients. As long as I don't care about comparing points with you, i can scale my points by whatever factor I like.
Edit: for the same reason, it's not worth fussing about whether the 100 in qbrain's formula is yards or meters. Since I swim SCY, I have been using 100 yards rather than 91.44 meters to do my calculations to make life easier. That means my "yards-points" are not the same as qbrain's "meter-points", but we already knew that we can't compare points across different swimmers. The only time the units matter is if I want to compare points earned in an SCY pool to points earned (by me) in a SCM or LCM pool.
Since my 400 fly time would undoubtedly be slower than 400 back, there would be more "energy points" for the 400 back.
I recalculated the ratios using the USAS time standards, but fly is faster than back, which isn't helping your case.
Swimming efficiency is plainly not within the scope of the system; it is aimed at tracking adaptations like lactate tolerance.
Are you sure? Both increased efficiency and increased lactate tolerance will result in more points if either adaption is used to swim faster in future workouts.
-- Q-points are not all that transferable between people of differing abilities.
True.
-- Q-points are, obviously, for swimming only.
True. I wouldn't have bothered if I could strap a data acquisition computer to me like cyclists and runners can do. At that point, the data is available for better analysis. The number of times I can keep in my head during practice to later put in the computer is limited.
Personally, I get a pretty substantial taper effect when I curtail cross-training activities, particularly weights.
The biggest taper effect from weights comes from decreased tissue damage I would guess, not decreased energy usage. I do not think there is a relationship energy usage and muscle damage for weight lifting.
A given energy expenditure on the bike will not yield the same effect on swimming performance as that energy expended in the water.
I think the tissue damage from cycling is minimal and recovery is based on time to restore glycogen stores. Swimming causes minimal tissue damage, but it builds incrementally because rest is not taken. Lifting causes a lot of damage, but rest is taken. This is based on my limited experience with heavy stationary bike used at the beginning of the year. Real cyclists who have to deal with things like hills and wind might experience a much higher level of tissue damage.
My point being, I think tissue damage and recovery from that is not easily modeled and is critical to taper.
I was thinking about getting multipliers the same way, qbrain. I think that makes perfect sense. I'm presumably expending about the same amount of total effort/energy in an all-out 200 of any stroke; BR is slower just because it's a less efficient stroke.
I might prefer to get the mulitpliers from my own PRs, though, because these are swimmer-dependent constants I'm not willing to sweep under the rug. I'm a much better breaststroker than freestyler, relatively speaking. Which means my breaststroke is less imperfect -- compared to the WR ideal -- than my freestyle is. My own personal multiplier for BR would be 1.35.
I'm a pretty lousy butterflier (especially over 200) so I get a multiplier of 1.6 there.
I kind of like the self-correcting nature of this approach. I get more points for swimming the strokes at which I suck the worst. Which is where I have the most room for improvement.
This turns qbrain's formula into
points = d^3 t*^2 / (t^2 d*^3)
where
d = distance swum
t = time taken
t* = PR time
d* = distance of PR
This completely renorms the point scale. (1 point is now equivalent in effort to swimming a PR for a 200 m race.) You can multiply by 100 or 1000 or whatever you like to make the point scale sufficiently impressive looking.
My thoughts on kicking:
I don't think kick sets deserve a multiplier. Sure, you don't get as many points as you feel you've earned based on how much kicking hurts. As qbrain points out, that's because most of your energy -- let's say 50% -- is wasted.
So when you train yourself to kick more powerfully, only 50% of that additional power is directed towards forwards propulsion.
The point system "knows" this, because all it sees is your speed. The points automatically rewards you only for doing things that make you go faster. I think that's more of a feature than a bug.
where did some of the things like intensity cubed is proportional to power come from?
As qbrain pointed out, this is because power rises as the cube of velocity, when drag is the most important force to overcome.
I bring it up just because I'm wondering if a different exponent might make more sense for some purposes.
qbrain's formula is perfect when you want to measure energy expenditures (as the name "energy points" implies). And qbrain is clear that this is what he is interested in -- he'd rather measure energy than count yards, as a way of encouraging quality over junk yards.
But (a) I'm not sure this is enough incentive for me to avoid junk yards. A 100 at warmup pace is worth about half the points of a race-pace 100 for me. Mentally, I'd much rather swim an easy 200 than an all-out 100.
And (b) for anyone (like me) who wants to use points of some sort to model training with an impulse-response model, I don't think energy output is the best thing to track. Cyclists, in particular, like to track power^4, because this has been found to correlate with lactate concentration in the blood. And lactate seems to be the standard for predicting training effect.
That would suggest that qbrain's energy points could be extended to define "pain points":
p-points = (d / t)^12 * t
Doing some sample calculations suggests that these track a lot more closely with the mental effort required to do the set, and with the perceived training benefit. For example, I would earn about as many "pain points" for doing 3x100 race pace with lots of rest, as I would for doing 10x100 with :10 rest. And to earn the same amount of points at warmup pace, I'd have to do 80x100 easy.
