-
There was some stuff by Seiler that showed what you're proposing had great effect, going from longer to shorter intervals. I can't think were I saw it right now.
I like to think about it in total work done. Which is time duration and effort level. If you go to hard, you can't do so much, but go too low and it's not a potent stimulus.
I have found 4x10min to be good. It's a good amount of work (done 1 or 2x/week) and hard but not super hard. If I do 4-6x4 then i need more time to recover.
Also due to my genetics, I really do not need to work Vo2 directly, I can do 5.5L/min. So I focus on improving efficiency etc. The more time I can accrue riding my bike at around 70% MHR the better I will be. That said if I needed to repeat 3-5 min efforts, then I would work it, but I don't.
That said, I have given up most structured training now. The simple fact is it does not make much difference. Years back I followed a rigid programme, with all sorts. Last year I just did whatever I felt like. My power numbers were better than ever last year.
8x4 at that sort of power with 90sec recovery is solid work. But not over the top hard. Not surprising it works. It's not far of my 4x10 efforts.
Thankfully I have chosen a division of cycling that does not reward finding the final 5%. But rewards being stubborn.
Interesting.
My primary goal is to improve my threshold power because I judge this to be the best metric to fit with the races I predominantly do (a typical year will include a focus on a couple of half iron distance tris and a marathon, plus a smattering of 10-50mi TTs and some cx and xc(running) races which I do for either fun and/or an alternative to an interval session).
My training is largely polarised with 10-12hrs/week structured around 2 high intensity sessions (usually 1 run 1 bike, but can be 2 of either sport depending on proximate goals), with the remainder at low intensity. I also try to include a regular longer low intensity endurance session for both run and bike (2-3hr run and 3-4 hour ride). The remainder are all short easy intensity runs/rides, often commuting.
My run training is slightly different but for cycling, this year I am experimenting with starting with longer intervals and decreasing the interval duration and increasing the intensity throughout the season. Each block of intervals is conducted for about 2-3 months before moving onto the next block. This means I realistically only have 3 “types” of interval session: 2x20 (completed), 8x4 (starting) and then hard start 30/15 VO2Max intervals (https://sparecycles.blog/2019/03/13/comparing-30-15-vo2max-intervals/ which I’ll start around June).
I do the 8x4 intervals on the recommendation of Mark Jones (@mork on TTF). IIRC his original recommendation was to target 10mi TT effort (so ~105% FTP) with 90sec recoveries. When he recommended this session he cited this paper (http://www.tradewindsports.net/wp-content/uploads/2013/10/Stepto-99-Effects-of-different-HIIT-on-TT.pdf) which has very low sample size but Mark said he had considerable success using this interval session with the athletes he coaches. As it happens, now that I read the paper it actually recommends targeting 85% peak power. Although their ramp test protocol is slightly different to mine this would be more like 113% FTP.
Either way, my “interval protocol” is always to aim for the hardest intensity that I can consistently maintain for the target duration. Or in other words I’m aiming to maximise my average power over the total duration of the “work” while minimising the variability both within and between intervals. For the 2x20s this has usually meant starting my interval block at nearer 95% FTP but as the block goes on (a) I hopefully improve and (b) I mentally adjust to the intervals and so this rises to nearer 100% FTP. It’s a while since I did the 8x4min intervals but I recall being able to complete them at 105% FTP but found it hard… I don’t think I could have done a 9th rep.