You can have some idea of the variations in the seasonal forecasts, including if any individual runs were close to what verified, in CPC's attribution of seasonal anomalies. This analysis does not focus on a particular region, like the Southwest, but it does give a broader perspective. For example, if you go to the slides for this past winter and check out slide 24 (top left panel), you can see some of the individual ensemble members did pretty well at predicting the North America precipitation pattern, whereas other members did quite poorly (spatial correlations less than zero).

What's important to remember is that the variations among those 40 model members are due entirely to tiny perturbations in the initial conditions (like the effects of tiny flaps of a butterfly's wings), and that's enough to be the difference between a really bad forecast (spatial correlation ~ -0.4) and a decent forecast (spatial correlation ~0.4). These are tiny uncertainties in the initial state that we cannot possibly ever expect to capture in a seasonal prediction system, and that's why chaotic weather and climate variability will always be a limiting factor in our seasonal predictions (and why they always will be expressed in probabilities). This is the same idea in my blog post from last month. 

I am not sure if any model members captured the magnitude of the cold and wet anomalies in the West, but I wouldn't be surprised if some did. Actually, if you look at the bottom right of slides 26 and 28, you can see that the best 4 model runs do capture the wet and cold conditions in the West pretty well. Again, the only things that distinguish those 4 best model runs from the other poorer model runs are those tiny butterfly-flap perturbations in the initial conditions. So, I agree that it's important to question why the forecast did not verify where extreme conditions did occur, and we should (and do) look deeply into the forecast data, but these results already tell us that it's difficult to underestimate the potential role of seasonally unpredictable, chaotic weather variability.