ENSO and the AO are only weakly related to one another, at best. While the northern winter ENSO state is fairly well predicted (since it is often a continuation of what exists in the autumn), the AO is poorly predicted, and knowing ENSO does not help much at all in predicting the AO. The AO is considered somewhat of a wild card as far as a winter temperature forecast is concerned, even though we know what its impact on eastern US temperature would be if we just knew what it was going to do. Some degree of predictability of the winter AO (or NAO for the portion of the AO that affects the eastern US) has been claimed in some research by Judah Cohen, using the amount of snow cover over Eurasia during the preceding autumn. He actually issues forecasts each year for AO/NAO based on this snow cover. I don't know that these forecasts have a proven track record, though. You might want to look up his journal publications. The phase of the AO/NAO has a greater effect on eastern US temperature than ENSO, so it is a big deal in terms of contribution to the uncertainty of the temperature forecasts. When AO/NAO is is positive phase, much more the eastern US is above average in temperature during an El Nino than when it is not, in which case the area of above average temperature from the El Nino is limited to regions farther north, such as northern New England and other areas currently forecast to have a tilt of the odds toward above normal in the current climate forecasts issued by Climate Prediction Center.