A fund management mantra: cull the Dogs, advertise the Stars

What's the game?
If a manager runs a family of ten boring old funds in a boringly average way he can expect at any time that five will be above average and five below. With any luck one will be a Star.

How can he can use this statistical inevitability to his advantage?

• He could advertise his whole fund family by quoting results only for the Star.
• He could liquidate the Dog and transfer its assets to the Star ("Our largest Fund ranks in the top 10% of all funds")
• He could liquidate any number of below-par funds to favourably skew the average performance of the family.
• He could keep starting funds with institutional friends as investors, kill the losers and run the winners. The performance of the winers can be used to pull in new investors.

....and that's why....

• Fund families tend to be so large. The more funds you've got, the easier it is to play this game.
• Fund managers have an incentive to take big bets, particularly in the early days of a Fund's life. If it goes right, he's got a Star. If it goes wrong, his mistakes are buried. Pity about the investors.
• Historical studies of performance are unreliable. They can be favourably skewed because the Dogs have all gone. It's called "survivorship".