There is something amiss with you data.The road speed is proportional to engine rpm and to primary sprocket size,that is to the product of the two.So if you take the ratios of these numbers they aught to be the same as as your measured speed ratios,and they are not see table:
Teeth------Product of krpm&teeth---Ratio-----Speed-----Ratio
13------------------122.--------------------------34.8-------------
14------------------129.1-------------1.058------36.5----1.049
16------------------147.2-------------1.140------37.3----1.022
The first ratio in the 14T collumn is the 14/13 based ratio, the second the measured speed ratios,they agree reasonably well.but on the 16T line the ratio between 16/14 ratio(1.14) is way above the measured speed ratios only 1.022 !!.Something is rotten in the State of Denmark.I haven't a clue as to what is more accurate, the speed or rpm measurements,but it does not add up, regrettably.What is odd is that the first ratios are in the ballpark but the second ones are not, so it's not a systematic error,I'm puzzled.Even if you had a 15t instead of a 16 t,we're still way off.If you conduct tests you must be able to trust your numbers otherwise you are just wasting your time.