Ramblings About Numerical Things is a series of articles sharing my thoughts and opinions on the meaning or application of numbers in a particular area of interest. I’ll try not to go overboard on the numbers themselves, but rather focus on how you can use numbers to meet your goals.
No, this particular blog isn’t about creating a mantra… the power of positive thinking… or visualization to help you endure the pain of racing. These may all be useful tools, but I don’t think they provide the key to making the most of your race. Have you ever gotten to the end of race, and thought you should have pushed some or all of it harder? If so… how much harder should you have pushed, and why? Have you ever ridden too hard, and found yourself walking a substantial part of the run leg? I believe that mistakes only become failures if we fail to learn from them… and the following is what I’ve learned, researched and implemented to help create realistic and achievable race pacing to allow me to race up to the edge of my capabilities… without going over (or at least too far). 🙂
To create a good plan, you first need to understand your current fitness level. This isn’t a guess based on a different length triathlon you did two years ago, but is a quantitative value on where your fitness is today, based on recent fitness testing. Fortunately a testing facility isn’t required… you can do fitness tests by yourself. Going backwards to forward (running to swimming), the best resources I’ve found for testing & planning are:
A lot of research has been performed on runners, from recreational to elite runners, by a coach named Jack Daniels. What was found was that running performance followed very predictable patterns, based on VO2 max (essential maximum cardio capability) and running economy. Jack Daniels combined these two factors into one number, a VDOT value. By doing a 5k all out race simulation, you can find a corresponding VDOT value which will tell you what your threshold running pace is (what you could maintain for a 1 hour effort), predictions of what your 10k, half marathon, and marathon paces would be (assuming adequate training volume), as well as recommend training paces. A great summary of Jack Daniels’ theories and strategies can be found here. After understanding the basics of VDOT (don’t worry it’s a fairly straight forward and relatively short discussion), you can calculate your specific VDOT value here. Basically you put in your distance (5K) and your time, and it calculates your VDOT value, training paces and race paces.
To use the VDOT concept, you first need to do VDOT 5k test. Personally I like to do my testing on a treadmill, at 1% grade, so I can be certain of consistency between tests (weather isn’t a factor). You can also do the test at a local track or even a flat road or trail. Make sure you warm up before the test, typically walking for 5 minutes, easy jogging for 5 minutes, and then a moderate pace for 5 more minutes, with some quick 20 second strides to get warmed up. Make sure you have worked up to a sweat. Then run your 5k, keeping track of our time, and heart rate. The key is not to blow yourself out in the first 5 minutes… pace yourself.
Once you have your 5k time, you can use the VDOT spreadsheet, linked above. Let’s say your time was 21:01, which results in a VDOT of 47:
There is a lot of great information on this page, but for triathlon pacing and planning, the most important number is the VDOT number (47) and the threshold pace of 7:12 / mile.
Now… how is this turned into run pacing targets for a triathlon? Basically there are recommended ranges based on your functional threshold pace and the distance of the event. Here’s a great graphic I’ve found, showing ideal pacing by race triathlon length:
Using the original example, with a VDOT of 47, you should be targeting 103% of your Functional Threshold (FT) pace, for a total time of 21:48. This calculator can convert your total time of 21:48 to minutes per mile, resulting in a target pace of 7:01 per mile for a sprint 5k run. As the race lengths get longer, the targeted percentage of your FT pace drops, meaning you run slower (common sense). If you were planning for a full Ironman, your target range would be between 80% to 75% of your FT pace, for a target 26.2 mile run of between 3:58 and 4:14 (9:05/mile to 9:41/mile).
What if the numbers don’t come out exactly to a whole number… i.e. what if my VDOT was 47.5? You could interpolate on the table above, or simply do the math as follows:
- 5k time of 20:50 results in a VDOT of 47.48 and a Functional Threshold Pace of 7:08. Converting the FT Pace to seconds, we get 7 x 60 + 8 = 428 seconds. Dividing by the recommended FT Pace target of 1.03 we get: 428/1.03 = 416 seconds or 6:56 per mile pace.
So… to get your triathlon running target you need to do a 5k at an all out race level effort, calculate your VDOT using the calculator, identify your target pacing based on your VDOT and race length on the chart above, interpolating if you are between VDOT points.
Of course, these targets are “ideal” ranges, assuming good running conditions. If you’ve hammered the bike too hard, are dealing with high winds, temperature, humidity levels or altitude, you will need to modify your ideal ranges accordingly. For modifications based on less than ideal racing conditions, you can see typical impacts to paces here.
As mentioned above, when doing the 5k test, it’s also good to keep your heart rate data as well. If you simply ignore the first 5 minutes when your HR is getting up to speed and average the remaining HR data, you get an approximation for your Lactate Threshold Heart Rate (LTHR). This should roughly correlate to your FT Pacing, although often temperature, hydration levels, cardiac drift, can impact HR causing it to rise above normal values and making this a little less reliable indicator, particularly for longer races. I tend to look at my HR relative to my LTHR to help keep myself in check early in the run leg.
Again, it is important to understand that VDOT is an “ideal” target, assuming you have ideal weather conditions, flat course, and (most importantly) that you have done an adequate amount of training volume to build your aerobic efficiency. The point is that if you simply train to optimize your 1 mile speed… you can’t extrapolate to your marathon time and have a realistic target.
I think the bike leg is probably the most important part of the triathlon in terms of planning and pacing. If you push too hard, forget about coming close to the ideal running pace above… and you could end up walking the run. If you are too easy on the bike, you will never be able to make up the lost ground running, as in all triathlon distances you spend the most time on the bike leg.
Like the running example above, the first key is to understand where you are today for fitness. The best way to do this is a Functional Threshold Power (FTP) test, which is the maximum average power you could maintain for a 1 hour effort. Bike power is the ideal way to quantify your bike fitness level. As power meters can be expensive, I’ll give some non-power meter alternatives later in this article. The good news is that like the 5k running test, you really don’t need to hammer the bike for an hour to see what you can do. Instead, you can do a 20 minute time trial test and extrapolate to your FTP value. To do the test, you warm up with progressive power building for 15 minutes, including some high power segments (for minute or two at a time), close to your expected FTP level. Once you are warmed up, you simply do a bike time trial for 20 minutes at an all out race effort, trying not to burn yourself out in the first five minutes. During the test, you also keep track of your heart rate and speed.
Once your test is done, you can estimate your FTP by simply taking your average (or normalized power, if you have it) and multiply by 0.95 to get FTP. For example, if you averaged 253 watts of power for 20 minutes, your FTP would be: 253 x 0.95 = 240 watts. Once you have your bike FTP, you can create a power target for your bike leg. An article discussing typical ranges for varying triathlon lengths can be found here. Essentially the distances and the corresponding power based on the 240w example above:
Where do you fall in the range? If you are a very strong cyclist and runner, do a lot of brick workouts, and are competing at the top spots for your age group, then I’d look at the upper end of the power ranges. If you are relatively new to triathlon, I would target the very low end of the ranges. Typically I tend to target and ride around the midpoint of the power range, and I’m typically able to come close or beat to the target ranges of the VDOT run pacing discussed above. For reference, some of my bike/run data is shown below:
As the races get longer, pacing becomes even more critical. Ironman length races have other pacing considerations… basically the total stress you place on your body… as described in this article.
Now that you know your power output, what does it mean in terms of pacing/speed/time? The best way to predict this is to use software such as Best Bike Split, which I have reviewed here. This software does a nice job of converting power into speed (and time) based on the actual terrain, weather data, and a model of you and your bike.
What if you don’t have a power meter? You can also use HR for setting intensity levels on the bike as well, although it is not as accurate as a power meter (as discussed above). To use heart rate, you will still need to do a 20 minute bike test to estimate your Lactate Threshold Heart Rate (LTHR). Like the run test, LTHR on the bike is simply your average heart rate for last 15 minutes of your ride. (Note that your LTHR on the bike is usually 5 to 8 bpm less than your running LTHR.) Then when you go racing, you can use the same target intensity listed above, by using HR as follows:
- Method: Target HR = ((LTHR – Resting HR) x Intensity Target) + Resting HR
- Example: If you LTHR is 165 bpm, your resting HR is 50, and you are targeting an Olympic Length race and your goal is to target 90% intensity, then your target HR would be: ((165-50)x0.9)+50 = 153 bpm. Keep in mind this is an average HR (and may be impacted by temp, hydration, etc.), so you will normally run slightly lower than this in the first half of the bike leg and slightly higher on the second half of the ride, maybe plus or minus 4 to 8 bpm.
My first recommendation would be to do the LTHR test on a trainer, with the protocol described for the power meter test above… but simply tracking HR. If you want an inexpensive way to do the test and estimate your power without an actual power meter, you can subscribe to TrainerRoad, which uses resistance profiles of common bicycle trainers to estimate your power output based on speed (I think TR is one of the best cycling training tools out there). Whether using TrainerRoad or just riding standalone, the key is to ride at an all out 20 minute effort. Another alternative is to ride a local TT (12k if possible), as this should be relatively close to a 20 minute test for most riders. With any of these options, the goal is simply to capture your average HR for the last 15 minutes of a 20 minute ride, to estimate LTHR.
Like cycling and running, the best way to start your plan for swimming is to assess your existing swimming fitness & pacing level. Swimming is a bit unique in triathlon, as form can play a much larger role in overall pacing than in the other sports. Like VDOT for running, the concept of Critical Swim Speed (CSS) can be used to estimate your threshold swimming pace. The testing is quite simple – just two all out swim tests, one at 200y and the second at 400y. Based on my experience, I do these on two separate consecutive swim workouts (two days apart), to make sure that the first test doesn’t negatively influence the second, which would skew the data. For more detail on CSS, see this article.
Although there are fewer published studies or data on triathlon swim pacing when compared to biking or running, a reasonable approach is to use the CSS calculation to create “ideal” swim paces. I created the following online spreadsheet to calculate your CSS value, and the associated times/paces based on distances. Again, these are ideal times based on pool swimming, and there are a lot of factors in open water swimming that will impact actual results, such as: swimming off line, sighting, no wall push off, contact with other swimmers, drafting, and wetsuit benefits. I’m in the process of documenting and quantifying some of these factors individually for a future blog, but based on initial data… if you wear a wetsuit and are able to draft between 30 to 50% of the swim… you should offset the other losses and come close to your ideal pool swim times.
Let’s say an athlete decided they wanted to do their first Olympic Length triathlon at Hy-Vee in Des Moines. They recently completed a 5 k at 24:30, have a bike FTP of 180 watts, and did a swim test with a 2:55 / 6:30 paces for 200/400y. What is their ideal pacing for the event?
- Using the VDOT Calculator, their VDOT would be 39.24, with a threshold pace of 8:20/mile. From the chart above, the ideal Olympic pace is 98% of your threshold. Converting 8:20 to seconds, we end up with 500s/mile. Dividing 500s by 0.98, we end up with a goal pace of 510s/mile (8:30 min/mile), or a time of 52:49 (using this calculator).
- With an FTP of 180 watts, we first want to establish the power target for the race, which is between 85 to 95%, as discussed above. With this being the first Olympic race, I’d recommend the lower value of 85% for the target, resulting in a target wattage of 0.85 x 180 = 153 watts. Using Best Bike Split, modeling a female recreational triathlete, the result shows a total time of 1:19:02 (see output page here).
- Using the CSS online spreadsheet, the CSS calculation is 1:48/100 pacing, with a corresponding Olympic time of 28:54.
Adding 3:30 to T1 and 2:30 to T2 times, would result in a total ideal race time of 2:46:45, if everything went perfectly. Realistically you may want to add a little time to compensate for heat, hills, wind, swim contact, etc., as makes sense. Typically I would look at a range for this athlete of 2:46:00 to 2:55:00, with a realistic race target of 2:50:00.
I’ve talked about testing, to understand where you are for the model. Keep in mind, if your training is fairly consistent, you should test every 6 to 8 weeks. Not only does this help with the predictions above, more importantly, it helps you to modify your training zones so you continue to build over time. Note that usually for long course racing (half or full IM’s), often testing is ceased after a certain point, which could be 4 months or more prior to the race. As long as you continue to train and build volume, likely you won’t lose VDOT or power over this time (or very little), so earlier tests are still reasonably valid for predicted performance.
Are you looking for an easy button for this information? Stay tuned for my follow up blog on this, where I will create a spreadsheet that will tie together VDOT, power data and swim CSS, so you can have one spot to input data and get an estimated race pace/time.
I read a great quote that said, “Essentially, all models are wrong, but some are useful.” (George E.P. Box, statistician). The point is that using calculators like these may not be able to pinpoint exactly what your predicted time would be in a race, but they may be able to get you in the right zip code… “thinking” your way into a faster triathlon.
- I would strongly recommend reviewing the following link to a summary of Jack Daniel’s work here. This series of articles (four) discusses the basics of building fitness, training paces and “junk” miles.
- A nice VDOT calculator is located here. One caveat to using the calculator is that Marathon Nation coaching suggests that the VDOT calculator / table tends to be your “ideal” race pacing, and for practical purposes half marathon age group athletes may want to derate their VDOT value by 1 to 1.5 VDOT points for planning half marathons and 2 to 3 VDOT points for marathons.