- FanGraphs+ | FanGraphs Baseball - http://www.fangraphs.com/plus -

Are Second Half Surges Predictive?

On July 1st, the San Diego Padres stood at 29-50, a worse mark than every team in baseball besides the Chicago Cubs. They’d been outscored by 80 runs, and simply looked like a team that was far from being competitive. Since that day, however, the Padres have won 42 of 70, a .600 winning percentage that is tied for the fifth best mark in baseball. The addition of top catching prospect Yasmani Grandal and a resurgent Cameron Maybin have the offense clicking, and their second half success has people talking about the Padres as contenders in 2013.

And certainly, the Padres do have several interesting young players, and their farm system was rated as the best in the game by Keith Law before the season started, so there are reasons for Friar-related optimism. However, before we get too carried away by their recent string of strong play, it would be helpful to know whether these kinds of second half surges have actually carried over to the following season.

Over the last five years, I found eight examples of teams that posted losing records before July 1st, but had a winning percentage at least 100 points higher in the final three months of the season than they did in the first three. While the Padres mid-season turn around seems unexpected based on how they played in the first three months of the year, this phenomenon happens pretty much every season, and sometimes multiple teams pull off large second half improvements in the same year.

So, how often did those gains carry over to the following season? Well, Padres fans, you might not want to read any further, because you’re probably not going to like the answer.

Year Team 1st Half 2nd Half Next Season
2008 Rockies 0.386 0.532 0.568
2009 Braves 0.474 0.581 0.562
2006 Phillies 0.456 0.590 0.549
2011 Dodgers 0.439 0.582 0.513
2007 Reds 0.383 0.506 0.457
2010 Orioles 0.312 0.494 0.426
2008 Indians 0.446 0.557 0.401
2010 Astros 0.392 0.542 0.346

Of the eight clubs that had similar leaps, only one team — the 2008 Rockies — actually improved upon their second half winning percentage in the following year, and that example comes with a fairly large asterisk, because the 2007 Rockies team made it to the World Series. They started their NL title defense with a thud, and their second half rebound and subsequent 2009 improvement was more about returning to established levels of performance. The Rockies played poorly in the first half of 2008, but those were really their only poor stretch of baseball over a three year time period, so they don’t necessarily fit the model of an upstart team having a strong finish as a precursor to what is to come.

Besides that Colorado team, every other club played worse in the following season than they did during their second half improvement. Of course, there was far more room to go down than up, so perhaps that was to be expected, but the magnitude of the overall declines doesn’t speak particularly well for the Second Half Surge theory.

On average, the seven decliners lost 67 points off their second half winning percentage, dropping from a .548 aggregate winning percentage as a group to just .478 in the following year. The 2010 Braves had the smallest decline at 19 points — and they did manage to win the NL Wild Card, so perhaps you would like to count them as a success story as well — but they were also the best of the first half teams that we were examining, having posted a .474 winning percentage in the first three months of 2009.

Four of the seven posted losing records in their follow-up season, and two of them — the 2009 Indians and 2011 Astros — posted lower winning percentages in the next season than they did in the first half of the prior year. Overall, the .478 winning percentage posted by these teams in the season after their “Second Half Surge” was slightly lower than the .480 winning percentage they posted during the season in which they appeared to be two totally different clubs.

And perhaps that small difference is the real takeaway here. At the group level, these teams posted an almost identical record in the next season as they did to their total record in the prior year, suggesting that the first half record of a team that improves significantly is just as important as the second half data. Had we just used total season winning percentage as a predictor of next season winning percentage, we’d have hit the overall mark almost dead on for these eight teams.

The Padres play of late is certainly encouraging, but it doesn’t mean that we can just ignore all the problems that were on full display in the first three months of the season. If we want to understand how teams will do next year, we should look at their entire season as a whole, and not fall into the trap of putting too much weight on recent performance. The Padres may very well be a team on the rise, but their record since the calendar turned to July is simply not enough evidence to suggest that they should be penciled in for a playoff run in 2013.