Pierre LeBrun writes:
"teams with the better second-half records have good playoff runs, which is why you should be wary, in some cases, of the final standings. Throw out the first half."
Now there's a huge amount of luck involved in winning a playoff series these days, but let's take a look at the relationship between playoff wins and 1st and 2nd half records from 1988-2009:
None are particularly good predictors, with the 2nd half actually faring negligibly worse than the 1st. The key, I think, is that you want the most games possible in your sample. This is true no matter how we group the teams - by 1, by 10 - whatever. In every case, the full season record is the best predictor, followed by the 1st half, with the 2nd half the least.
What if we look at how teams that made it to various stages of the playoffs did during the 1st and 2nd half of the season?
|1st Half||2nd Half||Difference|
Here, second-half records are good indicators - while the first-half records for all but the best and worst teams are essentially indistinguishable. At the same time, the Stanley Cup winners started out very strong and regressed very heavily in the second half of the season, as we might expect for teams with the absolute best records.
Here are the average records of teams that won 0-16 games in the playoffs:
|Wins||1st Half||2nd Half||Difference|
The general trend is for teams with better regular season records to win more games, and in this case the second half record has about 20% better predictive value than the first. At the same time, the average Stanley Cup winner over the last 20 years crashed in the 2nd half of the season. Betting against teams that looked like they got worse in the 2nd half has not been the best strategy.