Ankle proprioception is crucial for balance and relies upon accurate input from calf muscle spindles. Spindle input, in turn, depends upon the physiological and mechanical properties of surrounding muscle tissue. Altering these properties could affect ankle proprioception, with potential consequences for balance. Here we determine the effects of prior muscle cooling, stretch and contraction upon performance of a contralateral ankle joint matching task. Participants stood passively leaning against a board oriented 22° rearward from vertical. Their right ankle was rotated to a randomised position between ± 6° plantar/dorsiflexion. The task was to align the left ankle to the same position, without vision. In the first experiment, immediately prior to each testing session, participants either produced a strong calf muscle contraction in a fully plantarflexed (tiptoe) posture or underwent 15° dorsiflexion stretch. Contraction had no effect on task performance, whereas stretch produced a significant bias in ankle placement of 0.89 ± 0.6°, indicating that participants perceived their foot to be more plantarflexed compared to a control condition. In the second experiment, the right lower leg was cooled in iced water (≤ 5°C) for 10 minutes. Cooling increased joint matching error by ~0.4°, through a combination of increased bias and variability. These results confirm that conditioning the triceps surae muscles can alter perception of ankle joint position. Since body movement during quiet stance is in the order of 1°, the magnitude of these changes are relevant for balance.