Author Archive

What Data Can Tell Us About Kansas City’s Home Run Struggles

After getting out homered 5-0 by the Angels this weekend, the Royals sit at an underwhelming 20 home runs in 49 games, good for 30th in the league and less than half of the league average of 45. Early in the season, it can be tough to distinguish if under-performance in a certain outcome is due to random fluctuation or an actual decline in talent. Luckily, we have a litany of data at our disposal that can help to answer this question.

Since Kansas City does not have a lineup stacked with power hitters, and playing in Kauffman Stadium makes hitting home runs more difficult than many other stadiums, it’s preferable to compare current production to a projection system instead of league average in order to get a sense of the scale of the Royals’ current power struggles. This already takes into account both the team’s lineup and ballpark factors, giving us a better comparison. In the preseason, Steamer projected that the Royals would hit 126 home runs in 2014. Applying that projection to the 49 games Kansas City has played, we get that the team was projected to have scored 38 home runs through this point in the season. Using the linear weights from the wOBA formula, we can calculate that had the Royals hit 38 home runs as Steamer projected, they would own a (league average) .317 wOBA and a wRC of 202. Instead, Kansas City has a team wOBA of .296 and a 173 wRC. In essence, these 18 home runs have cost the team 29 runs in total, or 2.9 WAR.

Things should change going forward, however. Steamer posts daily updated projections that change as more historical data becomes available (i.e. more games are played) . Taking into account the abysmal start by KC, Steamer has updated their projected year end total home runs from 126 to 102. We already know that 18 of that 24 home run difference is historical, so the change in home runs projected through the rest of the season amounts to only 6 for the remaining 113 games. After factoring in playing time adjustments, Steamer has now discredited Kansas City 9 home runs that were expected at the beginning of the season. Although this represents a non-trivial  drop in home run rates, it is significantly less severe than the pace the Royals have set so far this season.

This does make some sense. Steamer has years of major league performance data to shape player performance for each of Kansas City’s starters, and centuries of baseball data on which to base aging curves. It seems pretty unreasonable to significantly change a projection based on less than two months of data from the current season. This would be especially unreasonable given that home run rates do not stabilize for a given player until about 300 plate appearances. Eric Hosmer has the most PA on the team at 218, so it will probably be another month before we have an idea of whether or not the Royals’ power outage is anything more than random fluctuations.

Another reason we might expect that this trend will not sustain is that much of it appears to be luck-based. Over the past five years, the Royals have had a HR/FB of around 8%, and the lowest they posted over a full season in that time frame was 6.9% in 2010. So far this season, Kansas City has a HR/FB of 4.5%. In addition, the team has hit 7 more doubles than Steamer projected for the season so far, supporting the theory that the Royals have had more than their fair share of balls land just on the wrong side of the fence. This does not account for all 18 home runs that were projected to be hit and were not, however. Bad luck only explains so much, and the majority of KC’s offensive woes still should be credited to poor hitting.