The pilot reported that, the day before the accident, he filled the airplane to capacity with fuel and then flew 50 miles to his home base. On the day of the accident, he visually examined the fuel tanks through the filler caps but did not use a dipstick. The right tank was full, and he determined that the left fuel tank contained about 15 gallons. He selected the right tank for takeoff and initial cruise, however the selector valve handle felt stiffer than usual. About halfway into the flight as he was about to switch tanks, he noticed that the right fuel tank gauge was still indicating full, but the left tank was empty. He confirmed the right tank was selected, but a short time later the engine lost all power. He performed trouble shooting steps and moved the fuel selector valve back and forth, but it now felt loose. A forced landing was initiated at a nearby airport, however, the airplane landed short of the runway, struck a fence, and the right wing sustained substantial damage. Post-accident examination revealed that the right fuel tank had been breached on impact and contained no fuel, and although the left fuel tank was intact, it was empty. There was no fuel in the line from the fuel selector valve to the carburetor, and the carburetor bowl was empty. Both fuel caps were in place at their respective filler necks, all fuel lines fittings were tight, and there was no evidence on the airframe or wings of staining or streaks to indicate an inflight fuel leak. The fuel selector valve handle was pointing to the right tank and could be moved between positions but felt tight and had a rasping action. Once in the respective tank positions, the handle was loose, and no definitive detent was felt. Further examination revealed that the handle was slipping on the selector shaft, which remained stationary at the left tank position. Mooney specifications called for the handle to be keyed in position with a roll pin fitted to the shaft and secured with a set screw. Examination revealed that the roll pin had previously broken, and the handle was instead held in place with an oversized stainless-steel screw in place of the set screw. The screw appeared to have come loose, such that without the roll pin it was not positively attached to the shaft. Under this condition, movement of the handle did not result in movement of the fuel sector valve, and it was stuck in the left position throughout the accident and previous flights. A photo of the instrument panel that the pilot stated he took about 40 minutes before the loss of engine power indicated that the airplane was flying at an altitude of 6,500 ft. The left tank contained between 8 and 10 gallons of fuel. This should still have been sufficient for continued flight beyond the accident location. However, the accuracy of the gauges could not be determined, and the airplane was not equipped with a secondary fuel quantity reference device such as a totalizer. Additionally, because the pilot did not dip the tank before takeoff, the true quantity of preflight fuel could not be determined.