Daily Suggestions / Do they 'work' ?

5 weeks before race time I would expect long runs around 2.5-3 hours assuming the average plan increases long run duration by 10-15mn a week (let's say a couple of miles).

The typical marathon plan is about 20 weeks. When did you start the Garmin racing plan?

I am wondering how Garmin fits a typical 20 weeks into fewer weeks with this new feature. Garmin Coach plans do not accept races that are too close, in particular longer races.

Let us know what you see in the next few weeks. Maybe the long runs will start scaling up fast.., but yeah, stick with your current plan :-)

The ACTUAL plan I'm following is adapted from the Garmin 16 week intermediate heart rate based running plan (except I run to power instead of hr zones). Its minutes on feet based, rather than distance based.

So yesterday I just did this 3hr effort, which as you say, is what would be expected.

- Run in Z2, easy pace, 135 minutes.
• Run in Z3, marathon pace, 45 minutes.

I can now see what its got for me next Sunday, which is, JUST a 10k race I have scheduled!!

In reality, following my plan, I'm running 10miles to the race, and then running the 10k at an easy 'marathon' pace, and getting the train back...to more or less match the below (the 90mins portion will prob be an extra 15mins for that easy opening 10miles) >>

• Run in Z2, easy pace, 90 minutes.
• Run in Z3, marathon pace, 60 minutes.

A better adaptive plan perhaps would advise me not to race a 10k, and instead offer up a suggestion for something longer, but I can understand why it would be limited in this way. Will be interesting to see what it offers up the following Sunday, as, that will be 3 weeks to go, and its not suggested any long run longer than 70mins yet !!

I wonder if 9.33 and race based adaptive plans on the watch maybe came along 'too close' to my October marathon and its struggling to adapt, assuming it doesn't have the ability to look backwards at what I'd done to date.

3673816 said:

I wonder if 9.33 and race based adaptive plans on the watch maybe came along 'too close' to my October marathon and its struggling to adapt, assuming it doesn't have the ability to look backwards at what I'd done to date.

If i get runs of 2h(max 2:14) for a half marathon then there is something wrong for you...maybe try to add another primary race in 3-4-5 months(no idea what's the ideal timeframe for a marathon) and see how(if) the suggestions chnage.

I am curious: how many weeks in advance of your race did you set up your race plan?

I am suspecting the race plans are struggling when the number of weeks is limited. 

I started this as soon as the feature was available in alpha, so a while back. I have followed it since then without fail and my base fitness was ok at the time with a 2 hour half-marathon time so nothing stellar but fit enough base to get me up to the ultra.

I never trained for longer runs than regular (half-)marathons, but for marathons, the typical plan is around 18-20weeks. I don't know how the algorithm squeezes 5 months into 2...

I ran a marathon in April so was just giving my half-marathon time as that was the last long race I ran.

I may have needed to give it more time but it would be good if they can add a meter to show how realistic it is to do X race in X time when you enter it in the calendar. Looking through the training plan suggested by the event, none of the training on that looks over the top for me so I feel that if I had just followed that rather than the Garmin suggestions I'd be better prepared now.

I read in this thread I think that it might not take elevation in to account for the training which seems crazy since it's such a big factor. The event type is marked as a Trail Run so you'd hope that it does take elevation in to account. Can anyone clarify?

If you attach a course to the race event, the race prediction will take into account the elevation profile of the course, as well as the weather the day of the race.

How it does it, I didn't find. If the race is a known race/course, I guess you can check the history of participants and see where the predictions would put you in past races. I admit I only run on road and pretty flat courses, so the impact of elevation or course design is negligible for me, and my race predictions are pretty much right on target.

The race prediction times (5k, 10k, half, full marathon) seem fairly accurate if I went at it with 100% effort on a flat course so I trust those enough to reference them but I'm not so convinced about the custom training for calendar events yet. I guess I'll know by this time next week!