What are the odds of losing 18 straight playoff games?

I honestly don’t have much to say about the atrocities that took place at Target Field over the past few days. It defies all logic that the Twins haven’t won a single playoff game in their last 18 tries. Regardless of the on-field product, you’d think they’d get lucky at least once in 18 tries. I decided just to do some math and share my findings.

If you flip a coin 18 times in a row, the odds of getting tails (or heads) all 18 times is .00038%. That’s about 1 in 261,985 (assuming that the Twins have had a 50% chance of winning any given postseason game). If you live in Montana, those odds are similar to the chance of getting struck by lightning (1 in 249,550).

However, I decided to handicap each individual game, rather than assuming the Twins actually had a 50% chance of winning each one of them. 2017, 2019, and 2020 were easy- Fangraphs readily provided our odds in each playoff game for those years. However, I was unable to find their formula, so I made up my own for the rest of the games in the streak.

My (admittedly very rough) formulation ended up like this:

Since the divisional era started (1969), the home team has won 53.8% of all games. This was my starting point, and I gave it a weighted value of 1.0, considering that it is a massive sample size. If the Twins were away from home, I used the away winning percentage, 46.2%.

Next, I looked at the regular season record of each team. I found the percentage that the difference represented (so, say, 92 wins is 91.1% of 101 wins, representing an 8.9% difference). I assigned this factor a weighted value of 0.5, considering that the team with the better record is far from winning every playoff game (I told you this would be rough).

Lastly, I looked at the starting pitching matchup. Given all other things are equal, I basically assumed that (especially in the playoffs) the team with the better starting pitcher normally wins (representing a greater importance than any other single player). Using ERA+, I calculated the percentage difference between the starters. I also assigned a weighted value of 0.5 to this factor, which may be on the heavy side, but I think my results turned out to be very reasonable.

The final equation: (Home field (Dis)Advantage) +/- (Record Difference % * 0.5) +/- (Pitching Difference % * 0.5) = Percentage Odds of Losing

When it was all said and done, only 6 of the 18 playoff games (12 calculated by me) fell within 5 percentage points of a true 50/50 split. We find that the Twins were basically screwed against a juggernaut Yankees team in 2009 (a team with 16 more regular season wins than the Twins), with all three games rating at a 70%+ chance of losing. The Twins highest win probability came in Game 4 of 2004, when a ridiculous pitching advantage and home field gave a win probability of 73.8%. Two games in the 60%+ win probability occurred in the Athletics series of 2006, due mainly to superior pitching (obviously, it didn’t matter). The 2010 Yankees series was basically a coin flip, which the Twins happened to lose three times in a row. The 2017 Wild Card game was long odds for the Twins, and 2019 and 2020 were coin flips we had a disadvantage in and advantage in, respectively.

It all adds up to this: using (rough) weighted probabilities, the odds of this 18 game losing streak come out to about 1 in 176,860, or .000565%.

There’s a popular lottery in North Dakota called the 2by2. Your odds of winning the grand prize of $22,000: 1 in 105,625.

Guess I’ll go buy a ticket.