Play off or play on?

Playoffs. Love them or hate them, they sure are entertaining. Which makes you think, the more the better, right?

Talking about playoffs when the regular season is barely underway would seem ridiculous, if it wasn’t for a piece of news that surfaced few weeks back here in Finland: the Finnish league is looking into expanding the playoffs in hockey. After all, the league itself is expanding next year. So it only makes sense.

More teams making playoffs, or at least staying in the run for longer, can only be a good thing, right? It’s more exciting so people tune in more and the game gets a boost in fan demand. That is the argument presented by the league, as expressed by the Finnish league’s board chairman Vesa Puttonen. Everybody wins, right?


Major League Baseball and the -69 and -95 playoff expansions

In his 2009 paper The impact of post-season restructuring on the competitive balance and fan demand in Major League Baseball (Journal of Sports Economics, 11, 136-156) Young Hoon Lee studied the playoff uncertainty and its link to attendance and found empirical evidence that the post-season restructurings of -69 and -95 improved the playoff uncertainty and had positive effects on fan demand. This seems to confirm the expectations of the Finnish hockey league. However…

What happened in MLB in the 1969 and 1995 was, that due to league expansions the old playoff system was no longer sufficient. In -69 both leagues (NL and AL) had 12 teams each. Before -69 the playoffs were basically just the championship series, only the winners of each league had a post-season. After -69, a second round was added to the playoffs, resulting in 2 teams our of 12 making the playoffs in each league, 4 out of 24 in total.

A similar expansion took place in 1995. Due to larger number of teams, another round of playoffs was added. Now 4 teams out of 14 in each league made the playoffs, a total of 8 out of 28 teams.

There’s expansion and there’s “expansion”

The key thing is to look at the number of teams making the post-season out of all possible. After the first structural expansion, 4 out of 24 teams had a post-season in baseball. That’s 17%. After the third round was added, the percentage goes up, to 29%.

Right now, before the proposed expansion, the Finnish hockey league has 71% of its teams making the playoffs. Even with the added team next season, we’d still have 67% of the teams playing for the championship after the regular season.

I highly doubt the fan demand is linear in this regard. Going from 17% to 29% making the playoffs may have had significant impact on fan demand. Going up from 71%? I’m doubtful.

The Swedish study in equal opportunity

Has anyone looked at the Swedish hockey league? They introduced the “pity playoffs”, the “pre-round” for last season*. What happened there? How did the demand react? ‘Cause they’re really getting all in with the playoffs, almost literally: 83% of the teams made playoffs last season. The league expansion to 14 teams isn’t going to bring the percentage down much, only to par with Finland: 71%.

The North American leagues are much more ruthless. Both NHL and NBA qualify slightly over half of their teams to post-season, with 53% both. NFL has probably the most confusing season schedule I’ve ever seen, but they mean serious business with the post-season: 38% of teams get a chance to go all the way to championship.

And what about baseball? It’s grown since we last spoke of it, the league now consists of 30 teams. Still only 10 make the post-season. That’s 33%.

Oh, and in case you were wondering about the KHL? 16 out of 28 teams. That’s 57%.

The percentages of teams making the playoffs in different leagues:

Playoff teams

Pointless games, at the beginning and the end

What would it mean, in reality, if we let more teams into the playoffs? It would mean weaker, lower quality teams getting a change at upsetting the, at least according to a lengthy regular season, stronger teams. I borrowed from a study The “Second” Season: The Effects of Playoff Tournaments on Competitive Balance Outcomes in the NHL and NBA by Neil Longley and Nelson J. Lacey (Journal of Sports Economics, 13, 5, 2012) and calculated the probabilities of each seed winning three rounds of playoffs. Like them, I used the average winning percentages for each seed over several seasons. Unlike them, I didn’t take into consideration the home advantage. I also assumed the “pity playoffs” are played as best-out-of-7, to simplify calculations.

Using the average winning percentage of each seed over 11 seasons (2003-04 to 2013-14) in the Finnish hockey league, I calculated the probability of each seed winning the championship, when weighted by the probability of each possible match-up.

The first seed, the team winning the regular season, wins the championship with 38,6% probability. The 10th placed team has a probability of 1,2%. Here’s the whole list:

Seed Prob.
1 38.6
2 25.4
3 23.0
4 22.0
5 17.7
6 14.2
7 5.3
8 2.8
9 2.1
10 1.2

Why include teams with very small probabilities of actually winning the championship? Besides, the extra round for seed 7 through 10 is what really diminishes their chances. So why increase their number? Are match-ups between the first and 10th placed (40% probability of that happening) really interesting to anyone, when the probability of the regular season winner winning is 72,1%? Against the 8th placed team seed 1 would have a winning probability of 67,2%. I’d prefer to go to the second game, if you ask me.

But because it’s not all math and probabilities, things happen. Weaker teams win. That’s why playoffs are so exciting. Longley and Lacey found, in their study, that the playoffs in the NHL had “considerable reranking effects” when compared to the regular season standings. Which begs to question: isn’t the whole point finding out who’s the best?

While increased playoff uncertainty may increase fan demand, it’s not always so straightforward. In their study “Playoff Uncertainty and Pennant Races” (Journal of Sports Economics, 12, 5, 2011) Anthony C. Krautmann et al. used monthly attendance data from MLB for the 1957-2006 period to better capture the effect increased playoff uncertainty would have on fan demand. They found that “the only month in which attendance is affected by [playoff uncertainty] is the month of September”, that is, right at the end of the regular season.

If it takes, on average, 1,3 points per game to make the playoffs in the Finnish hockey league (using the same data as before), can we really afford to bring that any lower? To make more early-season games relatively insignificant and thus more uninteresting to fans?

Can we really?


* The seeds through 7 to 10 play a round, usually best-out-of-3 or 5, to determine which two teams play against seeds 1 and 2.

Edit: Edited to add graphics on the number of teams making the post-season in each league at the reguest of my good friend and self-appointed social media consultant H, who doesn’t like to read too many numbers in a row. -Nov 10, 2014

One thought on “Play off or play on?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s