The more you know, the less you are sure

"A man with a watch knows what time it is. A man with two watches is never sure." Segal's law.

Superficially this old saying would appear to be arguing in favor of the KISS principle -- Keep It Simple Stupid. But look a little deeper and you see a more subtle difference. The man with one watch only thinks he knows the time. Both men are equally ignorant of the correct time.

Confused? Well, I am. Not about watches. My confusion is about training. Planning, analyzing, and adjusting. Today's variety of training zone systems is confusing enough. Add to that the quirks of websites and monitoring devices and the situation seems hopeless.

Anyone who has been in this sport for a while has gone through multiple generations of ever improving monitoring devices. When I started tracking my activity I was only riding, mostly my daily commute from Kahala to Kalihi. (Even today people ask me if I still ride to work. I don't, because I took up running and triathlon.) I got started monitoring my effort with a heart rate monitor. No training involved, just curious. First a Polar that just displayed time and heart rate, and later a Timex that added the ability to record a ride and upload the data into a special app. No GPS data, just time and heart rate. The Timex (an "official" Ironman watch!) came with a little book that introduced me to heart rate zones. As I recall we did two simple tests, then used the heart rate at the end of each test to work out our threshold, then divide the range into five equal sized parts. Primitive, but a start.

My first smartphone was an HTC, because they supported a pro bike team. Better still, the team all carried the same phone so people could track their location and metrics on a website in real time. I was sorely disappointed to learn I could not track my heart rate on my phone, because they used specially modified phones with an ANT+ radio. So much for truth in advertising.

My snazzy new phone did have GPS. I would record my rides with a free logging app from Google and upload the file to MapMyRide. Nice, but all I ended up with was a bunch of maps.

In 2011, probably while watching the Tour de France, I heard about TrainingPeaks.  I got started with a free account. That was around the time I decided to run the Honolulu Marathon. I started out logging runs with my brand new Garmin FR610. Soon I was logging my commute rides recorded by an Edge 800. I specifically recall asking Pattie about that, could I spend that much? She insisted I get the 800 because it did mapping, and she was convinced I would get lost without it.

For 2012 I moved up to a TrainingPeaks "Pro" account, which meant I could plan my workouts instead of just logging as well as a bunch of other cool features, like Annual Training Plan (ATP). I did my fist ATP that year, still not sure what I was doing, relying entirely on a really cool feature they since discontinued, the Virtual Coach.

After I got into running my friends all pointed out that I might as well do triathlon. I figured, why not? All I needed to do was work on my swim. Ha!

Lucky for me I found Joe Friel's book, "The Triathlete's Training Bible." I didn't have a clue who he was. I just happened upon it at Barnes & Noble, and it was the only book they had on triathlon. He had a lot more to say about training zones. I got really good at doing lactate threshold tests and setting my zones, going as far as recreating Friel's formulas in a Google Docs spreadsheet.

Triathlon training got serious when I decided to take on Honu, the Ironman 70.3 Hawaii race. Garmin had finally brought out their Vector power meter and I plunked down the big pile of cold hard cash it took to buy a set back then. That got me to read Friel's "Power Meter Handbook." On the bike, at least, power replaced heart rate as the standard for managing my training. For Honu 2016 I bought a Friel power-based training plan and liked it enough to use it again in 2017.

Friel uses a relatively traditional zone system. The dividing lines between zones are slightly different for the run and bike. Here, for simplicity, I will only consider the bike. He puts functional threshold heart rate at the start of zone 5. He then divides zone 5 into three parts, 5a, 5b, and 5c.

1 up to 81% of bFTHR
2 81% - 89%
3 90% - 93%
4 94% - 99%
5a 100% - 102%
5b 103% - 106%
5c 106% and above

For power he uses six zones, without any subdivisions. They are designed so that, on average, the HR zone and Power zones match. Riding steadily at the middle of Power zone 2 should put HR in mid zone 2, more or less.

Back when I took up running I looked around for help. At first I followed Bobby McGee. Great stuff, but not as directly useful to me as a just getting started runner. I learned a lot from him, but lacked the experience to turn it into a training plan. I still go back and review his excellent lectures.

Then I stumbled upon Matt Fitzgerald. The first book of his I read was about nutrition, but it went far enough into training that I used his plans to train for the marathon. Then I found his book "80/20 Running." Last year I think it was he brought out "80/20 Triathlon." Being a fan I snapped it up, and was pleased to see he that he goes much deeper into training zones.

In Fall 2017 I was still not sure I would do Honu a third time. I had a terrific bike, a Cervelo P3 and Vector power meter. What else could I want? Well, by then Stryd had made a big push of its running power meter. Dr. Andrew Coggan himself pitched the new technology. Seemed like a no brainer, especially when you consider how much cheaper the Stryd was compared to my Vector pedals. I decided to give it a try.

Having used a Friel plan twice for Honu I decided to give Fitzgerald a try. I was not convinced until I downloaded a sample and saw how detailed he gets specifying power on the run and bike, and making good use of TrainingPeaks relatively new structured workouts. In spite of all sorts of life challenges I followed the plan and had my best race ever.

One thing that makes Fitzgerald's 80/20 system unique is his zone system. He makes the case that too many endurance athletes spend too much time in between the optimal high and low intensity levels. Much of his book is devoted to making this point. Athletes should spend 80% of their training at low, easy efforts, and 20% at medium to high efforts. This is based on the work of Dr. Stephen Seiler. To drive home his position he created two zones to be avoided called X and Y, located below and above the traditional zone 3.

While all this was going on I was using -- let's call it playing around with -- WKO4. Initially aimed primarily at cycling, as Stryd usage grew I began to see some run specific material on WKO4. I only just started using the new WKO5, now referred to simply as WKO, but I expect running and other aerobic sports to be better supported.

At the risk of oversimplification, WKO differs from older systems in that it offers the athlete specific training targets based on recent performance. Previously, the athlete would perform a threshold test, plug the result into a formula and get a set of training zones. The formula used was based on data samples from a large group of athletes. The results work reasonably well most of the time for most people, but there will always be outliers and variation. WKO overcomes these limitations by tracking performance at all levels of intensity, not just threshold, and adjusts training zones in a manner best described as dynamic.

One of the many interesting features provided by WKO is a unique zone system called iLevels, based on recent work by Dr. Andrew Coggan. As far as I know it only works with power, but that being the gold standard for serious cyclists and the only practical way to monitor intensity during high intensity intervals, that makes sense. Coggan feels it is important that effort be focused on specific targets. Not so much physical parts of the body as physiological systems. Training at specific intensities stresses those systems in a specific way. To do this WKO divides the range above functional threshold into five zones (where Friel has three) and adds one more just below threshold. Based on the athlete's data, the program calculates the raw ingredients to design a workout -- power range, duration of hard effort, duration of recovery, number of reps, etc. These specifications are dynamic and tailored to the individual athlete, thus the "i" in iLevels (apologies to Apple).

Below are the iLevel zones as I had them defined in TrainingPeaks.

1 Recovery
2 Endurance
3 Tempo
4a Sweetspot
4 FTP
5 FRC/FTP
6 FRC
7a Pmax/FRC
7 Pmax

I must have gotten the percentage range off of a WKO video. I don't have them just now, but we don't need them to make my point. First, WKO uses nine, count 'em, nine zones. Which means some if not all are narrower than Friel's. Then there is the numbering. Who puts "a" after a number and follows it with just the number? 4a Sweetspot is lower than 4 FTP. Hmm.

So let's summarize. Three training zone systems, similar but different. Different zone names, different dividing lines. If I want to follow a Friel training plan, or use one of his excellent books, I need to use his zones. Same goes for Fitzgerald. I don't have a training plan that uses iLevels, but I do enjoy tracking my progress in WKO and iLevels are the best match there.

Speaking of iLevels, there is a detail involving TrainingPeaks that is really too technical to go into much detail here. To put it simply, Friel and Fitzgerald zones work fine with TrainingPeaks. The website even supports those silly X and Y zones names. It can handle iLevel zone names, but I am not sure how well it works with the higher zones, because, as I understand, these are derived dynamically from the very complex mathematical model at the heart of WKO.

I use an Garmin Forerunner 935 multisport watch and, on the bike, an Edge 520. Each has its own unique way of entering training zones. In both cases there are no letters, and no subzones like 4a and 5b. For the Friel zone system this is not much of a problem. I make 5a, 5b, and 5c one big zone 5 and try to target low, middle, and high within that range. The Fitzgerald and WKO systems are a lot more trouble, because to get the numbers to make any sense I have to combine zones that were separated specifically to avoid training at the wrong intensity.

But wait. It gets better. TrainingPeaks has a really cool feature supported by Garmin called Structured Workouts. This allows you to design a workout based on pace, heart rate, or power. You do not specify an absolute value, rather a percentage of the appropriate threshold. You are supposed to track and update your thresholds as your training progresses. When the workout is downloaded to the Garmin the percentages are replaced by absolute values, based on your current threshold setting.

When the time comes to do the workout the Garmin will provide one or more extra screens to display information about the workout. The really cool feature is a scale, red - yellow - green - yellow - red, with a mark to show current value. If the current segment is supposed to be at 120 watts and you are producing 120 watts, the mark will be in the middle of the scale. At the top of this post is a photo of me running with my FR 935. Walking in zone 1, actually, but even so it came out blurry.

This seems like a terrific feature, but it has some serious limitations. On the bike (where I use the Edge 520 but it works the same on the FR 935) I use power, and this works great. I do not need to know what zone I am in during the workout. I just see the watt range I am supposed to be in, time remaining for that segment, and try to center the needle. Perfect.

For running things are far from perfect. I wrote about this in a previous blog. That was at the end of last tear and nothing has changed. Garmin uses a dedicated ANT+ channel for bike power, but not for run power. The only way to get run power into the watch is through a ConnectIQ app. Using the Stryd ConnectIQ app I can put a power value in watts on a standard screen, but not the structured workout screen. When a power-based workout is loaded, you can see the bike power icon but no value. I can scroll to the screen with the ConnectIQ field, but that only shows current power in watts. For bike power I have the option of displaying the current power zone, as long as it is just a number. There is no way to display power zone. Just watts. Am I supposed to be at 100 watts? 150? 200?

I could say that I want running with power to work like running by pace, which is what I was doing in the photo, except that has a huge flaw as well. That big number by the scale? That is not current pace. It is average pace for the segment. So on a three minute interval if I look down and see I am slow, I speed up and see no change. That screen is not configurable. Bike with power shows three second average, which is useful and pretty much follows the needle. Running with pace, the needle tracks current pace but the number does not.

So where does all this lead me? I need to find the path that works for me. My initial reaction is to create my own workouts that blend the best of the Friel and Fitzgerald systems, but somehow take into account the new developments on WKO. Then I stop and think about how much work that will be. More on this topic in future blogs.