In an otherwise excellent piece criticising the Ottawa Senators for their collapse in game six against the Penguins, Wayne Scanlan remarks that the pressure was building as the Senators “set out to defend what Don Cherry always calls the worst lead in hockey: the two goal advantage.”
Now I know what he’s getting at: With a one-goal lead, a team keeps its focus, and with a three goal lead it would require a serious collapse to lose. But two goals? It’s close enough that you need to stay focused, but big enough to convince you that it is OK to relax. But is there any indication that a two goal lead is, objectively, a worse lead to have than a one- (or three-) goal lead? I find it hard to believe.
First thing to keep in mind is that all failed two-goal leads are also failed one-goal leads. That is, on its way to squandering a two-goal lead, a team must also squander a one-goal lead. So, it is analytically the case that the number of squandered one-goal leads is equal to, and empirically a certainty that it is greater than, the number of squandered two-goal leads.
There is one way the cliche might be true: a team might have a better record of holding the lead when it goes up by only one goal than when it goes up by two goals. In which case, the obvious coaching strategy when up by a goal would be to insist the players try not to score. Has there ever been such a team?