Cooks sometimes use the Leidenfrost effect to estimate the temperature of a frying pan by flicking a few drops of water onto the heated pan. I had no idea, before looking into this, that this could be such a rough estimate.
One site I looked at has the Leidenfrost point for water at 482°F (250ºC).
The Wikipedia article (without citation) places the temperature that this happens around 193ºC, but also notes that the estimate is quite rough,
The temperature at which the Leidenfrost effect begins to occur is not easy to predict. Even if the volume of the drop of liquid stays the same, the Leidenfrost point may be quite different, with a complicated dependence on the properties of the surface, as well as any impurities in the liquid [...] As a very rough estimate, the Leidenfrost point for a drop of water on a frying pan might occur at 193 °C (379 °F)
When estimates differ by more than 50ºC, that really is pretty rough (from a cooks perspective at least).
My first question is whether it is correct that the Leidenfrost point for water really is so difficult to pin down?
And secondly, is there anything cooks can reasonably do to improve this technique in terms of recipe repeatability?