Training Ride Efficiency

FWIW, I have just completed a review of my 2006 training rides through mid-October (181 ride files and ~340 hours) from the perspective of ride efficiency. What do I mean by training ride efficiency? I am primarily focused on L4-L7 efforts (the mix varies, but my primary ride goal always encompasses one or more of these levels), so I measure training ride efficiency as total minutes at L4-L7 divided by total minutes on the bike. I then summarize my ride time by week. Attached is the frequency distribution of my training efficiency index by week for this year. BTW, these are real minutes by level (e.g., L4 is >= 10mins duration and >=91%FTP) and not simply a frequency distribution of the watts data series.

My basic observation is that even though I work very hard at being efficient in the use of my training time, I can make still further progress. My overall ride efficiency index for the year is 44.5%, but some of my weeks were as low as 10-20%. My trainer rides are my most efficient rides, hands down. Even when I go out for a long L4 (I have the perfect route just 5mins from my front door -- 15 miles long, 1.6% avg upgrade, typically upwind), it's only a little more than 50% efficient due to the long descent back home. I think I can push my road rides up to ~50% efficient and my trainer rides are always >75% efficient. So, I think I can milk at least 10% more efficiency out of my rides. 10% is a pretty big deal when applied to >20,000 minutes year-to-date.

Attached Thumbnails

 

So how "efficient" are your workouts from a neuromuscular perspective? That is, do you think that just riding the trainer all the time (which would be the easiest way to increase training efficiency as you've defined it) is adequate/optimal preparation for competition out on the road?

Well, it's purely speculative since I still do the majority of my riding on the road. And, for a variety of reasons I wouldn't want to do all of my rides on the trainer. But, I will speculate that my racing performance would suffer if I did virtually all of my riding on the trainer, largely because I think my AWC would detrain. And, of course, the road rides provide the constant variations and high VI index that is so common in races. But, studying my rides from an efficiency perspective has allowed me to rethink my priorities for my road rides. I will be doing more L5-L7 efforts on my road rides since I can get most of the L4 time I want on my trainer. It's really the L5-L7 efforts that build my AWC and my confidence in both attacking and countering attacks in races. This will result in a lower ride efficiency index, but it's a tradeoff I'm willing to make. Also, most of my routes lend themselves to L5-L7 efforts.

Your various posts on this subject have inspired me to pay more attention to how much time I spend at or above L4 but I don't really understand how this works when applied to long rides.

I find it hard to believe that most cyclists with an accurately measured FTP would be able to routinely spend anywhere near 50% of a 4-5 hour ride at levels L4-L7. So you might argue that rides of that duration aren't necessary but my two biggest events next year will both be 5+ hours in length.

I'll say this: if you can do a 4-hour ride with 2+ hours at or above L4 then either your FTP is incorrect or you've reached the point where your training is no longer improving your FTP which means you're wasting your time.

--Steve

I think it would be difficult to spend near 50% of a 4-5 hour ride at L4+. I almost never do solo rides >3 hours, so when I look at my own ride files >4 hrs I'm basically looking at group rides with a lot of short pulls and few (if any) sustained efforts >91%FTP. These tend to be closer to 25% ride efficiency and, in fact, prior observation of these poor efficiency numbers on group rides has prompted me to comment on (most) group rides as being of limited value from an adaptation perspective.

My broader point is that I think I work as hard as anybody I know at riding as much as possible at L4 and above and I still have a total ratio less than 50%. I would be willing to bet that if everybody knew these numbers (they aren't easily obtained) that 45% would be among the highest ratios (unless someone rides predominantly on a trainer). The reason this is a key topic is because we all have limited training time (even the pros) and how efficiently we use that time has a lot to do with the physiological benefits and, consequently, race performance. I put my numbers down because it's difficult to discuss such subjects without specific data.

How would you define ride efficiency for a stage racer, an Ironman triathlete or alike?

I've been closely involved in training/racing aspects of long distance athletes over many years. In other disciplines though.

For an Ironman athlete, running imposes its dose of low level training. Therefore stating that these guys should aim at training at L4 + as much as possible would be close to ridicule in my opinion.

For a marathon swimmer, in the '90s, some volume would be done at the lower end of L4, most volume would be done at L3/2. The results have always been good. My best result was a second position in a world cup event. Best result of one swimming training with my boss was a victory in a world cup event (only five Canadians had won this event over 44 years).

That question has been bugging me for a while now. What is considered ride efficiency for a long distance cyclist?

To follow up with Solar's question...

I'm training for a one-day event of just over 100 miles. For the first 72 miles the lead pack stays together (only 4500 feet of climbing) but in the last 30 miles the pack disintegrates due to approximately 6000 feet of climbing. The winner (typically a domestic pro) will finish in about 5 hours with the next 50 finishers being spread out over the next hour.

How do I train for such an event and how do we define efficiency in this case?

Increasing my FTP with L4-L7 work is important but surely I need to spend at LEAST one day a week on a long ride and I need to do hill work. I also need to try to train myself to be able to finish the last 1-2 hours of my long ride as close to L3-L4 as possible.

--Steve

I am using the term "efficiency" very simplistically -- time spent doing efforts of the greatest value in terms of physiological adaptation divided by total training time. A more sophisiticated measure might be the TSS of high-intensity efforts divided by the total ride TSS, a measure of how efficiently I have allocated total training stress to efforts of the greatest value. So, yes, I am drawing a line at L4 and I am saying that my highest priority is time spent doing L4-L7 efforts. For a different training objective (e.g., longer races or rides), one might define the measure differently (e.g., L3+ time or TSS divided by total time or TSS). To compute such a measure only requires that one defines his highest priorites in terms of adaptations and the efforts that will result in the desired adaptations. How much of one's training time is he spending at these levels? It all comes down to maximizing the benefit of a scarce resource, in this case training time.

I think SolarEnergy hits on a very important question. I guess you could also ask, What are you training for? Maybe this is the scale you're essentially hinting at, that there's some optimal mix of high intensity work that needs to be done compared with overall work based on your target event. So maybe the inverse relation goes something like...

1 hour: 60%
2 hour: 50%
3 hour: 40%
4 hour: 30%
5 hour: 20%
6 hour: 10%

Of course this is all just made up. It also touches on the standard caveat, "What works for me may or may not work for you." Maybe that's another way of saying you're training for a 2 hour event and I'm training for a 3 hour event.

I also wonder about L4. Didn't I read acoggan suggest that L2-L4 are really different flavors of aerobic training? Given that, does L4 count in this high intensity mix? Likewise, does the 10% increase mark an unmaintainable workload over those 20,000 minutes?

Interesting discussion which brings up a lot of questions.

I will be training for such an event (longer, actually) next year (Everest Challenge), and I haven't gotten to the point of breaking down a complete month or week but I have given some thought to total volume and allocation by level. I think I would still target L4-L7 efforts, with a heavy emphasis on low-end L4 efforts at extended durations (e.g., 40-80 mins). This would necessarily compromise my available time at L5-L6, but I could probably handle ~15 hours a week with a minimum of 6 hours/week of L4. I would get a lot of my L4 time on the trainer (~40mins/day) and would use my road rides for L5-L6 and for long rides at net L3. My overall target would be ~50% L4-L7, but the trainer rides would make that fairly easy to attain.

Yes, absolutely any training plan should prioritize the adaptations of greatest value to a cyclist for his target events. And, this is going to be specific to the cyclist because we each have at any point in time a different MP/duration curve with different degrees of potential for increasing any specific energy system. For example, my 2006 data reflect a year-long emphasis on FTP at the expense of VO2MAX and AWC because I deemed my FTP to be my weak link. Next year will probably be more balanced because I will be carrying a respectable FTP into the beginning of the year.

There are different flavors of aerobic training, but L4 is the flavor of choice.

I don't think so. My TSS scores were ~950-1000 for much of the summer but I had one week at ~1300 and didn't have any ill effects. Raising my ride efficiency index by 10% will not result in a 10% increase in TSS (because it is a conversion of time from L1-L3 to L4-L7).

Well, that was my intent. I may be wrong, but I think many of us are using our time in the saddle less efficiently than is possible.

Assuming you get enough consecutive minutes at your L4 interval and you just happen to take more time resting (or waiting at a stop light, or talking on the phone, etc) did you really get a less effective workout?

Say you gain 10% more efficiency on your road rides, won't that just equate to shorter rides? or maybe harder breaking and harder sprinting to maximize your time in the "zone."

Nobody has unlimited time to waste, but many mornings I have done a ride and been back at home thinking "I'm just sitting here now doing nothing, what difference would it have made if I took 5 min between my L4 intervals or 15 min. to go explore that one street that I've been wanting to check out." This is especially more evident in the evening when I could have spent 1 hour on the trainer and 1 hour doing nothing on the internet, or 1:30 on the trainer and only 30 min doing nothing on the internet.

Do you equate being more efficient with being more effective?

This is true. I'm curious to what your average IF is for 2006. Something in the realm of .91?

You're not wrong. Since my last event I've been on the mountain bike once a week and the trainer 4 times a week. There's no question the trainer is dramatically more efficient than spending the time on the road. Even something like a light L2 warmup is more than coasting down a hill.

Of course, given that my events next year are endurance, it surely begs the question of just how efficient I need to be. But then, there's a case to be made for segmenting my weekly training into adaptation rides (L4+) and endurance riding (L2/3), then asking the question, how efficient are my adaptation rides?

Not to me it doesn't. What matters (to me) is my total minutes and TSS of high-intensity efforts. But, my training time is limited (as it is for most people), so I care about getting the most out of my available time. But, physiologically I don't think it matters whether you get 40mins of L4 efforts in 50mins (highly efficient) or in 4 hours (highly inefficient).

Not really. I haven't finished studying my ride files that were highly inefficient, but I think all I have to do is to eliminate the rides that are in the 20s. I'm not going to try to increase the ratio for all of my rides, only the worst of the bunch. And, if my least efficient rides are all group rides, they may stay in the mix because they serve a different purpose than purely adaptation.

If you mean more effective at realizing my targeted adaptations, then the answer is yes. Now, there would be a tradeoff if my highest priority adaptation was L6-L7 due to the higher recovery/high-intensity effort ratio. If I were a sprint specialist, I would probably be happy with an efficiency number in the 30s.

Offhand, I don't know, but I'll run the number. I actually compute it, but use it only internally to compute TSS. Maybe I'll output it.