When is a Game Out of Reach?Posted by rtmsf on April 14th, 2008
If you’ll allow us the liberty of jumping back two weekends ago to the Final Four, specifically the Kansas-UNC game, we’ll present you with the following call by the venerable Mr. Cash himself, Billy Packer:
Of course, Packer was swiftly excoriated for this comment by the public and blogosphere at large for being a crotchety old bastard who hates everyone and wants you to die national commentator whose arrogant, smug disposition is unsurpassed (third try – this is hard) knowledgable yet polarizing media figure who sometimes takes indefensible :-) positions as a product of his stubbornness.
The question we had at the time, and the question we still want to consider is this – is calling a game over when one team is leading by 26 points with 27:30 remaining in the game defensible? Packer’s statement sure looked stupid when UNC got the margin down to 54-50 with approximately ten minutes to go in the game. But UNC didn’t complete the comeback (losing by eighteen) and it’s quite difficult to find examples in the college game of really good teams (such as Kansas) blowing immense leads and still losing the game. The only one we can think of in recent history was Maryland’s blown 22-point lead in the first half over Duke at the 2001 Final Four (Duke won 95-84), and that Terp team absolutely could not get the Devils out of their heads (recall the Miracle Minute at College Park that season).
Even Carolina Fans Hate Packer
Therefore from a qualitative standpoint, Packer was probably right. The energy that a team like UNC would have had to expend to not only erase the 26-point deficit but also take the lead and win the game down the stretch would have been mindboggling. While good teams come back from 20-point deficits to win games a fair amount of the time, it rarely happens against other good teams. If you hadn’t noticed, Kansas was a pretty good team this year. UNC, as good as they were, was not going to come back and win that game. They just weren’t.
What about from a quantitative perspective? In a piece published at slate.com the week prior to the 2008 Tourney, Bill James (the original sabre-metrician) noted that he has a trusty heuristic that he uses for college hoops games to make a determination on whether the lead is large enough to “call it.” Too bad Packer didn’t talk to this guy beforehand. Here’s his formula:
- Take the number of points one team is ahead.
- Subtract three.
- Add a half-point if the team that is ahead has the ball, and subtract a half-point if the other team has the ball. (Numbers less than zero become zero.)
- Square that.
- If the result is greater than the number of seconds left in the game, the lead is safe.
Plugging the UNC-Kansas lead when Packer made his statement into James’ handy little calculator, we find that a 26-point lead with 27:32 remaining in the game is only 33% safe, which effectively means that the lead is absolutely and completely safe for the next 11.7 minutes of the game. To be clear, it doesn’t mean that Kansas had a one-third chance of winning – it means that KU was one-third of the way (statistically speaking) from holding an insurmountable lead at that point in the game.
Sorry Billy, You Jumped the Gun
So what would have been a truly insurmountable lead at that juncture, thereby making Packer look utterly brilliant (ok, difficult, we know)? According to James, it would have taken a 44-point lead to justify Packer (or anyone) calling the game over with that much time left. This seems rather high, considering that we only remember one thirty-point comeback in our lifetime (Mardi Gras Miracle), but them’s the numbers. As for us, we stand by the hard-and-fast rule of thirty points at any moment in the game. We’ll give odds on that number the rest of our lives and become a rich man doing it.