Ultramarathon Runners’ Toolkit for Monitoring Training

Jason Koop,
Head Coach of CTS Ultrarunning

In last week’s blog, I alluded to a upcoming article on running power meters. In the course of writing that article, I felt athletes would be better served with a bit of a setup and primer on how to track training so that a specific article on running power meters would have some context. With that as a backdrop, here are some overarching tools and themes I use to track training with my athletes.

Long before Strava existed, I killed a lot of trees coaching an athlete. On any given workday, I received numerous e-mail attachments of training files. I printed out every single one of them, and then I’d pore over each file, examining the critical data. Highlights of yellow, orange, and pink would adorn the sheets, denoting intervals, critical heart rate values, and anything else I found significant. I’d fill the margins with notes from conversations and debriefs on each workout from the athlete. Finally, I’d place the sheets of paper in individually labeled folders, one for each athlete. Thankfully over time, training-related technology and software has improved. Athletes have a number of different platforms that can assist in making sense out of the information collected from GPS watches and other connected devices such as a running power meter or heart rate monitor. Each system had its own spin. Some tools have been very good at taking individual files and breaking them down. Others have been better at aggregating many files to identify trends.

In ultrarunning, particularly trail ultrarunning, using these tools is tricky. Simple things like tracking training load become blurry for runners training on mixed surfaces. Markers of improvement are also difficult to come by, particularly in mountainous areas where the properties of the trail surface can change from one run to the next or even during a single run.

However, I still have cobbled together a system of tracking training for trail running. It’s imperfect, and likely always will be. But, it does provide insight into what is going on, how much an athlete in improving, and provides clues as to when to rest and when to pour on the workload.

Logging workouts: Strava, TrainingPeaks and FinalSurge

The first step in evaluating your training is to determine where your training log will exist. From low tech solutions like a paper and pen or free software like Microsoft Excel to sophisticated platforms like the recently released WKO5, I have seen them all. While each has strengths and weaknesses, my preference has always been to use third-party platforms like TrainingPeaks, FinalSurge or Strava. They simply do a better job in tracking training than any of the software from the native devices (Garmin Connect, for example) or hacking together a solution from Word or Excel. My personal preference is to use TrainingPeaks with WKO5 as an adjunctive piece of software. The latter is overkill in almost all situations, however.

The first question- ‘How did it go?’

While the remaining words in this article will be dedicated to numbers, algorithms and technology, I would be remiss to mention the most important piece of feedback an athlete can give: their subjective feedback on the run. This variable is so important, it is the first thing I turn to when opening up a training file. An athlete’s answer to ‘How did it go?’ provides context and color that pace, heart rate and the amount of vertical gain have no way of describing. Subjective feedback is an amalgamation of all training variables (volume, intensity, cumulative training load, etc.) all at the same time. However, because of this all-encompassing nature, it’s not specific enough to tease out specifically why certain things are happening (is the volume enough, is the intensity right, etc.). So, when evaluating training, I start with asking an athlete how it went, then dig into any individual variables from there.

Volume of training: easy, useful and practical info to track

Three metrics can describe the volume of training: miles, time and vertical gain/loss. These three metrics are a no brainer to track. Any GPS watch (albeit, with more or less accuracy than others) and any platform can track these variables for a given workout and over time. They all have become ubiquitous markers of training volume (‘how much vert did you get, bruh?’). In tracking these metrics with athletes, I pay particular attention to the ramp of time volume (keeping weekly time increases to <~10% for most runners in most situations) and the amount of vertical change per mile (trying to match it as closely as possible to race conditions in the final several weeks of training).

Training Intensity: Normalized Graded Pace and Grade Adjusted Pace

One way trail runners can get a better answer to work load and work rate is by using an algorithm in either Strava or TrainingPeaks that converts the cardiovascular effort (key word there) of running on uphills and downhills to flat, level running. TrainingPeaks’ Normalized Graded Pace (NGP) and Strava’s Grade Adjusted Pace (GAP) take your running and “grade adjust” it as if you were running on flat, level ground. For example, if you are running 10:00 min/mile up a 6% percent climb, your NGP would be 7:54. In other words, running 10:00 min/mile on a 6% percent uphill grade is comparable to running 7:54 min/mile on the flats. This gives athletes the ability to compare paces on different climbs and descents with the equivalent pace for flat, level running.

Both of these algorithms do a decent job of comparing the respective paces (and therefore work rates) of climbing and flat, level running when performed on similar surfaces (another key word there) and at <20% gradients. You can go out and do intervals on flat sections and climbs, compare the efforts, and determine which effort was harder or easier. However, there are two glaring flaws in utilizing these algorithms for trail runners.

Flaw #1: When the surface is different. Neither GAP nor NGP has the ability to account for the difference in work associated with running on different surfaces. You intuitively know that running through sand is more difficult than running on a track. Similarly, running over technical terrain requires more effort than running over smooth terrain. However, the paces and calculated GAP and NGP will not account for the difference in those surfaces.

Flaw #2: Descents. The calculations for GAP and NGP use the difference in energy cost between uphill, downhill, and level running to arrive at the equivalent pace for level terrain. While on flat ground, your pace is directly related to work rate. Your cardiovascular system has to work harder in order to go faster, which means a higher pace requires a higher work rate. But remember that for downhill running in particular, energy cost tells only part of the story.

For downhill running, other factors outside of the energetic cost combine to significantly affect the overall training stress. These include changes in foot speed, coordination, and musculoskeletal stress, all of which are different in downhill running than in level or uphill running. Neither GAP nor NGP takes these additional stresses into account, and as a result they underestimate the overall stress of downhill running. In a single or shorter run, that may not be a big issue. For ultrarunners, however, it represents a greater flaw as you try to sum up cumulative training stress during a very long run or over a longer period of training.

The take home message for athletes is that if you are running on generally consistent surfaces day to day, these two tools can be valuable for you to track and monitor.

Work load and work rate: The holy grail

If I could waive a magic wand and solve any coaching problem for trail runners, that power would be used to determine work load and work rate for trail runners, which are better determinants of volume and intensity than the ones previously described. If you remember back to your high school physics classes… OK, let’s be real, most of you were asleep. Any of you that were actually awake might remember that Work is a product of Force (the magnitude and direction with which you push on something. In a runner’s case, the ground) and Distance (Work = Force X Distance). A bigger Force applied over a longer Distance leads to more Work done.

In level running, the amount of Work performed is a function of your weight and how far you ran (distance). Therefore, for level running, work load can be well represented by how many miles you ran and how much you weigh. This is why road runners can better describe their training volume by miles, it is essentially capturing the work load performed.

Work rate (or Power) is the total amount of Work done divided by the time it took to perform it (Power = Work/Time or Power = Force X Distance / Time for the Algebra geniuses). The faster you run any given distance, the higher your work rate. For flat running, work rate can be represented by pace and weight. This is why road runners can accurately describe intensity by pace, it is essentially capturing the work rate performed.

But for trail runners, who run on undulating surfaces and terrain, using miles and pace to track work load and work rate are confounded. Running Power looks to solve that problem by accounting for the work necessary to overcome elevation gain, wind and different surfaces (to an extent).

Aggregating Total Training Stress

Within TrainingPeaks, your total training stress can be aggregated and trended over time. This provides one of the more valuable pieces of feedback when analyzing training load. Acute Training Load (ATL) and Chronic Training Load (CTL) provide snapshots of the long-term (>7 days) aggregate training load.

Acute Training Load is a 7-day weighted average of the training stress for each particular day. Cumulative Training Load is the same thing but over a 42-day period (the time frame for CTL and ATL can be customized). By looking at both of these numbers over time, I can obtain a big-picture view of historical training that provides clues as to when an athlete is the most fit (highest CTL), most fatigued (highest ATL), and most ready for performance (positive difference between CTL and ATL). TrainingPeaks provides useful charting capabilities of these metrics that athletes can use simply by uploading their training files on a day-to-day basis.

Monitoring Improvement

How do you know you if are becoming more fit? In the road-running world, your day-to-day paces and workouts can provide answers to this question. In trail running, the process is similar, though the answer requires more investigation, particularly if you are doing your specific interval work on trails (as you should be, right?).

Both Strava and TrainingPeaks provide tools to help you understand how your fitness is trending. These tools do not provide stoplight answers that turn green when things are good, red when they are bad and yellow for somewhere in between. Rather, they provide general trends you can interpret to see if your training is on the right track and if you are making improvements over a period of weeks or months. 

Strava and WKO5 segments. One of the great and convenient features of Strava is segment tracking. Strava segments are marked-out sections of road and trail created by users. You can even create your own segments for sections of trail you commonly use for specific interval work. This feature also exists in TrainingPeaks’ recently released WKO5. While you should not be trying to set a PR every time, the general trend line can provide clues to how your fitness is trending, particularly if the segment of trail is one you use frequently for training. The great thing about these segments (other than being addictive) is that they are automatically calculated and tabulated once you have uploaded your run. No need to print out pages and mark them up with a highlighter.

NGP between efforts. TrainingPeaks provides fitness tracking via the capability to break down files and create laps once the file is uploaded. In this way, after doing specific workouts, you can come back to the file and analyze the NGP of the specific segments. You can use this to gauge the quality of your workouts from interval to interval and by looking at different intervals across several days or weeks.

What I Use To Track Training

I will let you in on a secret. Very little, if any, of this information is actionable by itself. In all my years as a coach, I cannot recall a time when I have taken one piece of such information and used it to decide what an athlete should do. Rather, action is determined by the aggregate of the information, combined with feedback from the athlete. From a practical standpoint, I utilize the following system to drive the creation of and adjustments to an athlete’s training. I encourage you to do something similar.

• Use TrainingPeaks to assign workouts
• Use NGP and Strava/WKO5 segments to evaluate and compare workouts.
• Use CTL/ATL to track training load.
• Use feedback from the athlete to further gauge fitness, fatigue, and motivation.
• Use knowledge and experience to synthesize the information and drive action.