In this post I’ll discuss one way to analyze a swim set using the data from a Garmin 910xt Swim Watch. Similar data and graphs can be obtained from the Finis Swim Sense but the graphs look slightly different. Understanding the concepts of how pace, strokes per length and rate fit together is the most important key…if you understand that you can analyze the data in just about any system…or even by hand, tracking the information with paper & pen after each repeat.
Step 1: What is the intention of the set?
This is a 1000 yard set of 10 x 100yds freestyle posted by a triathlete early in his season. Like most folks, his goal is to get faster for his race distances. So lets take a look first at the dashboard and look at the key pieces of information.
The first thing I look for, if I don’t already know is a sense of what the set represents. It’s easy here to see this is a set of 10 x 100 yard repeats with short rest intervals. Essentially this is a “broken 1000 yd swim” due to the short rest, as opposed to a series of ten 100-yard efforts. The less rest between repeats, the more continuous the energy demand is, and the better it reflects demands of a triathlon swim leg or open water race.
Step 2: How is the pace control?
Next, the list if intervals on the left hand side shows us the distance and the pace of each interval. if yo don’t own a swim watch, this same data can be easily tracked with any sort of sports watch with a timer that allows you to start and stop each interval. Because pace is usually listed as pace per 100 yards, and these are 100 yard repeats, the pace and the time are identical for each repeat. You can see that his pace remained in a tight band of 1:28 – 1:32 seconds/100 yds for 1000 yards. While the rest interval isn’t show in the left hand panel, you can see from the blue & red panel on the lower right that the rest period looks to be about 10 seconds per 100 or less, just by comparing the width of the rest interval to the known 1:30 min:sec swim time.
So the second piece of data we collect is a gestalt of the overall pace, and how consistent the athletes pace is. He gets an A on pacing for sure. Do you agree?
Step 3: How is the Stroke Length Control, ie Technique or Skill Endurance?
The next thing we will look at his how this athlete constructed his pace, and is a reflection of how well his current skills hold up with fatigue and distance. This is the part that can really help accelerate improvements, because if we try to increase pace or performance before we are ready, the athlete can stall in his technique development which makes overall development limited (even if still he is still improving )
The top right panel shows us strokes per length for the entire set. I like this panel because it gives me a good feeling for stroke length control, which reflects skill, coordination and neuromuscular movement patterns. Since I already know this set is a 1000 yard broken set, ideally what I’d like to see in a well developed swimmer is a steady SPL for the entire set, and that’s exactly what we see here:
There are a few variations on his 8 stroke cycle average. Overall, rarely more than 1 stroke cycle variation from his average, and those variations are occasional, not on every length. For clarity, these swim watches count stroke cycles which is a full left and right arm cycle. The watches do not measure 1/2 cycles or an odd number of strokes. For that reason I prefer to have swimmers get into the habit of couting their own strokes when possible so they can begin to understand and feel the difference between say a 16 stroke length (8 cycles) and a 17 stroke length (8.5 cycles).
So now we have 3 pieces of vital information:
- What type of set
- Overal pace control
- Overall stroke length control (skill or technique endurance)
For this athlete we have gathered just by reviewing the watch data
- Type of Set: Aerobic Endurance, broken 1000yd Freestyle as 10×100 with short rests
- Overall Pace Control: Average pace of 1:30 with a range of 1:27 to 1:33
- Technique Endurance: Average Stroke Cycles per length of 8, with variation of 1-2 cycles in some repeats. (note that the watches are not always accurate in cycle counting either, sometimes adding a cycle or not completing a cycle depending on how you turn at the wall and what are you swim with or touch the wall with)
At this point you have everything you need to make a determination of how well the athlete performed this set, and what changes if will help contribute to the next stage in improving his racing speed.
There are a few other data points it would be nice to know in the big context, but this is enough information for a quick “microcycle” scope review of a standard type of swim set. Before you read the rest of this article series, think about any other data you’d like to know in order to make plans for the athlete in a larger time scope.
What do we do with SWOLF?
The next chart is the SWOLF chart. I present this here because it seems like there should be a gold mine of data here…but I’ll just go out on a limb and say at many times, this is fool’s gold. We have to ask our selves if we are really asking the right question. What does the sum of two different units represent? Do I want my SWOLF to be as low as possible? If I lower my SWOLF am I improving?
It’s far more important that you understand the concepts and meaning of the 3 data points we have already collected, but Several different patterns can emerge with this chart so it is worth at least thinking things through. SWOLF is the sum of strokes per length (stroke cycles) plus time in seconds per length. This number has no units and conceptually we can’t place an absolute value on it’s meaning. What we can see in this graph is that that number is steady as well…but we knew that it would be. In a differently paced set, or a less consistent swimmer, this chart would have a different “shape”.
Where do we go next? What’s my next workout, coach?
Since we’ve already extracted the important bits of data in the first 3 steps so let’s continue with what happens next! The next part of this article will discuss how to interpret the information we’ve gleaned from the watch data and plan future sets. Before reading it, think about what makes sense here to continue the athletes aerobic abilities, skill/techinque endurance and overall speed. What would you recommend this athlete do next? Several different options are available so lets’ brainstorm a little.
Feel free to comment below before reading further.
With the rise in swim watches that use accelerometer devices that can count strokes, detect how many lengths of the pool we’ve swum and keep track of our pace and time for every lap and repeat, triathletes and swimmers are now asking questions about what they can do with the information obtained.
The two most popular sites that currently track and report this type of data are teh sites of the watch makers themselves, Garmin and SwimSense. Training peaks also can import this data but as of this writing they lack a robust reporting feature.
The most easily obtainable and somewhat interesting metric reported is the SWOLF score. SWOLF is an abbreviation for “Swim Golf”, and is a score obtained by adding together your strokes per length, and the time for the length.
e.g. 25 yards swim at 30 seconds in 20 strokes is a SWOLF score of 50.
Below is a screenshot of such a reported metric from Garmin Connect:
And here is a similar example from Finis, the makers of the Swim Sense.
So this begs the question…what can SWOLF tell us?
It’s been suggested in the past by many coaches that SWOLF is a measure of efficiency…that a lower SWOLF score is a more efficient stroke and we should strive to get SWOLF as low as we can. However, that line if the analysis is misleading for two reasons. In the rest of this article I’m going to discuss these two reasons and at the end suggest a better way to use SWOLF…and possibly better ways to use your free time out of the pool.
The first reason that the SWOLF score is not a very good measure of efficiency…if it can be considered a measure at all, is that efficiency is the relationship between the amount of work energy that goes into an activity relative to the work output, or forward movement achieved. For example, if we could measure how much oxygen a swimmer was consuming while swimming in a flume at a specific speed, then we could measure his or her efficiency. External measuring tools of oxygen consumption are needed. So while we can’t get an actual efficiency number from a swim watch, what we can do is incorporate certain clues that may tell us something about the efficiency of the stroke.
But more interesting is the second reason that SWOLF can’t tell us much about the efficiency. Even though SWOLF is derived by adding two parameters together…namely Strokes per length (SPL) and the time in seconds for the length, we still need to know both the SPL and the TIME independently to make any sense of the SWOLF number.
Perhaps we are asking the wrong questions.
Let’s create a possible matrix of SPL and Time parameters, to keep it simple let’s just use HIGH and LOW for SPL and time and see what may happen to SWOLF and what it means. If it seems confusing at first, keep reading…please…you have to trust me…
Here are four hypothetical scenarios which I describe in a bit more detail after each one.
If if SPL * rate = time, then we are curious about what SPL + time or SWOLF can tell us?
A) SPL is high and time is high = high SWOLF and inefficient swimming
B) SPL is high and time is low = medium SWOLF and improved efficiency
C) SPL is low and time is high = medium SWOLF and very efficient swimming
D) SPL is low and time is low = very low SWOLF and inefficient efficiency
A) high spl and high time (ie slow) suggests a lot of drag and or inefficient catch. Scenario A swimmers can learn a lot from tracking SWOLF and watching it improve because experiencing either a lower SPL or a faster speed suggests they are improving swim efficiency
B) as speed increases an expected and normal response of efficient swimming is for SPL to increase (SWOLF may stay the same). Scenario B swimmers can expect to see a lower SWOLF for their easier swimming and a higher SWOLF for their faster swimming, and should strive to find a more sustainable faster speed…and be happy with a higher SWOLF. (but to discover what it is, or more specifically what their SPL & Pace targets are for various training distances and /or race settings
C) suggests very low stroke rate with lots of glide…little energy being put in compared for forward movement. Physiologically speaking a very efficient stroke (SWOLF may be the same is in scenario B). Scenario C swimmers should strive to increase their SWOLF, more specifically by increasing their tempo as they already have a very efficient stroke.
D) The ability to swim quickly and hold a low SPL requires high power, high strength swimming, and can be seen in Jai’s 2nd video here: SWOLF will be low but true efficiency is low. However if this is a race, fastest time, not highest efficiency wins. Scenario D swimmers are very skilled, fast powerful swimmers, and can pick and choose the stroke suited for their task.
In that video Jai swims two lenghts. Both have the same SPL, but in the 2nd length he adds effort and speed to reduce his time. Here are the numbers:
Length 1: 16 strokes 31 seconds SWOLF 47
Length 2: 16 strokes 18 seconds SWOLF 34
The slower length is actually the more efficient stroke considering the amount of effort compared to the forward movement. it takes Jai more energy to swim the faster length. Even without having the external measuring tools to determine true efficiency, it should not be difficult to believe that Jai could sustain the first length’s effort for quite a long time…the second length for possibly 100m or less.
A graph would be interesting here… plotting combinations of Speed vs. SWOLF at different SPLS/rates and seeing if there actually is a relationship we can glean that would be useful as a self-coaching tool. My guess is that there is not…we still need to know each metric…at least two out of the three (SPL, Time or Speed) in order to analyze data and know how use it to improve our swimming practices.
I find SWOLF most as a quick measure of consistency, rather than a measure of efficiency. When i look at my athletes watch data, the SWOLF graph can reveal if they are swimming with too much rest or too easily. Or conversely if they are improving their ability to be consistent with a given set (eg. 10 x 100 as an example). Pace alone doesn’t tell us, SPL alone doesn’t tell us, but SWOLF alone does not tell us either.
e.g. if I see a set of 10 x 100 with a SWOLF that does into change AND an SPL that does not change, they are swimming consistently which is good. I still need either pace or SPL along with SWOLF to learn anything about their swimming, it doesn’t stand alone. If I want them to swim faster and get fitter as a triathlete (my main market). SO I can reduce the rest, increase the repeat distance, increase the pace (which is possibly going to change the SWOLF ).
Hopefully that gives you more insight as to what some of the metrics mean and what you can do with the information.
I still find that practicing SPL ranges at will, finding your SPL, changing your SPL and holding your chosen SPL, layered with the use of the tempo trainer (steady, ascending, descending, etc) gives the coach and athlete much more control and direction than SWOLF had the possibility to do.
IN a follow up article I will use some specific watch data graphs to help walk through an analysis of a swim set’s execution and plan further practice sets for improved skill.
Please let me know in the comments section what your thoughts are and how you use the watch data you collect.