This question from my brother, the Hulk, who has fathered a little Hulk--great news, but also a reason not to go long on the bike.
Can you get fast without spending a lot of time on the bike?
The answer: it depends. Really, you'll probably not get any faster. Getting faster is almost always a matter of riding more and riding harder. If you can't ride, there's no shortcut to speed.
Whether you know it or not, every time you get on the bike you're doing some kind of interval. The times you coast, the hills, the sprint points--every change in pace is a change in training focus.
The real question the Hulk is asking is this: what kind of intervals can I do if I've only got time for a few of them?
Short, Harder Intervals
Most studies say you should do intervals shorter than 3 minutes. These intervals don't just make you faster at intervals of 3 minutes or less; they also improve your 60-minute, aerobic output--more, in fact, than if you were to just spend the equivalent time working on your 60-minute output.
There are slews of studies showing that these kinds of short intervals are more effective than any other kind of training over 2-12 weeks.
Should you do short intervals, then?
If you're starting your workout program now, the answer is no.
Most studies are set up over a limited time (at most, 12 weeks) and do not use, as subjects, serious cyclists. They use college kids or random freds. They're interested in mere improvement from the center of the bell curve, not improvement from the edge of the curve to even further up the edge of the curve.
Here's the typical study. Notice the details: "moderately trained athletes" (i.e., freds), 8 weeks, four types of training, and...you guessed it...interval training wins!
Studies tend to say, "do high intensity intervals for 6 weeks, and you'll see improvements on the entire curve, from your aerobic to your anaerobic power." The intervals can be as short as 10 seconds (tabata-style) or as long as 3 minutes (VO2max-style).
You should follow a maximal, short-interval-based plan if you've only got 6-8 weeks, if you're a "moderately trained athlete," and if you don't have a lot of time.
You shouldn't necessarily use a VO2 or tabata interval-based plan if these things aren't true of you.
(1) gains tend to plateau after 8 weeks;
(2) exclusive use of interval work can lead to burnout and sickness;
(3) intervals are mentally fatiguing;
(4) intervals may improve short-term aerobic output but for longer events (i.e., road races) there is less evidence of their effectiveness.
Short, intense intervals can be a shortcut if you're really out of shape and don't have a lot of time to get better. They are the shortcut most coaches suggest.
Longer, Easier Intervals
On the other hand, training for most professional cyclists is mostly about longer, submaximal intervals. Most coaches, particularly Chris Carmichael, point to these kinds of intervals. Here's a sample of the bullshit:
What we learned during Lance's comeback [the one after cancer] was that the maximal efforts weren't as necessary as we thought. Longer, submaximal efforts at power outputs just below his lactate threshold elicited similar increases in lactate threshold power, and because the training intensity was lower, Lance could do more of this targeted training in a given week or month than he could using the older, harder method. Going a bit easier actually made him stronger and faster.It's easy to dismiss everything associated with Lance as bullshit, but there may be some truth to the idea that submaximal intervals yield more benefits than maximal intervals. Hunter Allen, Joel Friel, Ed Burke--the luminaries of coaching all recommend submaximal intervals, even to those with limited time. There's a good amount of scientific data suggesting the benefits of such intervals--especially for serious, long-term cyclists.
My own experience involves a mix and a movement between the two types of intervals. In early season, I don't do as many short, hard intervals. Later, when I'm trying to build my peak, I up the number of short, hard intervals. I try to hold back on the eyeball-popping intensity until I'm within a few weeks of my target events.
I know this doesn't address the Hulk's question; instead of telling him how to make the cake, I've just told him to use flour and eggs--not how much of either.
Different people respond differently to different programs, so I don't want to recommend how much of either type of interval to put in a training plan. I also don't want to limit long, submaximal intervals to the early part of the season or short, maximal intervals to the build and peak periods.
If you're like the Hulk and you seem to respond to long rides at submaximal tempo, you may want to try to simply shorten your long rides as you must (rather than substitute maximal and short intervals sessions).
The real challenge of creating an annual training plan is figuring out when to do what kind of intervals--not only what to do during particular daily workouts, but when to schedule blocks of intensity and blocks of rest--or, as the Hulk's case may be, time spent caring for a newborn.
The upside of having limited time on the bike is that, when you ride less than five hours a week, your marginal gains will be more than the guy who rides 20 hours a week--you're getting way more benefit per minute on the bike.