Is UCLA's Bench a Positive or a Negative?

Since the beginning of the year UCLA's bench has been a concern. How have the team's trio of reserves done in conference play? One partial way of making that assessment is to look at their +- figures for when they are on and off the court. Statsheet.com has +- data on players. So here is the +- data for our three reserve players (I calculated it based on raw data from Statsheet.com--there is data for 17 conference games for Welsh and Goloman, excluding the home game against WSU and 16 games for Allen, excluding the Arizona and WSU games which do not have +- records for Allen in those two games) The players' +- totals are calculated for the minutes they have played and then compared for the +- of the team when they are off the court. Here are the results:

So, for example , Welsh played 285 minutes with a +- of -29 and he was off the court for 395 minutes and the team was +72 for those minutes. If Welsh had played zero minutes and you extend out +72 at the same rate you would get +124 (So +52 more theoretically than if Welsh played 0 minutes) So the effect of Welsh playing would be a net negative of 81 points (the difference between -29 and +52 or almost five points a game). The same analysis with regard to Allen yields no difference (the team would be +15 with Allen off of the court if he played no minutes and he was +15 when he was on the court for those minutes). For Goloman, the team would have been +25 if he played no minutes so a net negative of 39 points (or a little over two points a game)

Obviously, there are limitations to looking at +- because of different line-ups used and it may be that there is more of an effect against good teams and/or home vs road, but with the possible exception of Welsh it does not look like we're getting hurt that much by the bench ( and to be fair to Welsh he is primarily coming in for Parker, one of our best players).