The user data series use a syntax that is described here


I already have Jim's book, so if you tell me the page number I can check what you want and try and recreate it.

Mark

Oh its on p99, and really easy to calculate (its speed in meters per minute / power).

For my running I get a value between 0 and 1.1, which doesn't neccessarily mean much since I'm a poor runner and I used the GC running power estimation tool, not a Stryd. So I took Jim's calculation and multiplied by 100 to get an index that is a bit easier to work with (0-100+).

You can add it to a chart really easily:
((SPEED * 16.66667) / POWER) * 100

Will look to add it to the chart DB so it can be downloaded by others, I need to get to grips with Jim's book (it arrived yesterday).

CHEERS

Mark

Wow thanks a lot, that worked!

I am not sure why it did not work previously. I notices how Speed & Power were not shown in red before, maybe that indicated something?

Jim introduces several metrics in his book, I will try if I can add them all.

Also, I will try to get averages of the whole workout in the summary view. I will be very interested if you have added the chart to the DB

Keep me posted :)

You will need to create custom metrics for the summary (and trends charts), but relatively easy to do if you use the example you get when you add one (average power) as a guide.

Mark

Jim's "efficiency index" is what I have been calling "running effectiveness" for about a year now (as I believe he acknowledges in his book).

I have created a running dashboard for WKO4 that displays running effectiveness along with some other potentially relevant metrics:

https://www.google.com/...B3sbxktgk2Tb6WEk_QBw

ETA: I first calculated the average effective horizontal force in December of last year, but can't find where I first called it running effectiveness:

http://club.stryd.com/...parison-help/1204/33

Note: I receive royalties from Velocious Software, the developers of WKO4, and am a consultant for Stryd.

Hi Mark,

I could still use some help with the custom metrics;)

I agree it should be relatively straightforward. For average EI for the whole run I should replace speed & power with average speed & power in the formula you mentioned.
- When I type SPEED it turns into a red font, when I type AVERAGE_SPEED it does not. Does that mean the program does not recognise the parameter?
- What should I do with the text in the program box? Can I leave all that out and just put the formula there?

Any help is appreciated, I really like the Golden Cheetah program. I just have to get more familiar with the advanced functions ;)

There are a crapton of builtin metrics.

When working at the sample level (e.g. plotting second by second samples) then the telemetry data is available in variables like SPEED, CADENCE, POWER (the autocompleter will help you here).

When working at the ride level (i.e. plotting metrics across date ranges) the metrics all have names like Average_Power, NP. Again the auto completer will help you here.

Lastly, when working at the sample level you still have access to the overall metrics for the ride. So you could, for example calculate a running percentage of current power vs NP.

Stated simply: the autocompleter is your friend !

As previously mentioned, all the variables and syntax is documented here.

Hi Mark,

Thanks for the response.

Maybe I did not phrase my questions clearly enough. I have read the variables documentation, and the auto-complete is very helpful. Could you explain what the difference is in the colouring of the words? For example, POWER turns red, but AVERAGE_POWER remains black.

When I add a new custom metric, many steps are already present in the program window. For example, the relevant {} function, init{} initializing aggregating variables, sample{} calculating for each sample etc. According to the syntax list, the only function required is the value{} function. Does this mean that the only thing that needs to be in the program windows is: { value{ MY FORMULE }} ?

Again, thanks a lot for your incredible quick help!

AVERAGE_POWER does not exist. It is Average_Power.

I need to document it but

When calculating a metric, each sample is iterated over from start to finish. Each time the sample() function is called so it can calculate or aggregate or do its thing.

At the end of the iteration the value() function is called to get the result (this is what is stored/displayed).

Before the iteration starts the init() function is called to initialise any variables / running totals.

The relevant() function is called right at the start to make sure this metric is relavent for the workout.

So as an example, calculating Average Power:

relevant() {
Data contains "P"; ## must actually have power recorded
}

init () { total <- 0; count <-0 } # initialise our aggregating values

sample () {
if (POWER > 0) {
total <- total + POWER * RECINTSECS; ## aggregate JOULES
count <- count + RECINTSECS;
}
}

value () {
total / count; ## joules / seconds == watts
}

This may not give the expected results if you have pauses in the workout as average power is computed for the whole duration but average speed only when speed > 0, probably would be better to compute Average EI from scratch using the same formula at the sample level and accumulate just like the My Average Power example, or to use Time_Moving and Duration metrics to correct the pause effect.

---

Ale Martinez
www.amtriathlon.com

Hi Mark,

Thanks for all the help. I managed to get all the metrics working in GC, and it really takes using Stryd to the next level.

I now have trouble creating a trend graph with one of these custom metrics (Efficiency Index). I can easily create trend lines of other metrics (average HR, average cadence etc.) however for efficiency index the graph remains blank. GC does however recognise the metric; the axis are correctly labeled with their units. The value also appears on the summary page of each ride and works perfectly, it just doesn't appear in the graphs.

Do you have any idea what I could be missing? I have also repeated the steps done here but to no avail: http://www.bigdatatriathlon.com/...stryd-goldencheetah/