Yep, you're absolutely correct. My bad. The fresh air going through the cylinder would cause it read as a leaner mixture. Perhaps a ~15:1 ratio measured by the sensor would actually give you close to the ideal 12.5:1 ratio for maximum power in the cylinder? Would you assume the readings would vary some depending on how far from the exhaust port the oxygen sensor was placed in the exhaust pipe? Due to that very reason you listed above with the fuel and air being sucked back into the cylinder?