Welcome To Fetcheveryone

Our awesome training log doesn't hide its best features behind a paywall. Search thousands of events, get advice, play games, measure routes, and more! Join our friendly community of runners, cyclists, and swimmers.
Click here to get started
Already a Fetchie? Sign in here

New Training Log Prototype

17 watchers
May 2018
3:04pm, 14 May 2018
4749 posts
  • 0
larkim
Just looking at info after a long (!) run at the weekend, and I'm confused how the benchmarks and predictions work together (if at all).

See the image below. In the "benchmarks", my fastest benchmark gets a WAVA of 59.x%, but as it was a long run there is a range. On the "predictions" there is a reference to it seeing 56.41% as the best WAVA, but that is slower (ironically for the same distance - 5km) than I actually covered that day. So the result is that it "predicts" I'd run a marathon in a slower time than I actually did and they don't seem to correlate much.

Just curious how it all works, tbh, unless I'm seeing something that is a gremlin (in which case I take credit for spotting it ;-) )

May 2018
3:30pm, 14 May 2018
19345 posts
  • 0
fetcheveryone
The predictions are based entirely off the total time and distance for your run - in this case, 33 miles in 5hrs (nice one). They are based on the ubiquitous Riegel method of calculating race times at different distances. They are most likely to be useful if there are no other factors come into play e.g. if the course is flat. As I've touched on in my marathon prediction article, the standard Riegel exponent falls away for most normal people at longer distances - so a 5k prediction off a 33 mile race should be taken with a lot of salt.

On the other hand, the benchmarks are based on the fastest portions of your run. So within that 33 miles, at some point you ran a 4:00:06 marathon, and your fastest mile was a 7:33. The benchmarks are likely to be a bit more all over the place, particularly if (as in this case) the course profile offers some opportunities to go "wheeeeeeeeee" for several miles.

The two systems don't intermingle at all, although I've often thought about plotting them in the same table / graph.

Does that clear up a few mysteries?
May 2018
3:54pm, 14 May 2018
4751 posts
  • 0
larkim
Yes, that clears it up!

I suppose it starts off "odd" with the "Best WAVA" bit in the Predictions heading giving me the impression that it was predicting from that best WAVA, but what you're saying is that that Best Wava figure is just reading from the table once the whole thing has been "Regieled" from my total time?

This run was odd because of the nature of the trails I didn't lose as much between 26 and 33 miles as you'd normally expect, so the mara prediction is more or less bang on with the mara benchmark, which is what made me query it.
May 2018
10:01am, 25 May 2018
878 posts
  • 0
Oranj
Does the new log not like capitalization for tags? I always enter mine with a capital letter e.g. "Intervals", but when I come back the next day it's changed to "intervals" (see fetcheveryone.com/t-15006956 ).
May 2018
10:03am, 25 May 2018
2270 posts
  • 0
StuH
Yep, I've noticed that. Doesn't like spaces or symbols either.
Jun 2018
8:54pm, 2 Jun 2018
892 posts
  • 0
Oranj
I like how clicking on the fetchpoint icon in training now means that it's always switched on, rather than having to refresh it as you move about on the training view map

Got something to say?

To contribute to the discussion, you need to either sign in or register as a user.

About This Thread

Maintained by fetcheveryone
Try the new training prototype.

Vote in the poll: fetcheveryone.com/polls-view.php?id=6179

Check my blog for more updates / history: fetcheveryone.com/blog/3

Related Threads

splits graph pace log line default split mile distance view change
Back To Top

Close