Sometimes it feels like everyone you know is either coupling up or breaking up all at the same time - and new research suggests that when it comes to divorce, at least, there really is a seasonal pattern couples tend to follow.
After analysing all of the divorces filed in the US state of Washington between 2001 and 2015, researchers have shown that there are two months every year when divorce spikes - March and August.
The researchers from the University of Washington say this is the first quantitative evidence of biannual patterns for divorce filing, and suggest that the peaks might be a sign of divorce following a "calendar ritual", where couples want to stay together for the Christmas and summer holidays before making their decision.
"People tend to face the holidays with rising expectations, despite what disappointments they might have had in years past," said one of the researchers, Julie Brines.
"They represent periods in the year when there's the anticipation or the opportunity for a new beginning, a new start, something different, a transition into a new period of life. It's like an optimism cycle, in a sense. They're very symbolically charged moments in time for the culture."
But at the same time, the holidays can also be filled with stress and disappointment, leading many people to file for divorce once they've had time to sort out their finances and legal representation after the holidays.
The team presented their research on Sunday at the 111th American Sociological Association annual meeting in Seattle, but their findings have yet to be peer-reviewed, so the pattern needs to be independently verified before we can take it too seriously.
Another study in 2014 also showed a spike in divorce filings after the December holidays, but this time in January, not March, so more research needs to be done to show just how soon after the break families are most likely to split up.
The delay between the December holiday period and the March spike in divorce proceedings could be to do with waiting for children going back to school, or the start of spring in the northern hemisphere, Brines suggests.
The Washington researchers also only took into account divorce filings across one state, which presents a skewed and limited sample size.
They're now looking into whether the pattern translates to other states, and have already looked at Ohio, Minnesota, Florida, and Arizona - although they're in the early stages of analysis.
"What I can tell you is that the seasonal pattern of divorce filings is more or less the same," said Brines.
The interesting thing is that the researchers weren't looking for divorce patterns at all - they were going through social data in the state of Washington to try to get a better understanding of how the global financial crisis had affected people's lives.
But when looking at the divorce records, a very clear pattern emerged that they couldn't ignore. "It was very robust from year to year, and very robust across counties," said Brines.
There were almost 25,000 divorces filed in Washington in 2014 alone.
The pattern remained even when they took into account other seasonal factors such as unemployment and the housing market, leading them to predict that family holidays were driving the trend.
To test this, the team looked at the trends of another family legal decision: guardianship filings, where parents file for custody of children.
The hypothesis was that if the divorce filings were tied to family holidays, then other family court actions would also follow the same patterns. That was exactly what they found - guardianship filings spiked after the holidays, but non-family legal matters, such as property claims, didn't follow this pattern.
The only time the biannual divorce spike seemed to change was during the recession, which was a time of general volatility.
It's too soon to read too much into this research - it needs to independently verified and the sample size expanded before we get too carried away.
But it's an interesting preliminary insight into the season ebb and flow of family relationships that could help us better understand and predict break-ups and legal proceedings in future.