Big Data Training Plans?

4 watchers
Jun 2015
8:29pm, 9 Jun 2015
3770 posts
  •  
  • 0
Kieren
Just an idea and not an original one but I wondered if a big data driven training plan might ever be on the road map? Quite possibly a pipe dream.

I'll explain below what I mean:

According to the homepage, there are over 70,000 registered users on Fetch. Even if only a quarter of those users are logging runs, there is quite a large sample size. There is already the "Training Group" area of the website where you can say you want to run a distance at a specific time and it tells you information about typical training for other hitting that target. Training Groups are under the [Race]section of the website and the link below shows some info for a 25 minute 5K

http://www.fetcheveryone.com/training-group.php?distance=5.00km&time=25

With a large sample size of user data, if heart rate and GPS data is included it might even be possible to gauge what types of training provides better results. For example, it might turn out that Y-interval with X-rest is better for most people than some other variation.

The training logging could extrapolate how quickly people improve on any given plan, correlate the plateau busters etc

---------

Some of the polar watches and phone apps also use the heart rate monitor to measure heart rate variability as a way of determining if the athlete is over or under trained. Doing that daily will let you know if you can train hard of need to back off.

In theory, that could be part of a dynamic training plan. So in the morning, take your HRV and the training log will look at the workouts you need to do and tell you to go hard or easy depending on your recovery state.
May 2020
5:46pm, 15 May 2020
3928 posts
  •  
  • 0
Kieren
5 years later, it looks like "Train as One" does this now with machine learning / algorithms the call ai

trainasone.com

Probably a whole host of other tools too now that cost to entry is low.
Oct 2020
11:12pm, 16 Oct 2020
1715 posts
  •  
  • 0
Longwayround [LWR]
I've just started using trainasone, as a runner who gave up a few years ago due to injuries and lost mojo. So far, it's going very well. Thanks for the heads up.
Oct 2020
12:59pm, 27 Oct 2020
4034 posts
  •  
  • 0
Kieren
I hope it works for you. It would be great to get an update / review in this thread after a 12-26 week training block.

I used TAO for a short while but found they gave me everything a bit too easy. Maybe that is the best way for longevity but at the moment I prefer a cookie-cut plan from the Jack Daniels Running Formula book.

Just came back to this thread as there is an article on using big data to put together a marathon plan:

podiumrunner.com

.

Got something to say?

To join the discussion, sign in or join us.

About This Thread

Maintained by Kieren
Just an idea and not an original one but I wondered if a big data driven training plan might ever be...

Related Threads

  • training

Heart Rate Analysis

See your heart rate training data from the last 12 months broken down in technicolour.

Flat as a pancake?

We've collected thousands of your reviews from events, and worked out which races give you the best chance of a fast time!

UTMOST

Beat your league rivals to gain weekly promotion. Compete at every distance from a mile to a marathon.

Marathon Prediction

We delve deeper to give you greater insights when working out your goal marathon time.


Back To Top
X

Free training & racing tools for runners, cyclists, swimmers & walkers.

Fetcheveryone lets you analyse your training, find races, plot routes, chat in our forum, get advice, play games - and more! Nothing is behind a paywall, and it'll stay that way thanks to our awesome community!
Get Started
Click here to join 108,778 Fetchies!
Already a Fetchie? Sign in here