Sorry for the misinterpretation of your response...glad to read that you didn't think I was attacking you; cause I wasn't.
My set up and #'s are not applicable to your set up. I'm 5'7' and 160lbs. Your 6'5" and 160lbs. Two different zip codes.
My suggestion to you was to not look to Mr Day for the next generation of you. Look to Dan or Dave Luscan, or Jim Martin. Pick one and go with it.

Good luck with this, Dave. From what I can see and how I interpret the findings so far, I think you'll be better in 2018 than you were in 2017....Move on from Frank D and, in my opinion, you'll be even farther along.

Cleveland would be a great test for these
changes. Think about it.

I just found the one guy on Strava. I could not for the life of me log THAT many miles indoors on any kind of trainer living in CA with the weather they have for any purpose like cranks or even an IM race. And do soooooo many treadmill miles and miles on the same route.

I'd shoot myself in the face.

Why are there literally like no outdoor rides?

Also tons of seeming manual entries with no power and no HR. There are velotron rides with data from Strava users. Why not this one?

A super expensive velotron but no watch or bike computer/pm/hrm data? I'm a schmuck nobody, but I've got data that says I'm a schmuck nobody.

I'm done reading this.......

H2O Fun actually is in a well known CA facility where they tend to "contain" people for their own sake (jokingly also referred to as a "Country Club").

As he is behaving quite well (outside of trolling the Interwebs) by cycling and running his brains out in place, he is now and then allowed outside for racing (only in USA[T] kit, which contains a sewn in tracer) in a controlled environment.

So we can all stop speculating about the "why" and "for what reason".

I have to admit that I haven't figured out yet what incident caused him to end up at the place with Frank Day, but these posts give me at least a pretty good idea why.

As I read that chart the cost of pedaling is about 150 watts at 1.42 m/s and about 120 watts at 1.19 m/s. This is a 30 watt difference and we saw an average of a 23 watt improvement. Coincidence I guess.

I earned, Dr. - I did all that work, which helps me understand all this stuff

Wasn’t this calculated from the residual from the mechanical watts, was that calculation multiplied by 5?

The devil is in the details.

The other issue is the shape of the curve. Eyeball suggests that it would intersect the Y coordinate between 25 and 50 when, in fact, it has to intersect at zero.

Do you recommend that those reading this thread ignore the possibility of an easy 3-6 watt increase because you think it small? Do you think this advantage should be ignored because, well because it it just so strange to ride slower cadences and you didn’t think of it? Do you think efficiency is important? (some here don’t).

Figures 1 2 and 3 all have metabolic power on the y axis. The residuals shown in figure 2 are in metabolic watts. I think the figures are pretty clear. If things are unclear it might help to read the paper.

Indeed.

No it doesn't. As pedal speed gets really low, metabolic cost goes up. This is probably due to the high force required and the resulting recruitment of type II motor units. That regression just says that low pedaling rates are less efficient than average.

As I wrote earlier in this thread (post 585), one might see around one percent change in metabolic cost associated with a reasonable range of pedal speeds. I'd still say that's in the right range. So, should you worry about 1% change in metabolic cost? Sure. It will alter speed by around the cube root of 1.01 or about 0.3%. In a one hour ride, that efficiency would improve your time by about 12s. As others have pointed out, pedaling at higher force might impact the run so that might be something else to keep in mind.
As you said, the devil is in the details and I would add "the devil is in the big picture".

I have no idea how someone is going to prove how the bike would directly impact the run with data

Gee, lets think. Maybe two or more pedal speeds with controlled power followed by running trials?
Looks like cadence mainly effects the early part of the run. Google scholar is your friend.

J Sports Sci Med. 2005 Sep 1;4(3):342-53.
The effect of cycling cadence on subsequent 10km running performance in well-trained triathletes. Tew G.
Abstract The aim of this study was to examine the effects of different pedalling cadences on the performance of a subsequent 10km treadmill run. Eight male triathletes (age 38.9 ± 15.4 years, body mass 72.2 ± 5.2 kg, and stature 176 ± 6 cm; mean ± SD) completed a maximal cycling test, one isolated run (10km), and then three randomly ordered cycle-run sessions (65 minutes cycling + 10km run). During the cycling bout of the cycle-run sessions, subjects cycled at an intensity corresponding to 70% Pmax while maintaining one of three cadences, corresponding to preferred cadence (PC), PC+15% (fast cadence) and PC-15% (slow cadence). Slow, preferred and fast cadences were 71.8 ± 3.0, 84.5 ± 3.6, and 97.3 ± 4.3 rpm, respectively (mean ± SD). Physiological variables measured during the cycle-run and isolated run sessions were VO2, VE, RER, HR, RPE, and blood lactate. Biomechanical variables measured during the cycle-run and isolated run sessions were running velocity, stride length, stride frequency, and hip and knee angles at foot-strike and toe-off. Running performance times were also recorded. A significant effect of prior cycling exercise was found on 10km running time (p = 0.001) without any cadence effect (p = 0.801, ω(2) = 0.006) (49:58 ± 8:20, 49:09 ± 8:26, 49:28 ± 8:09, and 44:45 ± 6:27 min·s(-1) for the slow, preferred, fast, and isolated run conditions, respectively; mean ± SD). However, during the first 500 m of the run, running velocity was significantly higher after cycling at the preferred and fast cadences than after the slow cadence (p < 0.05). Furthermore, the slow cadence condition was associated with a significantly lower HR (p = 0.012) and VE (p = 0.026) during cycling than in the fast cadence condition. The results confirm the deterioration in running performance completed after the cycling event compared with the isolated run. However, no significant effect of cycling cadence on running performance was observed within the cadence ranges usually used by triathletes. Key PointsCompared with an isolated run, completion of a cycling event impairs the performance of a subsequent run independently of the pedalling cadence.The choice of cadence within triathletes' usual range does not seem to influence the performance of a 10km run.The results reinforce the necessity for triathletes to practice multi-block training in order to simulate the physiological responses experienced by the cycle-run transition.Further research into the effects of cycling cadence on subsequent running performance is required.

BioMcGeek wrote: "No it doesn't. As pedal speed gets really low, metabolic cost goes up. This is probably due to the high force required and the resulting recruitment of type II motor units. That regression just says that low pedaling rates are less efficient than average.”

Wrong again my friend. Figure 4 is for unloaded pedaling. Metabolic cost does not increase as speed slows if contractility also decreases. Unloaded is unloaded. So, as speed slows contractility needed to pedal also falls. When pedal speed is zero required contractility is zero. Hence, metabolic cost at zero pedal speed is zero. Metabolic cost of pedaling increase somewhere between the square and cube of the pedal speed, the number depends on the chain and bearing losses.

Frank Day

Are you still running an 80.5cm seat height? How tall are you?

BioMcGeek wrote: “As I wrote earlier in this thread (post 585), one might see around one percent change in metabolic cost associated with a reasonable range of pedal speeds. I'd still say that's in the right range. So, should you worry about 1% change in metabolic cost? Sure. It will alter speed by around the cube root of 1.01 or about 0.3%. In a one hour ride, that efficiency would improve your time by about 12s. As others have pointed out, pedaling at higher force might impact the run so that might be something else to keep in mind.
As you said, the devil is in the details and I would add "the devil is in the big picture”.”

Wrong again. There are plenty of people here on 175 cranks riding at 100 or so cadence (that is similar to when Dave was riding 200’s at 85-90 rpm). Figure 4 puts that cost at 300 metabolic watts. 20% of that is 60 watts. Bringing that down to a more reasonable cadence of 77 reduces that metabolic cost to 150 Watts or 30 real watts according to you. That is a 30 watt change if one keeps effort (oxygen consumption) the same. 30 watts is a lot more than 1% of what most of the people here race at. Do you really consider that small and unworthy of trying.

Whatever the improvement, small or large, isn't this what people like to call "free speed”?

Frank Day

during the first 500 m of the run, running velocity was significantly higher after cycling at the preferred and fast cadences than after the slow cadence (p < 0.05).
And summary is what? I just see it saying cadence basically makes no difference. Am I reading this wrong?

Does the bold underlined font help you see the cadence effect?

No because there is not number on what a fast or slow cadence is. Different folks will have a totally different view on what is fast or slow.

What I have found, and talking to many other athletes, if one is not doing bricks, enough running, etc. it can take a long time to "warm" up off the bike for the run.
Since I do a lot of right off the bike to my treadmill, I can get off the bike and hit full speed at the beginning of the run. Is one of my assets with my training.

So I see lots of real reasons most folks are slow in the first part of the run, and it has nothing to do with crank length or cadence they are spinning, in my experience.