I had been wondering for some time if the statistics for the two-leg format in the MLS playoffs were all that different from elsewhere. The higher seeds have complained in the past about having to host the second leg at home, instead of opening the playoff series at home, which is what happens in all other team sports in North America. So I decided to look behind the numbers of the two-leg format in MLS Cup playoffs since 2003 to determine whether higher seeds are disadvantaged by hosting the second leg.
Since the 2003 season, there have been 28 MLS playoff series that use the two-leg format.
Half of the first leg matches ended in a win for the home team. Of those matches, the first-leg host has gone on to win the series 43% of the time. The second-leg host was able to turn around the deficit 50% of the time, and one series (7%) went to a penalty kick tiebreaker.
If the first leg match ended in a draw, which happened in 32% of the series (9 matches), the odds for the first-leg host lengthened considerably. In this situation, the first-leg host has won only 2 series (22%) while the second-leg host has won 6 (67%). One series went to a penalty kick decider.
If the first leg match was won by the visitors (5 matches), the advantage swung decisively to the second-leg host. They won 80% of the series with only one going down to penalties.
These trends are the same as those of the two-leg series in European club competitions, but the magnitude of the trend isn't as large in MLS. In the UEFA Champions League, winning the first leg as host improves the chances of progressing (roughly a 63% probability of winning the series). A draw or a loss shifts decisively the advantage to the side hosting the second leg. Since the 2002-2003 season, on just four occasions (out of 106) has a visiting team won the opening leg yet lost the two-leg series. A draw improves the second-leg host's chances of winning the series to between 63-75%.
So to conclude, what we're seeing in the two-leg series in MLS isn't all that different from what we see in Europe. The biggest differences are the parity between competitors and the inexperience of MLS sides in killing off matches and taking full advantage of home field, which could explain why the probability of winning a series as first-leg host is lower.
I'll include the data later and say a lot more about it on Soccermetrics, but Typepad's editing tools are screwed up at the moment.