singlespeed gear selection - why a 40/23, why Gerry Pflug... why?

[robotfury]

I have been racing, riding, training ss for 3 years now and I am constantly intrigued/fascinated and sometimes confused by gear ratio decisions - I think a standard characteristic of any ss rider/racer (maybe not the confused part).  I am usually riding 32x19 or 20 here in So Cal with all the climbing.  Currently I am toying with 34.21 and 34.22 to test the kook-aid of the Rotor 34t SS ring. The step from 32 to 34 is hefty .  I find the climbing harder - duh -  but manageable on steady climbs; however I am not seeing a significant positive change or an "oh, I get the 34t".

Then I read that, Gerry Pflug just won the Mohican 100 overall on a 40/23 and that he trains on a 36/18. Aside from Pflug being superhuman, that gear choice just seems impossible to me. Why would someone choose a 40/23? Or even choosing a 34t, a reason that eludes me at the moment.

Long way short: Technically speaking, what is the thesis behind adding more teeth up front?I know it is a personal choice to some extent, but there must be some method to the madness... on paper at least to explain this.


Thoughts? 

[THE LONG RANGER]

Larger rings/cogs are going to be slightly more efficient, than small rings/cogs of the same ratio.

For this particular ratio(s), could it be just a, "what's lying around" type of situation? I've done 45 minute cyclocross races and 150 mile gravel grinder races on the same gear ratio. It's usually from laziness/lack of time/lack of interest to dial a gear ratio to the *tooth* or whatever. I wanna ride my bike, not do math.

[THE LONG RANGER]

Well, someone did the math on my hunch #1:

http://www.g-cog.com/VBMX/spicer.pdf

TLDR: It was found that larger sprockets provide more efficient trans-
fer of power while smaller sprockets proved to be less efficient.
Simple, frictional loss models were developed that could be made
to have sprocket size loss variations that agreed with those varia-
tions measured experimentally. Typically, a 2–5% loss difference
was measured between the 52–11 and the 52–21

My hunch #2 still stands as, "let's ride bikes!"

[adelorenzo]

More teeth on the cogs is measurably more efficient. The did all the analysis in Bicycling Science but I don't have my copy handy. But go pedal around in your 11T cog sometime and you'll feel it. Other reasons are to prolong drive-train life (more teeth and links to share the load) and more teeth wrapping around helps keep your chain on. Downside is a bit of weight and less chainring clearance for your trials moves.

The good folks at Surly bikes have a pretty good take on the whole issue too I highly recommend it.

I run 34x19 as my normal gear for a 29er. Also have a 21T cog for loaded bikepacking although so far I have been too lazy to change it out for any trips.