Even assume all the data are correct, but Shiller obviously is not good to extract information, knowledge, and especially insights from those data. First, it's claimed inflation adjusted, then we should first compare the home value against inflation, which seems to be shown on this chart by the most misleading 100 bold line. Does that means points above 100 line beat the inflation? No, not at all. The whole graph is based on 1890 benchmark, if we choose 1894 as the benchmark, everything will be totally different, but the home value against inflation is independent on the starting point.
This reminds me the charts from Fund managers, which always give you a 10% (or more) return from their date range.
The most wrong part is the dashed line, which was out from Shiller's mind. It doesn't make sense the home values will be back the 1890 benchmark, if he predicted the line would be flat out, I would agree.