Telfair did show some promise during the last 26 games of his rookie season, after Cheeks was fired and Pritchard stepped in as interim coach. However, his increased production was mostly a result of increased minutes. I don't know of a site that lists PER splits, but if you just look at his percentages and per minute stats as a starter and reserve, you'll see that his FG% went up, but his 3FG% went down, as did his FT% (slightly). His PTS/36 actually went down, but his AST/36 went up slightly. His TOV/36 also went down as a starter.
Overall, since PER is adjusted based on playing time, I suspect his PER as a starter was slightly higher, but not much, than as a reserve. Also, I don't think any statistical improvement was due to coaching, but due to opportunity (increased PT) and natural growth (adjusting to the NBA game).
His PER his second season, under Nate, was still substantially higher than it would have been as a starter his rookie year. His PTS/36 was much higher (14.5 vs. 11.6), but his AST/36 was lower (5.4 vs. 6.3). His TOV/36 was also down under Nate (2.4 vs. 3.1) compared to as a starter under Pritchard. So, his AST/TOV ratio was better under Nate. However, keep in mind that PER adjusts for both playing time and pace, and as we all know, ever since Nate started coaching the Blazers his teams have been at, or near the bottom of the league in game pace.
Again, I don't know of a site that gives splits on advanced stats like PER, but given that he played over twice as many minutes as starter than as a reserve his rookie year, had a PER of 9.7 for the year, and his scoring (PTS/36) went down as a starter, I doubt if his PER as a starter would have been above 11.0. Most likely, it was somewhere between 10.5 and 11.0. Actually, if you look at his PER 36 numbers as a rookie starter, you will see they are very similar to what they were last season in Minnesota when he had a PER of 10.4. Under Nate, in his second season, Telfair's PER was a career best (by a substantial margin) 13.0.
BNM