Change text placement for ROC curve figure

met/docs/Users_Guide/appendixC.rst

@@ -924,14 +924,14 @@ MET produces hit rate (POD) and false alarm rate (POFD) values for each user-spe
 
 A ROC plot is shown for an example set of forecasts, with a solid line connecting the points for six user-specified thresholds (0.25, 0.35, 0.55, 0.65, 0.75, 0.85). The diagonal dashed line indicates no skill while the dash-dot line shows the ROC for a perfect forecast.
 
+A ROC curve shows how well the forecast discriminates between two outcomes, so it is a measure of resolution. The ROC is invariant to linear transformations of the forecast, and is thus unaffected by bias. An unbiased (i.e., well-calibrated) forecast can have the same ROC as a biased forecast, though most would agree that an unbiased forecast is "better". Since the ROC is conditioned on the observations, it is often paired with the reliability diagram, which is conditioned on the forecasts. 
+
 .. _appendixC-roc_example:
 
 .. figure:: figure/appendixC-roc_example.jpg
 
 	    Example of ROC Curve
 
-A ROC curve shows how well the forecast discriminates between two outcomes, so it is a measure of resolution. The ROC is invariant to linear transformations of the forecast, and is thus unaffected by bias. An unbiased (i.e., well-calibrated) forecast can have the same ROC as a biased forecast, though most would agree that an unbiased forecast is "better". Since the ROC is conditioned on the observations, it is often paired with the reliability diagram, which is conditioned on the forecasts. 
-
 Area Under the ROC curve (AUC)
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~