Here is a link to the testing data and training programs which provides info on Eliud Too's preparation for the Dublin City Marathon. James and myself (mostly James) have been working hard over the last two days to put together all the information in one location.
Essentially the way the training programme is designed is that the athlete gets a lab test done at the start of the training block. This provides data on maximal aerobic capacity, but more importantly heart rates and speeds which correspond to specific blood lactate concentrations. All training intensities are primarily based off heart rates. James uses pace and speed simply as a guideline for attaining steady state heart rates. No pressure is ever put on the runner to achieve a set time. The perfect session is when heart rate was smack on target, regardless of the split times. A training adaptation is defined as an increase in pace for no change in steady state heart rate. These adaptations are very small, take months to develop, but over the course of 4 years can add up to significant improvements in performance.
From a periodization perspective, we use a block periodization model, similar to what Issurin first proposed back in the mid 90's. A full training block is usually a series of 5-6 x mesocycles. Each mesocycle is 4 weeks long and will have a different focus, based on how far out from the race we are.
So the 5 x monthly cycles for Dublin were divided into:
2 x aerobic development (June and July)
1 x transition (August)
1 x VO2max/Lactate tolerance (September)
1 x Speed/Alactate (October)
Feel free to post any questions or comments. I'm sure James will fill in more of the specific details, as he is the main guy devising and implementing the training on a daily and weekly basis.