Saddle Height/Rotor Q Rings

Saddle Height

I've finished a few rides now at a lower saddle height. My old saddle height of about 36.5" was lowered to about 35.5". At first I tried it at 35" because that corresponded with most of the studies I've read, but that felt too low. The 35.5" feels ok now and I think it's reduced my movement in the saddle.

A final note on the subject: A couple of the studies I read (in High-tec cycling and somewhere else) concluded that higher saddle height had the direct result of higher power output. The studies recommended raising the saddle until limitations of injury or discomfort kicked in (I had neither - I just looked like my saddle was too high).

Q Rings

Although I've been somewhat interested in non-circular chainrings for about 20 years, it has always been just a curiosity until now. I first was exposed to cycling in college when I picked up a hand-me-down Schwinn bike from a friend for $50 so I could get to class. It had non-circular chainrings. They were probably Shimano BioPace, but I don't remember.

In my recent research about saddle height, I saw lots of articles about non-circular chainrings. I say non-circular because there are lots of variations:

1. Shimano BioPace (introduced in the 1980s) were not circular, but were not elliptical either. They had varying, non-perpendicular radii. The largest radius occurred near at the dead spots in the pedaling stroke, 12 o'clock and 6 o'clock. That doesn't make sense to me. Why would you want the torque to be highest where you had the least muscle power to turn the pedals??? They look like this:

2. O-Symmetric rings are available now and have been used by a few pros in time trials. I don't know much about them, but they are not elliptical (they look sort of square, as you can see in this photo of Bobby Julich's TT bike:

At least they appear to have the larger radius at the 3 o'clock and 9 o'clock 'power' positions to take advantage of the pedal downstroke.

3. The studies I read varied in their love of non-circular chainrings, but the one that looked most promising to me was the Rotor System, which isn't really a non-circular chainring at all. Its a complicated crankarm gearing system that accelerates the pedal through the dead spots and takes more advantage of the downstroke. It uses circular chainrings, but your pedals are not always in a line, or 180 degrees from each other. When your right leg is at the bottom (6 o'clock), your left leg is a little past the top (at about 1 o'clock). The system weighs about a pound more than a conventional crankset and is pretty expensive ($700 to $1000 depending on the model). It looks like this:

Now, even though I'm open to the idea of a non-conventional chainring system, I'm not about to spring for $1000 for a system that, if I loved it, I'd have to buy 4 times. (For a road bike, a TT bike, a Computrainer bike, and a tandem). Because unlike the other systems I'm describing, I think this one takes some getting used to.

But Rotor realized that too, so they recently introduced Rotor Q-Rings. They are an elliptical chainring that has the larger radius occurring when the pedals are at the power positions (3 and 9 o'clock). They are designed to approximate the effect of the Rotor system, and the elliptical shape is supposed to be smoother than the asymetrical shapes of the BioPace and O-Symmetric rings. I just can't figure out why this is the 3rd or 4th generation of non-circular chainrings. I don't know much, but this is the first thing I would have tried. They look like this:

Any way, I was intrigued by the idea, they aren't terribly expensive (about $200 new), they don't add weight, and it's winter experimentation season, so I bought a set. The big ring is called a 53, but the radius varies from an effective 56-tooth on the downstroke to an effective 51-tooth at the top and bottom of the stroke. The small ring is called a 42, but it's 44 in the long direction and 40 in the short direction.

I had to raise my front derailleur about 1cm to clear the long axis of the big ring, and now there are maybe a couple of centimeters of space between the derailleur and the big ring when the small axis is up. In training I had no shifting issues at all, but the first time I tried shifting to the big ring on a group ride this weekend, the chain momentarily jumped off the rings. It worked fine after the first time. It might need a little adjustment and I might need to get the feel of it.

I had already decided to go to a bigger rear gear (I ride a 23 now in training and was going to go to a 25). But with the bigger front rings, I've actually gone to a slightly smaller gear ratio even after adding the 25. So I guess I need a 26 now to get the ratio I want for training.

I'll have to admit that the rings felt great the first couple of times out, and I felt very strong on Saturday's group ride -- but that could be due to other factors. After a couple of weeks of acclimation and adjustment, I'll do a couple of time trials to find out if there is any benefit. If I get something between 4 and 9 percent power increase as shown in the research, I'll be amazed. But you never know. It may be one of those things that are beneficial to some people, but not to others, which can make the research inconclusive. A 5% increase in power would be about 12 watts for me. That's a number that will be noticeable in my testing.

Musical Interlude

As a momentary antidote to the analysis paralysis that normally clutters up this space, here's a sample of the tunes on my Rio Cali that help pass the basement Computrainer time:

• My Chemical Romance - Welcome to the Black Parade
• The Racontours - Steady, As She Goes
• Radiohead - Creep
• Pearl Jam - Black
• Dave Matthews - Two Step
• Red Hot Chili Peppers - Californication
• The Killers - All These Things That I've Done
• The White Stripes - Seven Nation Army
• Rush - Closer to the Heart
• Hoodoo Gurus - Bittersweet
• Tracy Bonham - Mother, Mother
• Nirvana - Lithium
• The Rolling Stones - Gimme Shelter
• Yes - Changes

20-min FTP test

Several references I've seen say that a 20-minute time trial in training should give you the equivalent average power as a 1-hour time trial in competition. And a 1-hour competition TT, which is functional threshold power (FTP), should result in about 103% of the average power you'd see in a 1-hour training TT. I think yesterday I proved those relationships to be true.

My 1-hr, training TT average power on 12/2/06 was 231 watts, which yeilds a FTP of 231 x 1.03=238 watts.

Yesterday, 12/19/06, my 20-minute training TT average power was 240 watts. That's almost a perfect match, and it may be an exact match if I've increased my FTP by 2 watts in the past 2 weeks.

It's good to know that some of these calculations actually work.

Saddle Height

I’ve been told many times by the rider behind me in a pace line that I rock too much in the saddle. If I’d only heard this once or twice I might dismiss it as a bad day, because I don’t feel like I move around too much. But I obviously do, because I get the comment more than infrequently and it's usually from folks who know a little bit about riding a bike (translation: they’re faster than I am). If I thought efficiency wasn't involved, I wouldn't waste my time trying worrying about it, but methinks it does makes a difference, maybe a significant one.

I’ve experimented with saddle height before, I've read many opinions about how do determine it, and I've been fitted on a bike once or twice. But I think my saddle height sort of migrates higher over time because I’m accustomed to it more than anything else. So I have decided to take a more scientific approach and see what happens.

My saddle height has been about 36.5” from saddle top to center of pedal axle, which is shown as length 1 in the figure below.

Saddle Height is dimension 1 Trochanter height is 76 in this figure

Here are the studies that I’ve found and the corresponding saddle height that would be recommended for me:
35.3” Hamley & Thomas, 1967 – 1.0 x (trochanter height)
34.9” Hamley & Thomas, 1967 – 1.09 x (inseam height)
35.3” Shennum & DeVries, 1975 – 1.0 x (trochanter height)
35.3” Nordeen & Cavanagh, 1975 – 1.0 x (trochanter height)
37.1” Nordeen & Snyder, 1977 – 1.05 x (trochanter height)
??.?” Genzling, 1978 – 25 to 30 degree knee bend
35.4” LeMond, 1987 – 0.883 x (inseam height) – 3mm

Some of these studies are a little vague about whether the measurements are including shoes, etc. It’s obvious from the dates of the above studies that saddle height research is no longer a hot research topic, but it has been researched a lot – the above list is just a sample of the available studies.

So it looks like I need to lower my saddle about an inch to 35.3 inches. I plan to try that and see if I feel more stable on the saddle and see if I notice any differences in power, fatigue, etc.

It’s interesting to note that based on the above guidelines, my wife, Betty Jean, needs to lower her saddle almost 2 inches! I think I’ll recommend that she lower it about an inch and see what happens. An inch is an awful lot in saddle height. And even thought we probably should lower it gradually, we’re going to go whole hog and see what happens.

I might find out that I was already at my optimal saddle height and that I'm just a Jim Furyk of cycling, but I doubt it.

The Chicken or the Egg?

I've recently been discussing lactate threshold. I sometimes tend to overanalyze or mathematically dissect and issue to death because that's part of the allure to me. But I also think that anyone who would like to be a stronger rider can ignore the more simplistic approach to training at their own peril. Eddie Merckx's advice might be the best two word training advice ever: "Ride lots."

I've always wondered if the strongest riders have been riding many years because they are good at it (most people naturally enjoy activities that they are good at) or if they are good at it because they have been riding for many years. This post from the Google wattage forum gives this less gifted athlete a little hope that the latter argument holds true:

Intervals do well improving ones fitness in the short term, but they lead to burnout. Long term improvements comes from chronic long term training specific to the sport of interest. In order to do that, it's best to keep what ever cycling training you're doing enjoyable. Keep a long term focus, without obsessing on the details and you'll do fine.

This is exemplified by a study of 14 competitive cyclists with nearly identical VO2 max values that differed substantially in their lactate threshold determined during cycling (ranging between 61 and 86% of VO2 max). When the cyclists were divided into a "low" and "high" LT groups (66% vs. 81% of maximal oxygen consumption), it was found that the two groups differed considerable in the years of cycling training (2.7 compared to 5.1 years on average). However, they did not differ in years of endurance training (7-8 years of running, rowing etc.) When the low cycling LT and high cycling LT groups performed a lactate threshold test while running on a treadmill, the two groups were no longer different. Measured while running, the lactate threshold in both groups averaged over 80% of VO2 max.

In other words it wasn't genetics, but the length of chronic training and competition that made the difference. What's most important is that it be enjoyable so that you'll stick with it and make it into the second group.

My VO2max is about 65 mmole/l/min, which is pretty decent. Maybe that means my LT will gradually rise with time. And if not, it'll be a good long term experiment anyway.

TSB for personal bests

This chart was posted by Andrew Coggan. It shows the Training Stress Balance (TSB) values where the most personal bests are recorded. Looks like TSB of about 10 is a good target for best performance.

Performance Manager

Cycling Peaks software has a new feature developed by Dr. Coggan that's pretty interesting. Finding good cycling form has always been more art than science, and maybe it still is, but his Performance Manager, which was incorporated into the last update of Cycling Peaks WKO+ makes an effort to determine form scientifically if you load the power data file for every single workout that you do into the software. Here are the basics:

1. For each ride, the software computes your Normalized Power (NP), which is the steady power equivalent of the entire workout. This number, which is usually higher than your average power is calculated using a rolling 30-second average and some fourth-power polynomials to approximate your body's response to irregular power output.

2. Your intensity factor (IF) is just NP/FTP (Your Functional Threshold Power [FTP] is your steady 1-hour TT max power). An intensity factor of 0.7 means your average effort was about 70% of threshold.

3. The software computes a Total Stress Score (TSS) for each workout. TSS is (time in hours)*(IF^2)*100. So an easy one and a half hour ride might be TSS=(1.5)*(0.7*0.7)*100=73.5.

4. Performance manager defines your Acute Training Load (ATL) as the average TSS per day over the past 14 days, and it defines your Chronic Training Load (CTL) as your average TSS per day over the past 6 weeks.

5. Finally, Total Stress Balance (TSB) is designed to represent your cycling form on any given day. It is calculated: TSS=ATL-CTL. Theoretically, the higher your TSB, the better your form, and you'll normally feel strongest when your TSB is a positive number.

See my recent Performance Manager chart below, and my 2005 chart below that. The red line is ATL, the blue line is CTL, and the yellow bars are TSB. Click on the charts to get a legible version.

November 5, 2006 to present:

January 1, 2005 to current:

Double-deflection Lactate Threshold

After combining information I've obtained from reading, from lab tests and discussions with Tony Myers at ATS in Atlanta, and with Richard Wharton, Online Bike Coach, in Dallas, I've determined that I have a somewhat unusual characteristic. At a low effort level (for me it's at about 150 watts, or 135 bpm, which is in my endurance training zone), the lactate concentration in my exhaled breath (and bloodstream) starts to noticeably increase. For most cyclists, that's an indicator that with a little more effort, a pretty good muscle 'burn' and a limited amount of exertion time is in store. But since I know that my functional threshold is at 238 watts and 160 bpm, that doesn't apply for me. I have a double-deflection lactate threshold. Tony says he has seen one other client exhibit the same traits, and Richard said that he has seen it once, in Charles Kulp, to whom his new book is dedicated.

The 'diagnosis' is both good news and bad news. The good news is that one of my negative genetic traits, a low initial lactate threshold, is offset by a positive genetic trait, a high capacity to buffer lactic acid. It probably means I can ride for a long time at very near my threshold (about 4.25 hrs yesterday at 89% intensity factor). But I think it also means that once I do exceed my threshold, I can't stay there very long, but can recover quickly for another short burst. Essentially, I may have lots of matches in my book, but they aren't very hot.

In the next few months, I plan to: 1. Perform testing to see how my blood lactate levels relate to what I've seen from the ventalatory threshold testing I've done so far, and 2. Figure out if I should modify my training in any way to better accommodate my type of body chemistry.

Thoughts on threshold measurement

I like data, so about a year ago, I had my lactate threshold professionally measured using a breathing apparatus that measures accumulation of lactate in your exhaled breath, which represents an accumulation of lactate in your bloodstream. In theory, lactate accumulation means that you have reached or exceeded the point where your body can clear lactic acid faster than it's being produced by the work of your muscles. A buildup of lactic acid eventually requires you to lower your power output (effort).

Two tests in winter 2005 indicated that my threshold, based on lactate accumulation, was at a heartrate of about 135 bpm. I hadn't been training much in fall 2005, but that seemed very low to me. I knew from hard training rides and training for triathlon that I could maintain a heartrate of 160 for an extended period, 20 or 30 minutes at least.

Over the past year I've done some bike racing and have determined purely through trial-and-error that my threshold is around 158-160 bpm. (I'm defining threshold here as the effort on the bike that I can sustain for a lengthy period, specifially, I'm using the Dr. Andrew Coggan definition of Functional Power Threshold (FTP). He says FTP is the max power that you can hold for 1 hour, which is a very practical definition as far as I am concerned.) I recently bought a power meter, and from last week's test, I know for a fact that my current FTP is 238 watts. I also know for a fact that my FTP heartrate is 160 bpm (it was very steady during my test).

I don't question whether my 2005 lactate threshold test actually reflected an increase in blood lactate concentration around 135 bpm. So that leads me to this conclusion: Be very careful about how you define threshold effort. It might be accurate to state that my lactate threshold is 135 bpm, but of what use is that information? I don't want to use that number to set up my training ranges, and it doesn't tell me how hard to ride in a breakaway. It's just the effort at which my blood chemistry starts changing.

I choose to define threshold as a number that is useful for me in training and racing. I think the Coggan FTP is most useful in that regard.

FTP test

Now that I've had my Ergomo power meter for a few weeks, I've become both comfortable with it and confident in what it's telling me. So yesterday I performed my first functional threshold power (FTP) test. I use the term FTP as defined by Andrew Coggan's work: it's the highest average power that you can hold for a 1 hour time trial in a race (which equates to about 103% of your training TT results). It can be estimated by doing a 30 minute time trial in training or by taking 97% of a 30-minute time trial at a race (where motivation is theoretically higher).

I decided to use the 'gold standard' 1-hour test. My average power for a 1 hour TT in training was 231 watts. 231/0.97 = 238, so I'm using that as my FTP now to establish my training zones. (My normalized power was 235 watts for the test interval.) Last February I estimated my FTP at about 210 watts, so it's increased by about 13% in less than a year. I'd like to get it to 250 by mid 2007 (3.6 watts/kg).