Training Fundamentals for Competitive Runners
Physical Fitness
Since no runner can maintain maximum fitness all year around, several questions have to be answered.
- - -	How long does it take to bring a runner to top condition? How long can he maintain it? How much below his peak can a runner retreat and still be able to return within a certain period of time?
Assuming that a runner's condition follows a sine curve, i.e. peak followed by trough followed by peak, etc., we need to know:
- - -	The number of times a season the runner can physically (and psychologically) peak. The minimum physical condition which would still allow him to regain his peak within a specified time. The competitive runner's race schedule.
In this example, we shall use an interval training exercise to define fitness. Say that:
- - -	Minimum fitness is 10x400 meters @65 seconds/400 meters with a 60-second rest between intervals. Maximum fitness is the same exercise but at 55 seconds/400 meters. The season lasts 24 weeks and the runner is to peak every 8 weeks.
In the beginning, the runner can do 75 seconds/400 meters. To get him down to 55 seconds/400 meters in 8 weeks, the pace has to be reduced by 2.5 seconds/400 meters each week. For the next 4 weeks, the pace is increased by 2.5 seconds each week, and so forth. Notice that all the graph segments are linear, i.e. the pace changes by the same amount each week. Suppose the coach wants most of the change to occur at the beginning or at the end of a segment, i.e. non-linear changes. Figure 2 shows three general patterns, two of which are non-linear.
In decelerating improvement, most of the change in pace takes place in the beginning of the period, as opposed to accelerating improvement where most of it is towards the end. Applying accelerating improvement to the first 8 weeks of Figure 1 gives Figure 3.
The smooth patterns of Figure 2 are unlikely in the real world; more likely, improvement will follow a step-like pattern. Figure 4 shows this pattern for the first 4 weeks of Figure 1.
A word of caution. Although the KIP RunPacer is the most effective "carrot" for maximizing workouts, it shouldn't be abused. Instead, try varying exercises so that one day is easy and one day hard. For example, a hard day's workout might include 10x400 intervals at a certain pace and rest. The easy day has 10x300 intervals at the same pace and rest. Or, a runner may do 5 kilometers at some pace on a hard day and 3 kilometers at the same pace on an easy day. Also, resist the temptation to increase the pace "just a little more" than what the runner really should be doing. If he doesn't succeed and becomes frustrated, he may balk at using the RunPacer and miss out on some great training.

Pace

Outside of negative splits - running slower in the beginning and faster at the end - constant pace is probably the most efficient way to run. So, we'll concentrate on the art of pacing in general and how to teach specific paces.

The best way to learn pacing is by doing all workouts at constant speeds. Over time, the runner will relate running at a steady speed to how his body feels and will know when to push as his body gets tired. Should he want to test his pacing skills, he can always choose a KIP RunPacer exercise where he runs with the beeps, without them and again with them.

Competitive runners also have to learn specific paces. Suppose the goal is 1500 meters in four minutes at a constant pace. The runner starts with 800 meters at the target pace. The next time 900 meters, eventually up to 1500 meters. Along the way, he can test himself by running segments without the beeps.

Interval training is another effective way to teach specific paces. In the beginning, the runner may do, say, 5x100 at the target pace with the same rest at the end of each segment. Next time, he does 6x100 at the same pace and rest, and so forth. For variation, the rests can be reduced while the distances are being increased.

Although true negative splits are difficult to execute without a KIP RunPacer, they are worthwhile trying, if only to discover one's potential. And, a runner who does master the art of negative split running will find himself at a strong advantage over his rivals. At the very least, negative split exercises will help make workouts a lot more interesting.

Optimal Race Pattern (ORP)

Ideally, a runner should arrive at the end of a race with all of his "fuel" spent. Also, he should arrive first! Suppose there are two runners with identical times for a certain distance. The first is a strong sprinter while the second maintains a constant pace from start to finish. In a race, the sprinter will try to "convince" the pack to run slower than his constant-pace rival would like. Likewise, the constant-pacer tries to convince the pack to maintain a faster pace. Since it is more natural for the pack to run slower, a "persuasive" sprinter should win more often. In short, for the sprinter, the pacer, or anyone else, not running his own optimal race pattern is a losing proposition.

Developing an ORP is relatively straightforward, at least in theory. For a particular distance the trial time is set at, say, 10% slower than the runner's best time. Various race patterns, such as constant pace, sprint-pace-sprint, or negative splits are all run in the same training session. Data, like pulse, respiration, and lactic acid level, as well as the runner's subjective reactions, are collected. Exogenous factors like temperature, humidity, and altitude should also be taken into account.

If the success that swimmers have had with negative splits is any indication of their efficacy, then serious consideration should be given to it by runners. Figure 5 show some typical negative split patterns.

The two factors determining the characteristics of a negative split exercise are the total time for the distance and the skew. The skew is the deviation from the average - the "tilt" of the line. Suppose the average pace for 1500 meters is 16 seconds/100 meters (total time is 4:00). If the skew is 10% then the first 100 meters would be run in 17.6 seconds and the last 100 meters in 14.4 seconds. Holding the total time constant, the coach can experiment with various skews to determine the best one for a runner. Also possible are modified negative splits where the beginning and end skews are different or where the pace changes instead of at each 100 meters, at each 200, 300, or 400. Figure 6 shows a modified negative split pattern.

An important part of the ORP is determining where a runner should begin his final kick. To do this, we have to calculate what we call his "Relative Sprint Advantage."

Let's assume that one super runner holds all the world records, from 100 to 5000 meters. For example, if we divide his 200-meter sprint time by his 200-meter average time for 2000 meters we get an Ultimate Sprint Factor of 0.69. (A sprint factor of 1.0 would have meant that his 200-meter sprint time was the same as his 200-meter average and, therefore, no sprint advantage at all.) The following table illustrates the Ultimate Sprint Factor at 100, 200 and 400 meters for various distances (it's about the same for 100 and 200 meters). Interestingly, the factors for men and women are virtually identical.

TABLE 1. ULTIMATE SPRINT FACTOR
Distance (meters)	100/200	400
800	0.78	0.85
1500	0.70	0.78
2000	0.69	0.76
3000	0.67	0.73
5000	0.63	0.70

Let's look at a runner whose best time for 1500 meters is 4:00. His best sprint times for 100, 200, and 400 meters are 14, 27, and 58 seconds, respectively. His average 1500-meter times for 100, 200, and 400 meters are 16, 32, and 64 seconds, respectively.

From Table 1 above, the Ultimate Sprint Factor for 1500 meters at 100/200 meters is 0.70. Were this runner a slower version of our super runner, his Ultimate Sprint Times would be: 11.2 seconds @100 meters (0.7x16); 22.4 seconds @200 meters (0.7x32). For 400 meters the Factor is 0.78 which gives an Ultimate Sprint Time of 49.9 seconds. Table 2 displays all the information needed to determine where he should begin his kick.

TABLE 2. RELATIVE SPRINT ADVANTAGE
Sprint Distance (meters)	100	200	300
Ultimate Sprint Factor (USF)	0.70	0.70	0.78
Average 1500-meter Pace (AVE)	16	32	64
Ultimate Sprint Time  (UST) = (AVE) x (USF)	11.2	22.4	49.9
Best Personal Time (BPT)	14	27	58
Sprint Advantage Factor (BPT-UST) / (AVE-UST)	0.58	0.48	0.57

The lower the Sprint Advantage Factor, the greater the advantage. A value of zero indicates that the Best Personal Time equals the Ultimate Sprint Time. In other words, our runner is just like the super runner...almost.

No sprint advantage is indicated by a Sprint Advantage Factor of "1," where the runner's best time for a sprint distance is the same as his average pace for 1500 meters.

For our runner, the Sprint Advantage Factor of 0.48 indicates that he should start sprinting 200 meters before the end of his 1500-meter race.

Suppose that a runner's theoretical ORP for 1500-meter is constant pace with a sprint 200 meters from the end. However, if he has difficulty holding a steady pace, his practical ORP might be to stay in the middle of the pack and begin his sprint according to the times of the last 800 and 400 meters. For example, if the times have been relatively slow he may begin his sprint, say, 300 meters before the end.

Race Simulation

Think of the KIP RunPacer as the perfect competitor for building up race experience. Not only can you re-run an actual race from the past, but you can also compete against invented runners. A word of caution, though. Invented runners have to be human. They can't stick in world record sprints in the middle of a 1500-meter trial nor can they vary their paces in an unrealistic manner.

Splits of actual races are easily obtainable, requiring only a videotape of the race and a stopwatch. For middle-distance races the 100-meter splits can be taken while for longer distances the 200 or 400-meter splits may be more relevant.

An advantage in inventing races is that the coach knows the strengths and weaknesses of his runner and can custom design a more challenging simulation. For example, a runner with a strong finish might be racing against an even-paced simulation where splits are somewhat faster than called for by the runner's ORP. Should the runner stick with his ORP he would find himself falling further behind as the trial progresses? Only when he starts his final sprint will he catch up with and “defeat” the KIP RunPacer.

So, how does a runner prepare for a major competition? The first phase is to bring his Optimal Race Pattern (ORP) time down to the target time. If he does it linearly (Figure 2), he just divides the difference between his ORP time and target time by the number of weeks to the competition. Then, he reduces total time each week by that number of seconds.

The second phase involves competing with the beeps while trying to maintain the ORP. In the first trial, the runner knows the total time of the exercise and that it is run at a constant pace. In another trial, the runner is told the total time of the exercise, but not the splits. In still another trial, the runner is told neither the total time nor the race pattern of the beeps. The object in all these trials is to see how closely the runner can follow his ORP in simulated competition.

Sometimes, a runner may have to run sub-optimal race patterns according to the progress of the race. Suppose his ORP is constant pace until the last 400 meters and then a sprint. However, during the race the runner stays with the pack and the actual pace is slower than what he would have liked. The question is where to begin his final kick. Having done many race simulations with the KIP RunPacer, the runner has already encountered a similar situation and, hence, knows exactly where to begin sprinting.

To add realism to a race simulation, some of the runner's team-mates could run exactly with the beeps. If none of the team-mates can handle the entire distance, several could alternate running different segments of the race. This inclusion of other runners is especially useful when the runner's ORP dictates that sometimes he be ahead of the beeps. Although he can't see where the beeps occur, he could check himself by glancing back at his team-mate.

Physiological Testing and Exercise Design

First, we assume that pulse rate is a reliable indicator of work output. Then, we use an exercise bicycle (or similar machine) to obtain a relationship between pulse and work. Finally, we make inferences between pulse and pace to help design exercises.

Beside the direct relationship between pulse and pace, there are two other important relationships. The first has to do with the time it takes to reach a steady-state pulse rate. For example, the extra energy expended in reaching the desired pulse rate more quickly might be put to better use at the end of a race. The second relationship is between pace and recovery time, in particular with respect to interval training. Here, the trade-off is between doing intervals at faster paces and longer rests or slower paces and shorter rests.

The runner is tested on the exercise bicycle at increasing loads and steady-state pulse rates are recorded. As the load increases, the pulse will increase steadily to a point from where it then rises sharply. Figure 7 indicates the situation.

If we assume linearity in the above graph and linearity in the relationship between pulse and running pace, then we only have to measure the steady-state pulse rates at two different paces on the track (using the KIP RunPacer, of course). Since we already know the maximum pulse breakpoint from Figure 7 we can derive the graph in Figure 8.

Suppose that the runner wants to train at 80% of his maximum pulse rate of 140 beats per minute (about 112 beats per minute). The corresponding pace is about 20 seconds/100 meters which is then programmed on the KIP RunPacer.

As the runner becomes more fit, the graph in Figure 8 shifts to the left and the maximum pulse breakpoint increases. Consequently, at the same pace, the steady-state pulse rate will be less. It is recommended that the graphs in Figures 7 and 8 be up-dated periodically.

The second relationship necessary for developing exercises is that between pace and the time it takes to arrive at the desired steady-state pulse. Suppose the runner is to do intervals where the pulse is to be 80% of the maximum. If at that pace it takes 30 seconds to arrive at steady state and if he is doing repeat 100's at 20 seconds/100 meters, he will never get to the desired pulse rate.

Suppose we want to know how long it takes to reach steady-state pulses for 60, 70, and 80% of the maximum workload. From the first experiment on the exercise bicycle, Figure 7, we already derived the steady-states pulse rates corresponding to these workloads. To find the time to reach, say, 60%, set the corresponding load on the bicycle and take a pulse reading every 10 seconds or so until steady-state is reached. The same is done for 70% and 80%.

We also know that steady-state pulse rates can be reached more quickly by starting at higher loads. So the bicycle is set for a load corresponding to, say, 90% of the maximum, the pulse is taken every 10 seconds and the time is noted when it reaches 60%, 70%, and 80% of the maximum. Assuming linearity, we obtain the graphs shown in Figure 9.

In deciding how fast the runner is to reach his steady-state pulse, the benefits of constant pace have to be measured against the extra lactic acid the runner has to carry because of his greater efforts in the beginning.

The third relationship between pace (pulse) and recovery time is especially important for interval training where the rest should be long enough for the runner to recover to a prescribed pulse rate. Using the exercise bicycle set the workload for, say, 80% of the maximum pulse rate. When the runner reaches steady state he stops and his pulse is taken every 10 seconds. The times for recovering to 60%, 50%, and 40% are noted. The same experiment is repeated at workloads corresponding to 70% and 60% of the maximum pulse rate and the recovery times are noted. Assuming linearity, we obtain Figure 10.

For short-distance interval training, the runner's pace does not correspond to a steady-state pulse rate and data has to be gathered in the field. Suppose we want recovery times for an interval distance of 100 meters. The runner sprints 100 meters at a certain pace and records his pulse at the start of the rest period and, say, every 10 seconds. He repeats at another pace and if we assume linearity the graphs of Figure 11 are obtained. For the sake of completeness, each runner who does interval training should have graphs for 200, 300, and 400 meters.

At this point the coach should have most of the physiological data for designing practical and efficient exercises. It goes without saying that he should also try to keep exercises and workouts interesting and challenging so that runners won't lose their motivation to do their best.

Conclusion

With all of the KIP RunPacer's obvious advantages as the most sophisticated and productive training system available for runners, its greatest attraction is that it makes serious training fun! When the runner also participates in setting his own challenges, the pleasure of using KIP is multiplied. What could be more rewarding than setting worthwhile goals and achieving them through intelligence and hard work?