September Sprints results

I was skeptical that I could improve my short-interval anaerobic power very much in only 5 weeks, especially at the end of a season where I'd had a fair amount of unstructured short interval work in races and hard group rides. And my 5-week plan didn't include as many sprint intervals as I'd planned to do.

But as the graph below shows, just a few weeks of about two sprint interval sessions per week (8 x 15sec) have resulted in a pretty good uptick in my 30-second, 60-second, and 2-minute max power. I've thought about whether there is another reason for those improvements, but have no other excuse than the sprint intervals. As I've said before, I'm no sprinter, but improvement in those time intervals should help me stay closer to the front in bunch sprints.




Like they say, your body can't do what you don't train it to do, and I've always hated sprint intervals and not done many. I found that the more I did them, the less I disliked them. I think even one sprint interval session a week throughout the season would make a noticeable difference, and they can be done quickly before a longer ride so they don't mess with your schedule very much.

It's also interesting to note that my winter training plan includes L7 (10 to 15-sec sprint) intervals one day per week during the L1-L4 (base) training phase. According to the folks who study this stuff, you can do L7 intervals and if you recover completely between intervals, they won't negatively affect your aerobic adaptation becasue you aren't really stressing your anaerobic metabolism in such a short time, you're just stressing your musculoskeletal systems. But if you try to mix in L5 or L6 intervals during the base training period, your aerobic adaptation will be reduced, so avoid them if possible.

That tells me that the sprint intervals probably aren't helping me through better anaerobic energy system development, but through neurological adaptation. The muscle memory and nerve signal systems are improving, which is the same thing that happens when you start lifting weights after a long layoff and you get big gains in the first few weeks without any muscle growth.

New 2008 Training Schedule

I recently posted a training schedule for '08. Ignore it. This is the real one. I developed it after reading "The Road Cyclist's Guide to Training by Power" by Charles Howe with Dr. Andrew Coggan's input.

I read all kinds of stuff about training with power, but this is by far the most concise and applicable document that I've seen. And its free at http://www.freewebs.com/velodynamics2/rcgtp1.pdf

My trainin plan looks like this:

TT improvement

My experiment with a new TT position has to be considered a success. I feel just as powerful as before and I'm much more aero. I will do a final tweak this week by lowering my bars another cm and bringing my seat about 1-2 cm further forward, then I think I'll leave it alone. It's difficult to see my improvements without a power meter on the Cervelo, but I know they are there based on my recent rides.

Three times in the past 5 weeks I've done the Thursday Worlds course in Macon with the group ride:


Chasing or being chased helps tremendously with the motivation factor, which is critical for an hour and a half TT ride at full power. The first two times I averaged 21.9 and 22.1 mph. I skipped a week and last night averaged 23.2 mph. That's a huge improvement. Some of it is due to a shortened course the third time that did not include Moran Road, which has a long false flat. But I know it was not all due to the course change, because I was averaging about a mile per hour faster at the Montpellier Rd turn this week compared to the other two weeks.

Also very importantly, my back was not sore this time. The rearward saddle and low bars position I tried before put too much strain on it. The new position is much more comfortable. I could have ridden for another hour at the same pace, I think, if I'd had food. Long TT skills don't really help me in Cat4, though, because our TTs are usually only 10 to 25 miles long.

I've ordered a replacement for my slow Tufos (Vittoria Chrono EVO CS tubulars - hopefully with at Crr 0.0004), but they won't arrive in time to be mounted and dry properly for the State TT in Rome in about 10 days, so Tufos and a Crr of about 0.006 it is. I didn't need that 30 seconds anyway.

It's the TSB, stupid

I last posted wondering why my performance had improved in the last month and provided a few possible theories. Then I received a comment on my post from Stuart Lynn (http://tt-training.blogspot.com/). He suggested that the increases in performance may be due to a positive Total Stress Balance (TSB). One glance at my TSB trend over the summer (see my recent "2007 Training Load" post) almost proves that he's right. My TSB hovered around or below zero from March to mid-August, but has been between 5 and 15 for much of late August and September. That's mainly due to the ATL/CTL spike I got at RAGBRAI.

If I put any faith in Andrew Coggan's Performance Manger, then the the obvious answer is that the RAGBRAI effort has indeed paid dividends 4 to 8 weeks later. That means I'm crazy if I don't try to repeat that type of effort a few times a year at the appropriate times.

Thanks, Stuart, for helping me to see the obvious.

RR power and Crit power

Sometimes the more charts I have to study, the less each of them means to me. If I wanted to boil my long-term cycling training progress down to two numbers, they would be:

1. Functional Threshold Power (FTP) - sustainable power for an hour

2. Crit Normalized Power (CNP) - sustainable power for 35 minutes

The first is well-known. I just invented the second one.

Those two numbers are indicative of my sustainable power in road racing (FTP) and criterium racing (CNP). Here's a chart of how those numbers have increased for me over the past season:

Both numbers showed substantial improvement through the winter and early spring, then plateaued through the summer, and have started trending upwards again. My FTP is now at 280 and my CNP is at about 292. Why the late season increases? Here are some theories as to why the numbers might be going up again:

1. Cumulative effect of entire year of hard training (this one may be the most likely explanation)

2. Delayed effect of huge weekly effort at RAGBRAI in late July (8 weeks, that seems reasonable for adaptation to the effort?)

3. Effect of sprint training that I started in mid-August (unlikely, I think, but possibly a contributor).


Keeping the numbers headed northward is one thing. Next and just as importantly, I have to learn how to use the power more efficiently in races - staying at the front of crits, attacking at the right time in road races, working effectively for teammates, etc.

2007 Training Load

Here's my 2007 CTL chart. I started last November at about CTL=40. CTL is chronic training load, expressed as TTS/day. It's not an average, but a power function with a half-life of 42 days. (The yellow is total stress balance [TSB]. Check out the CTL spike and the TSB valley of doom that I endured for about a week after RAGBRAI in July.)




Anyway, you can see that it took me all winter last year with Peach Peloton rides on weekends (see the little spikes in the blue line from November to February that faded back during the following weeks) and riding the trainer and doing weight workouts on weekdays to get from 40 up to about 65 by the time the season got underway and I started working on some intensity.


This year I plan to use erg files, computrainer racing against others, and more riding during the week to prevent my CTL from dropping below about 70 or so. I'll also keep some intensity included in my weekday workouts. Hopefully, I'll be more prepared for intensity when next season's races start.


Also, I hope to maintain my current FTP of around 278 watts so that the Peach Peloton rides this year will give me an intensity factor of about 0.74 instead of the 0.85 that I got last year. The 0.74 should be more effective in giving me the endurance mileage that I need without requiring me to take so much time for recovery during the early part of each week. (Those numbers assume that my normalized power on Peach Peloton rides will be about 205 watts, like last year. If I work more at the front or if the rides are faster than last year, I might be looking at IF=0.85 again).

Tracking Functional Power

Functional Threshold Power (FTP) is the highest average power you can hold for a 1 hour steady effort. I performed a 1-hour road test for FTP in December 2006 and found it to be 240 watts. I tried again in April and got a result of 268 watts. My next sheduled test was August, but when I attempted the test, I couldn't get through it mentally - I stopped after 20 minutes. I guess it was just one of those days.

The difficulty of accurately measuring FTP on any given day got me thinking. I compared my FTP testing results with my 60-minute best normalized power (NP) for the same period and saw a very strong correlation. So I think I might make the formal FTP test a thing of the past and just monitor my 60 min NP to determin my FTP - it seems to be just as good or better of an indicator that one test on a single day and it uses data from hard training and racing, requiring no training-to-racing conversion factors. (With FTP testing in training, I always threw in a 1.03 factor to account for motivation of racing, etc., and I'm always a little wary of generic factors). So from now on I'll use racing data and keep the factors out of it.

FTP is a good indicator of road racing and TT fitness, but to get a better handle on my crit fitness, I'll also track what I'll call CTP, or Crit Threshold Power. It's nothing more then mean maximal 35-minute NP. It includes more of the anaerobic effort required for short crit races like we see in cat4 races, so it might not track exactly with FTP throughout a season.

As you can see below, my FTP climbed over the winter and early spring, levelling out at about 275 watts, which I think is accurate. My CTP pretty much mirrored FTP except for a jump in the last month, which I attribute to the sprint training I've started to do. I look forward to see where that goes with more work.

One last caveat: in order for this chart to accurately reflect my fitness, I need to make sure to do at least one or two hard 35-minute and 1-hour efforts in each two month period, even during the winter. Do crits exist in January?

October off? Nope.

Once upon a time, about a month ago, I had decided that I'd take October completely off the bike again like I did last year. Now I've changed my tune. The Sprints experiment is going well and I don't want to lose too much form heading into winter, so I've modified my plans to keep my TSS from dropping below 65 or 70 (it's about 85 now).

I'll do three workouts per week in October: Each week I'll ride the Tuesday Worlds in Macon, I'll do sprint intervals on Thursdays, and do a 4-hour endurance ride on Saturday or Sunday. In November, when the time changes, I'll begin Computrainer workouts during the week and ride with the Peach Peloton on Saturdays.

It will look like this through December:

A little explanation of the rides listed on the calendar:TSS is the Total Stress Score I expect for each ride.

CT 90-90-90 is a Computrainer ERG workout that I've created. After warmup, ride 20min @ 90% FTP, then do 2 x 5min @ 90% 5min max power, then 3 x 30sec @ 90% 30sec max power and cool down. Takes about an hour - TSS 83.

The sweet spot is 90% of FTP (250w for me). I'll do three 15-minute intervals at 250w with 7 minutes rest between intervals, then finish with three 20-second sprint intervals and a cool down (TSS 97).

Sprints are working?

I think my September Sprints experiment is working.

It makes sense that your body can only do what you have trained it to do, but I didn't fully appreciate it until I saw the numbers. After only two or three weeks of very unstructured sprint workouts (in some cases that means just throwing in a few sprints on my endurance rides), I can already see a difference. Yesterday I was doing a just tool around town kind of recovery ride. I decided to give one big effort up Persons Street to open my legs. I went hard, but didn't feel like I was hell-bent. My 30-second power from that spur-of-the-moment effort was 575 watts, 30 watts higher then my previous 30-second best power. That info nugget along with the data from my Grant Park crit on Saturday (best 40-min normalized power ever by 20 watts @ 295w).

Based on this preliminary anecdotal success, I'll continue the experiment into October to see where it takes me. I'll never be a sprinter, but I think this type of workout could really help me race crits better next year.

End of Season Assessment

I've completed my last race for 2007, so it's a good time to look back and see how the season went. I'll take the next few posts and look at several variables and how they changed from 2006 to 2007 and throughout the 2007 season.

First, I'll compare overall fitness. Two charts pretty much tell the story when it comes to my 2006 fitness vs my 2007 fitness. The first is Mean Maximal Power 2006/2007, shown here:




My Cycling Peaks WKO+ software searches the data from all of my power meter downloads throughout the season to compile this chart. The x axis is the time interval on a logarithmic scale, and the y axis is my best 2007 average power for each time interval. It's worth noting that my time trial data is not included because I don't yet have an Ergomo bottom bracket for the Cervelo.

It's pretty obvious that I've made huge gains since the 2006 season. My peak sprinting power has increased from about 800 watts to about 1000 watts. My 1-minute power has increased from about 300w to 460w, and my 1-hour power (Functional Threshold Power) has increased from about 210w to about 280w. Some of that is not reflected directly on the chart, but most of it is there.

Here is the same data presented using Normalized Power, which is essentially equivalent constant power (it needs at least 5 minutes of data to work, so there is no 0-5min data here):



My power has increased across the board, but the increase from about 240 watts to 300 watts normalized power for the 30-minute interval has probably made the biggest difference. That improvement has allowed me to stay in all my competitive division crits this year. With last year's power, I think I'd have been dropped in all of them. I held a normalized power of about 290 watts for 40 minutes in the Grant Park crit last weekend and still only barely was able to hang onto the end of a very hard charging front group. Maybe next year I can get that number to 310 or 320 so that I can contest those races instead of just hanging in there.

I'd say my 2007 improvements were due to:
30% - Peach Peloton rides last winter providing me a much better early-season base
30% - Tuesday Worlds hard rides making the longer high-intensity rides easier to do
20% - A fairly regular racing schedule (about 25 races over 7 months)
10% - Computrainer ERG workouts to maintain some intensity in early-season training
10% - Targeted training and self-analysis made possible by Ergomo power data and Cycling Peaks WKO+ software

TT Position - Phase II

Aero is important for going faster - that's obvious. But cycling past age 40 would be nice, too, and I'm 39. My new aero position is undoubtedly more aero than my old position, and therefore it's faster; but it puts more strain on my lower back than my old TT position did (and much more than my road position does) because my legs-torso angle is more acute. The last thing in the world I need is a bad back - that'll shut you down fast and for a long time. So I'm going to change again to save my back.

The reason my torso angle is so acute is that I lowered the bars and slid my saddle back at the same time when I developed my new position. I lowered the bars to flatten my back and drop my frontal area - that's standard. But I figured my power output would be higher if I slid my saddle back to something similar to my road position. I think it works - my power seems good. And moving the saddle back gave me an important unexpected benefit: it increased my reach so that my upper arms are at about a 45 degree angle with vertical. That lowers my shoulders and upper body more.

So I need to increase my legs-torso angle without going back to the higher shoulders-torso position. The first obvious thing to try is to slide the saddle back forward - maybe not as far as it used to be, but several inches, at least. That'll raise my shoulders because my upper arm angle will get back close to vertical. To lower my shoulders back down, I guess I'll take spacers out and lower the bars again.

One thing that I have learned in messing with this stuff is how cool Levi and Floyd's positions are on their TT bikes. I mentioned that I changed my upper arm angle away from vertical and lowered my shoulders by stretching out more. They have taken the exact opposite approach and changed their upper arm angles by bringing their bars in closer to the saddle, giving them the 'mantis' look. I'd love to try that, but I'd need a smaller TT bike frame - probably not going to happen. Before I move my position again, I'll photograph it and compare it to my old one so I can see how the angles are changing.