How can you predict your performance in a forthcoming marathon? And how can you measure improvements in your fitness if you are not racing regularly? The answer is to work out your 'critical velocity'! What's that? Another sophisticated performance variable to deal with - just when you were finally getting comfortable with the intricacies of vVO2max, tlimvVO2max, and lactate-threshold running speed?
So what is this critical velocity - and how do you determine your own? All you need to do is perform four treadmill runs at different, very high-quality velocities. Make sure that each of the speeds you choose produces complete fatigue (inability to maintain the speed) within 10 minutes. If your chosen velocity allows you to cruise along for more than 10 minutes before prostration, go for a faster one.
Complete the four-speed critical velocity test during a single workout. Warm up properly first, and then select the fastest of your four speeds - one you are confident you won't be able to sustain for more than 90-120 seconds. Once the treadmill is running at this speed, hop aboard, start your stopwatch, remain as relaxed as possible and try to hang in there as long as you can. When you're forced to grip the handrails or hop off the treadmill altogether, stop your watch and note how far you ran at the chosen pace.
Time and distance limits
Jog or walk easily for 12 minutes to recover, and then select another velocity which is just a little slower than the first one. Hopefully, this will produce exhaustion in 3-4 minutes. Time how long you manage to maintain this speed and note the distance covered. Allow 15 minutes for recovery, then repeat the exercise twice more with successively slower speeds, making sure neither allows you to keep going for more than 10 minutes. Once you've completed the fourth surge, your critical velocity test is finished. Now comes the slightly trickier stage of working out its implications.
For each of your four chosen velocities, the amount of time you were able to run before you became too exhausted to continue is called the time limit, or TL. For each velocity and corresponding TL, the distance you ran is known as the distance limit, or DL. Pinpoint each TL with its corresponding DL on a graph which plots distance as a function of time (see example below). Connect your four (TL, DL) dots; when you do so, you should have a straight line extending all the way back to the vertical y axis of your graph. This line will, of course, have a slope, which represents the change in distance per unit change in time. By definition, this slope is your critical velocity.
As you look at the graph, you should be able to see why critical velocity should have decent predictive power for runners and other athletes. (Cyclists, for example, can determine what is known as their 'critical power' on a cycle ergometer, while rowers can do the same on rowing machines). When the critical velocities of various runners are compared, those runners with 'flatter' slopes and thus lower critical velocities cover less ground during each additional second or minute of running than those with steeper slopes. Even if such runners have similar y-intercepts when their DLs are plotted as functions of TL, the competitors with higher slopes (greater critical velocities) will pull steadily away from than those with more gradual slopes as race distance increases.
However, there are limits to the predictive power of critical velocity. If runners with identical slopes are compared, those whose slopes appear 'higher' on the graph - ie those covering greater distances for each point in time - will obviously be faster than those whose slopes are lower, even though their critical velocities are no different. To grasp this, you have to understand that the general equation for our relationship between distance limit and time limit is: DL = a + b(TL); 'b', of course, is the slope and thus the critical velocity, while 'a' is the y-intercept of the graphical relationship between distance and time. Interestingly enough, 'a' is also considered to represent a runner's anaerobic capacity.
Interpreting the y-intercept
Think of it this way: if you asked a runner to run at a maximal speed which he/she could sustain for no more than 10 seconds or so, the time limit (TL) would obviously be very short, so short that the point (TL, DL) would lie very close to the y-intercept. All of the energy needed for this 10-second blast would be produced 'anaerobically'. Thus, the value of DL (its height on the y axis) would be determined fundamentally by the runner's anaerobic or 'oxygen-independent' power. Those with high DLs would have large anaerobic capacities, while those with low DLs would have smaller anaerobic reserves. There's another aspect to this, of course. Running, as you know, is also a neural thing, so a higher DL might represent not so much a greater anaerobic energy supply as an ability of the nervous system to generate and co-ordinate power, given a fixed supply of oxygen-independent energy. Thus, a high y-intercept might reflect excellent neuromuscular co-ordination and efficiency rather than a surplus of anaerobic enzymes.
Given that critical velocity is usually determined by treadmill speeds which produce complete fatigue in less than 10 minutes, how can it predict success in a prolonged test of endurance like the marathon? Can it really be better than lactate threshold and VO2max at foretelling marathon outcome?
To answer these questions, researchers from Columbia University in New York City carried out critical velocity research with 12 marathon runners (six men and six women). The mean age of the runners was 29, their average body weight was 140 pounds, and all 12 were due to compete in the 1994 New York City Marathon. The athletes had been involved in running for an average of eight years, and all had completed at least one marathon before the New York event (1).
The subjects' critical velocities were determined as follows: each runner warmed up for five minutes by walking on the treadmill at 1.6 metres per second (16:45 per mile pace), then straddled the treadmill belt while the chosen velocity was set. Once the appropriate velocity was reached, the runners began running while holding the handrails until they had adjusted to the speed. Once they released the rails, timing began and did not stop until the subjects grabbed the handrails again at the point of volitional exhaustion. TL and DL were recorded for each effort.
All 12 individuals finished the marathon on race day, despite fairly oppressive environmental conditions (the temperature averaged 72¡F with 82% humidity). The runners, by no means elite, fared very well under the circumstances, with finishing times ranging from 3:12 to 4:21.
Did their critical velocity as measured by the Columbia researchers predict their finishing times? You bet it did! A simple linear regression analysis revealed that critical velocity correlated more highly with finishing time than either ventilatory threshold or maximum aerobic capacity (VO2max). The actual regression equation was MT (marathon time) = 445.3 - (50.3 X CV), where CV is critical velocity. In other words, if your critical velocity happened to be 5 metres per second, your predicted marathon clocking would be 445.3 minus (50.3 X 5), or 193.8 minutes. Of course, the higher the critical velocity, the faster the marathon time.
By contrast, VO2max and ventilatory threshold did not predict marathon time so well. In fact, when a stepwise multiple regression was conducted, VO2max was left completely out of the predictive equation; only critical velocity and ventilatory threshold 'made the grade', with critical velocity a stronger factor than VT.
Why should the CV test, which uses high speeds producing fatigue in less than 10 minutes, be so good at predicting performance in a much slower event lasting more than three hours? This is simple: note that the y-intercept of CV basically reflects a runner's power - ie max running speed, co-ordination, quickness of force application, and efficiency. All of these attributes are important for the marathon: the faster your max speed, the easier it is to keep up a respectable marathon pace; the better your power and co-ordination, the more explosive your footstrikes and the shorter the time you spend with your feet 'planted' on the ground. Remember that a three-hour marathon is a race of more than 32,000 steps. Reduce the stance phase of the gait cycle by just 20 milliseconds, and you carve 640 seconds - almost 11 minutes - from your finishing time.
CV itself (the slope of the line linking distance with time) represents a number of things. The higher the slope, the greater the fatigue resistance during fast running, and fatigue resistance is itself a function of vVO2max, lactate threshold, running economy, and running-specific strength. Thus, CV brings together the key predictive physiological variables; in effect, CV is a 'barometer' of how well your nervous system, heart, and muscles are working together.
CV and your training
What implications does this have for your training? The take-home lesson from this critical velocity research is that the ability to run extremely fast and display great fatigue resistance over an array of running velocities is a terrific predictor of marathon performance. This means that quality training - the kind of training which optimises maximal running speed as well as efficiency at high velocities - translates into superior performance. This principle holds true even in competitive events in which running velocities are far below maximal and even below lactate threshold - many well-trained runners run a marathon at about 80% of vVO2max and around 94% of lactate-threshold speed. For the marathon, think neural training: lots of high-quality work at around vVO2max which leads to improvements in CV.
The Columbia researchers studied sub-elite athletes: would the CV test also work for elite athletes - and for people who are slower than the Columbia runners? There's little reason to think otherwise. Elite athletes and back-of-the-packers can also be tested for CV, and critical velocity will vary within each group. Those with the greatest y-intercepts and CVs will always be the best runners within their group, whether you are talking about 2:08 or 5:08 marathoners.
Why was VO2max such a relatively poor predictor of marathon performance in the Columbia study? Remember that it is simply the maximal rate of oxygen utilisation and says nothing about the other key predictors of performance, notably vVO2max, lactate threshold, running economy and even CV. One might have a voluminous VO2max and good or bad economy with high or low vVO2max, lactate threshold and CV. VO2max simply can not be relied upon to predict performance, since it can be accompanied by good or bad physiological baggage.
The other variables, however, tend to be more highly interrelated. if you have a great lactate threshold, for example, there's no way you can have poor economy; if you have a high-class vVO2max, you are assured of good economy and lactate threshold; while CV is a function of vVO2max, lactate threshold and economy. The variables other than VO2max simply contain more information and are thus more reliable when it comes to predicting outcomes.
If you have a treadmill or cycle ergometer, I advise you to work out your own CV, as described at the beginning of this article. Just make sure you choose four test intensities which each produce fatigue in less than 10 minutes, and try to 'space out' the paces as much as possible - ie don't make them too similar. For example, you might choose an intensity (running speed or Watt level on the bike) which wipes you out in two minutes, a second which you can tolerate for three, a third which defeats you in four minutes and a fourth which you can manage for six. To monitor your progress, repeat exactly the same test every 4-6 weeks. On top of its usefulness in assessing your training progress, it is a sensational workout in its own right, usually featuring at least 15 minutes of incredibly high-quality running.
If your training is going well, your CV should improve fairly steadily, and so should your y-intercept. As those two factors are upgraded, your maximum speed improves and you display increasing fatigue resistance. As a consequence you will become a faster marathoner. And - although the research remains to be done - you should also run faster at lesser distances.
Owen Anderson