Immersion cooling is still in experimental phase, b/c it is not yet clear what secondary benefits can arise. That's the gamble investors take.

Secondary benefits include:
- rapid deployment
- less/no impact from ambient temperatures
- operating temperature may be adjustable
- higher clock rates may be possible (e.g. 10% increase reduces the capital cost requirements for hashing power, adding to the ROI of the DataTank)

However, to be fair investors need to determine for themselves whether spending bitcoins is worth the gamble. BTC strikes me as undervalued right now, which may confirm what Spondoolies is asserting. But then, by the same logic any investment into bitcoin mining is questionable at the moment, if you don't have access to equipment at cost.