The cross-tabs on this poll look weird.... overly high male sample, excessively high Republican sample, too many college grads, and too many people from metro Portland....
This seems like an outlier to me, unless additional polling substantiates these numbers.
Actually, the big partisan fault is low independents. The 2004 exit was 34I-34R-32D. Realistically, the Republican numbers are
very unlikely to be up on 2004, so 37R is obviously wrong, but 41D is probably high, too. (This is, of course, assuming that Oregon hasn't experienced a massive increase in partisanship in the interim, but I find that... unlikely.)