You would never invest in a company like PartyGaming, which came to the market with a yard-long wealth warning pointing out, among other things, that its activities might be illegal in the US, its main market, and directors with, well, a gamey past.

Maybe.

Still, there's a good chance you did invest in it - if your pension scheme followed any half-conventional investment policy, or if you have money in almost any middle of the road packaged investment with a broad spread of so-called "UK" equities.

Worse, it is unlikely that you bailed out when several of those directors cashed in their chips earlier this year.

Some of the £61 million dividend declared a couple of weeks ago would have been yours, if only they hadn't cancelled it yesterday.

Almost certainly, your investment manager chose only whether to be over-weight or under-weight in PartyGaming. To avoid it altogether would risk a bonus-losing run of under-performance if ever it had a winning streak and beat the rest of the Footsie for a while.

Few fund managers have that kind of courage.

If you, or your fund manager, played safe in a tracker fund, you stand to lose more still as the tracker's computer now scales back your holding in PartyGaming at a price little more than a third of where it stood on Monday morning. The computer is set to track its reduced status as a proportion of the Footsie, which will doubtless be reduced further still when it drops out of the index in December.

This is the way conventional, middle of the road investment works. Size matters - even in an "active" fund run by someone paid generously to use his, or her, judgment, knowledge and commonsense. In real life, if a company is big enough to be in the Footsie, an investment manager who is paid to meet a benchmark associated with the Footsie wanders any decisive distance from it at his peril.

As to the real tracker funds, true, most of them track the All-Share index. Even so, that still means keeping most of the money in the companies with the biggest market value - then dumping them at a loss if things go wrong and it gets relatively smaller.

The dispiriting thing is that a computer tracking size, regardless of quality, beats judgment, knowledge and commonsense almost every time - not only because it costs a lot less.

It ought to be possible to beat the index by just avoiding self-evident duds. In practice that doesn't work because duds get taken over far more often than they go bust, while plenty of well, but cautiously run, companies lag at the back of the herd because the stock market considers them boring.

Why, though, cannot someone devise an index that leaves out all those exposed to a single, identifiable legal hazard?

PartyGaming and the rest have fallen victim to a distasteful alliance of anti-gambling puritanism and a protectionist tactic to shield American casinos from offshore competition.

A special case, certainly. Still, are you quite sure nothing like it could happen to tobacco?