It’s a tricky one for the efficient market theory. Last Wednesday, the Wall Street Journal reported that BlackBerry was cutting 40 per cent of its staff. The company couldn’t even manage a denial. This looked like final confirmation of the Canadian company’s demise. Reports of unsold inventory had been swirling (the company called for an investigation of the analysts responsible, betraying its own fears). BlackBerry had already said it was exploring “strategic alternatives”. Yet news of the layoffs did not move the shares. Perhaps, at $10.25, the stock was already trading on the assumption that BlackBerry would be sold for scrap. But then the group said on Friday it would report a terrible second quarter, with revenue falling by half from the first. The shares then tumbled by a fifth – stopped only by a conditional $9 buyout offer yesterday. Had Friday’s sellers thought everybody was getting fired because things were going well? How many hints did these misbegotten souls need?

The best explanation of the bizarre trading may be the same one that behavioural economists use to explain the allure of lottery tickets. Small probabilities of big gains do strange things to people’s risk appetites. If the chance of a gain moves from, say, 51 per cent to 52 per cent, most people will pay just a bit more to invest. But if you move the probability from 0-1 per cent, some investors (or punters, if you prefer) will stump up big. This is the “possibility effect” and it is ugly when it goes into reverse. Those who held on to BlackBerry until Friday were emotionally attached to the big gain they could make if BlackBerry fixed things. This sort of hope cannot be extinguished by strong evidence – only ironclad proof. The lesson is to be careful about buying a beat up stock because “the worst is priced in”. You may be co-investing with dangerous optimists.