BIKE FITTING

A light bike is nice. All your friends get to pick up your bike, say "Wow!" and make envious comments, and then they proceed to drop you like a hot rock once the fast riding begins. Why? Physics. Evil, oppressive physics.

Weight has no effect on wind resistance and only a minute influence on rolling resistance. Once a climb lasts over about a 1/2 mile, a heavier bike starts costing (a little) time, but on flat to rolling terrain (95% of Southern rides & races) the performance difference between a 23lb. bike and a 17lb bike is negligible.

Why? Once a bike is up to speed, assuming the same "frontal area" (what the wind sees), the power needed to keep 167lbs. (17lb. bike + 150lb. rider) rolling down the highway is about the same as needed to keep 173lbs. (23+150) rolling. The real difference is about 2 watts (out of about 190watts total).

But if the 173lb. combo has a smaller frontal area of, say, 2.7 sq.ft., vs. the 167lb. combo with 2.9 sq.ft. (a small change), who's going to roll down the highway faster? Right, the heavy. At the same power output, "light" will go 22mph and "heavy" will go about 22.9mph. For 112miles, that's 12 minutes faster!

For an average rider in an average aero position, to increase speed 2mph from 18 to 20mph takes an additional 34 watts (117 to 151), for the same rider to go from 25 to 27mph would take an additional 58 watts (264 to 322), and to go from 28 to 30mph would take an extra 71 watts (355 to 426). You don't even want to know what it takes for Cippolini to go from 40 to 42mph (okay it's a 166 watt increase, from 1115 to 1281. That's about enough to turn a Ferris Wheel).

Good equipment is nice, and every little bit helps, but we as triathletes have really gotten carried away (I think). This includes me.

For instance, we think we can't race without a new disc wheel, when most testing agrees that compared to a regular 32 spoke training wheel, it's probably only worth about 45 seconds over 40k @ 25mph, less at slower speeds. That's .3mph. Now that's not bad, and I'd take it, but I don't know if it's worth $1200. (They look nice, though.)

I suggest put first things first. Optimize you position, where the big speed improvements are, then add on the toys. Remember: Christmas tree, then ornaments.

None of us seem to be perfect these days, and that goes for our weird-ass legs as well. We all seem to have leg length discrepancies, strength imbalances, pronation, pigeon-toes, slew-feet, rickets, etc. All this has to be taken into account when getting comfortable on your bike. Sometime shims are needed, asymmetrical cleat placement, or even different length crank arms in some cases.

Small amounts of extra efficiency can be found in other areas, such as finding optimal crank length for your personal femur length (not your height), cleat adjustment fore/aft, and minimizing pedal stack height (foot-to-pedal axis distance).

These alone may be worth only 5 or 10 watts, but if optimized, the total watts savings spread over 112 miles is worth more than a new disc wheel or 100 six a.m. swim practices. For example, I lowered my pedal stack height 12mm, and instantly increased my power output 18 watts at the same HR. That was probably more than average, but it is a real-world example of what I'm talking about.

These extra increases aren't aerodynamic, an therefore can't be measured in a roll down test. They can be measured on a CompuTrainer though, and tend to increase with adaptation over time.

Performance Testing

Is one position really faster aerodynamically than another? How about wheels and tires? How do you find out without coughing up a couple of grand for a session in a wind tunnel?

Here's how: With accurate, repeatable roll down testing. This is basically rolling down a hill in a set position without pedaling. The tricky part is to find the right incline, of the proper length, on the right surface, with the right weather conditions, and time each trial accurately enough to make it of real statistical value. Each position has to be tested enough times to establish a mean, which usually takes 6-8 trials. The high and low are discarded and an average is found.

Now this is not a good way to test Joe vs. Betty, any person-to-person testing. But it is accurate for individuals and individual variables. Joe, on the same bike with the same wheels can change and test positions, or in the same position on the same bike can test his wheels, or with the same position, bike and wheels, can test tires, or with the same wheels, tires and position, can test bikes. You get the idea.

The key is an accurate, repeatable timing method. I have one that works well, and I'm trying to improve it even further (and perhaps needlessly) to perfection. Using accurate data, it's possible to extrapolate this data into an idea of real-world speed increases, over whatever distance (30 or 40k, 56 or 112 miles).

This is a procedure that takes all morning. The top-secret testing location is north of Covington, and the testing involves a bit of riding repeatedly up a hill, so there's a free hill workout thrown in with the deal (we can all use more hills in our bike diet).

Testing is by appointment only, Gatorade and snacks are provided, as is all necessary position changing/parts swapping. The testee is responsible for their own transportation.

Current testing data will be posted here soon.