Blog Homepage

Using histograms to identify your threshold power

One of the great things about modern day software is the array of functionality they offer.   One of my favourite tools is Golden Cheetah (GC) and it provides some really nice histogram functionality that I use to keep tabs on how my power is developing.

This post provides a couple of examples of histograms from a few of my training rides and races and shows how you can use this tool to hone in on your threshold power.

First up a quick explanation of what a histogram is.  Putting it simply, a histogram is a bar/column chart that provides an indication of how much time was spent at each intensity.   A good histogram will allow you to change the size of the ‘buckets’ which then increases/decreases the range that each bar captures.  This can be very useful, as you will see shortly.  Histograms can also be used to review heart rate, speed or cadence information.

Histogram #1 is from the Blayney to Bathurst 70km event.

image

The first thing to note about this chart is the amount of time I spent not pedalling (about 13%).  This is a handy bit of info as it can help identify how hard the race was, however in the context of this post it actually hides the information we are after as it makes all of the other numbers much smaller (relatively).   So, the next step is to hide the zeros and GC can take care of this with one click.   That leaves us looking at the next chart.

image

This new view gives us much sharper focus on the changes between bars.   You will see that I have highlighted one particular drop, along with its power band (260-270 watts).    Basically, the theory goes that the body knows when its reached the threshold point and as spending time over that point is quite demanding it’s natural that you do not spend too much time over it.   What this chart shows is that I spend most of my race time between 160 and 270 watts with a significant drop in time spent above this range.  This is an excellent indication that my threshold power is somewhere between 260 and 270 watts.  Lets zoom in and see where exactly it occurs.

image

For this step I prefer to start with a bucket size of ‘3’.  In my experience it provides a good indication of what’s going on at a micro level.  As you can see my power still holds up until I get to the 267-270 watt range.  After that it drops considerably.  So its a good guess that my threshold power for this race was 270 watts.  However, its worth considering the second point I highlighted in the chart.  This is the drop off the highest part of the chart.  This occurred at 240-243 watts.    Considering the race was a scratch race and that it’s common to spend some time over threshold, pulling turns, climbing, etc, its possible that the higher range reflects a pace just over my threshold (perhaps my climbing threshold).     For this reason it’s good to use a separate race or training file to complete the analysis.   As it so happens I did a team time trial today and I have the power file all ready to go.  

Ok, so starting with the bucket size of 10 again this is what we see.

image

Again there are two points that could be the threshold.  But as this was a time trial its more likely that I would spend as much time as possible at my threshold.   The team time trial does require some time over threshold as well but too much time there puts me into the red and out the back.  This TTT also had two climbs each lap and that will have boosted the power a bit.    So at this stage it’s looking like my threshold power is somewhere between 250 and 280 watts. 

Let’s see what it looks like with a bucket size of 3.

image

Not quite as revealing as we had hoped but if you ignore the prominent point (its only .5% higher than the others) then its still that second drop off that catches the eye.  Where to now?  Well this is where you start playing with the bucket size.  Start by goes to 5 or 6 and see what that highlights.  

image

We can see a nice drop-off at 276 watts.  I also tried buckets of 7 and 8 and these showed the same drop but the ‘6’ reduces the power range. 

So what can we conclude from this.  In Bathurst my main drop-off point occurred between 260-270 watts, with 267-270 watts being the likely range.   The Team Time Trial revealed that my threshold was likely to be between 270-276 watts.    The number 270 seems to be popping up a lot doesn’t it.  Who wants to guess that this is my current threshold?   I do! 

How would we confirm this?  Well, checking other files helps as well but you need to remember that to use this method you need to be quite selective about which files you check.  The training session or race must have been a hard one and required you to spend a reasonably amount of time at threshold but without it being too erratic.  Criterium files tend not to be as good as road race files with time trial efforts or races being the better ones.

What else can you use histograms for?

E1 zone confirmation.

image

This histogram is from my Saturday Aerobic Development ride.  The intent of this ride was to spend as much time as possible in my E1 zone – so working, but not too hard.   Ultimately these sessions go a lot off feel with the odd look at the power meter numbers to confirm I am working hard enough.   You can see quite clearly that I spend the vast majority of my ride time between 200 – 230 watts.   I would rate this ride as a good example of endurance work and I could have kept this pace up for quite a bit longer.  That’s good enough for me to know that I am on the money with 230watts being the upper limit of my E1 at the moment.

Now consider this ride from 31 March this year (5 weeks ago).  Same type of session with a shorter duration (1.5hrs vs Saturday’s 2hrs)

image

The power range is much lower, 180-215 watts.    This is a great indication that my aerobic power has improved over the last 5 weeks (including two off the bike due to illness prior to Bathurst).    This also shows my training is working with 15-20 watt improvement in my E1 power zone.  Gotta be happy with that.

What about preferred cadence?

Yep you can find that as well.  Check out this histogram from my Canberra Tour time trial a few years ago.

image

It should be pretty clear that (at that time) I preferred cadences of ~100rpm on flat to undulating courses.   You can do the same thing with hills but its a slightly different method.  This next one is from my hills ride Saturday of last week.

image

What you need to do here is to create an interval/lap for the hill climb (or several hills) and then select these in the ‘intervals’ list in GC.   You can then see the cadence histogram changes to show two streams of data, the first is the entire ride while the second is the selected interval.  You can see from this chart that my currently preferred climbing cadence (threshold intensity) is 90-96rpm.  

Anyway – that’s enough charts for today.    I hope this post has shown you a few new ways you can use histograms to track your training progress, especially your threshold power.   

 

Jason Mahoney Monday 14 October 2013 at 08:04 am | | Tips




No comments

(optional field)
(optional field)

Comment moderation is enabled on this site. This means that your comment will not be visible until it has been approved by an editor.

Remember personal info?
Small print: All html tags except <b> and <i> will be removed from your comment. You can make links by just typing the url or mail-address.