Here's something I posted over on Gordoworld that discusses a similar issue:
As regards the easy,steady,tempo,hard,very hard progression. I think there's a lot of truth in that, however, IMO one of the reasons tempo nukes novices and causes so many recovery problems is that they are told to do it just ABOVE FT/LT. In RW parlance, tempo = 10M-10K pace, and the same seems to be the case in many traditional single-sport coach's minds. What I'd propose would be better termed FAST-STEADY (a term I heard from UK TT champ Michael Hutchinson and Brighton Uni-based physiologist Jamie Pringle). This pace is significantly easier than Tempo and with the right training, as I say, you can lay down a lot of that in a week (particularly on the bike) without being too nuked (See * below).
TG76: I sort of agree with what you're saying. Certainly as you work harder you have to recover more, and certainly at super-LT1/AeT the increase in glycogen use has a role to play. However I'm wary of such a blanket statement as 'above AeT...' because the exact levels of glycogen used will be dependent on many factors (Cadence, individual's Muscle-fibre distribution, and possibly also things like hydration and blood sugar levels).
As regards FTII fiber recruitment, again, I think you're oversimplifying a bit. FTII can be 'forced' into play early by a low cadence, which is basically what Big Gear work does, and a lot of IM triathletes do a lot of Big Gear work (some do nothing else - we certainly tend to have much lower cadences than pure roadies).
Finally, (*) Although our goal for IM is to be able to produce the best possible performance at Aet, I actually think that training just sub-FT/LT is the best way to produce this. This goes back to real-world issues like time constraints and the 'bang for buck' thing in my post above. The problem with Aet is that I think it often serves more to make you knackered than it does to really boost your fitness as fully as it might. (I'd be willing to bet that 4hrs of Aet would be more knackering and have less effect on your fitness than 4hrs slow pace but with 4x20 mins of Fast-Steady inserted into the ride.)
I don't know to what extent you could take a novice and build their training this way from scratch - I've never coached a REAL novice athlete at all (they've all had background fitness that needs to be considered). What I am satisfied with can be summarised thus:
1) After an end of season break, it is possible to build a Base training programme by moving fairly quickly from, say, one cycle of Progressive volume increase at slow paces to a series of cycles of progressive increases, not of volume but of proportion of Fast-Steady work.
2) This training is more time efficient than JRA or AeT in terms of training effect.
3) This training is not more expensive in recovery terms than slow OR AeT when related to training effect by volume, and so it seems probable that relative IM novices SHOULD be able to train this way without breaking down unnaturally quickly.
4) Fast-Steady (or sub-FT/LT or Max Aerobic) pace has the added benefit of enabling you to race better at faster paces, too (Important because I rarely see IM athletes who wouldn't ALSO like to race well over shorter distances).
Take a triathlete called Bob. Bob is a good runner. His Endurance/slow pace is 8-min/mile. His Aet pace is 7-min/mile and his Sub-LT/FT pace is 6-min per mile. He's going for a 10 mile run.
He can run 80 mins, 70 at AeT or 60 at Sub-LT/FT.
If he does AeT, he has 10 more minutes of recovery available to him in life.
If he does Sub-LT/FT, he has 20 more minutes of recovery available to him in life.
Obviously the running example is compromised by impact stress, but you get the idea. I'd imagine that, even factoring in artificial extra recovery minutes for the harder efforts we'd find that, UP TO FT/LT, the actual recovery cost linked to time would make the FS run the most effective method of training.
Of course it's not quite this simple. I've just also had this thought as a further refinement:
The training effect and increased recovery cost might be factored in as follows (the higher the score the more worthwhile the session):
The 80min run is worth a training effect of +1 because it increases repeatable endurance/efficiency. So a +1 total
The 70min AeT run is worth a training effect of +1 because it increases repeatable endurance/efficiency, and a second +1 because it involves a higher pace, leads to better glycogen use/storage, switches on the Lactic energy system. So a +2 total.
The 60min sub-FT/LT run is worth a training effect of +1 because it increases repeatable endurance/efficiency, and a second +1 because it involves a higher pace, leads to better glycogen use/storage, switches on the Lactic energy system, and a third +1 for coming close to the maximum level of aerobic training, increased training of lactic clearance systems. So a +3 total.
However there's also an effort cost to factor in:
The effort cost of the 80 is 0. it is the base line.
The effort cost of the 70 is -1. it is the next step up.
The effort cost of the 80 is -2. it is a step up again.
(FWIW, I'd imagine the effort cost of the next step, crossing LT/FT, would be a -4 or -5 because you're taking the system into a new, and more stressful situation of lactic imbalance. Why isn't the effort cost of the 80 -1? Because the scale has to start with an acceptable cost as O, not a no cost. Desirable training effect means doing something must happen, so the lowest point on the scale must be a training action.)
Finally we add in the time-efficiency cost and additional rest/recover time factor:
The time-efficiency score of the 80 is 0. it is the base line.
The time-efficiency score of the 70 is +1. It saved us 10 mins of time.
The time-efficiency score of the 60 is +2. It saved us 20 mins of time.
The final 'training value' scores:
80min endurance run = +1
70min AeT run = +2
60min sub-LT/FT run = +3
I realise it's simplified and not based in science, but it helps explain the thought processes that underpin my statements. I'd imagine that, if these hypotheses were given scientifically determined values, that even if a continuous AeT session and an endurance session containing sub-LT/FT sections went herad to head, the sub-LT/FT session would 'win' in the overall value stakes.
Stuff I like:
PBscience Triathlon Coaching and Lab Testing