Reexamining the DRIP

I have been DRIPing all my dividends, based on the fact that positions grow with no new commissions. I pay about $8 per trade with my broker, so most trades are costing me about 0.2-0.3%, as my average purchase is in the $3000-$5000 range. At the rate I contribute to my qualified plan, that allows me to purchase a new equity every other month or so. However, dividends are coming in about $2000/month. If I weren't doing the DRIP, I'd have twice as much to deploy on discrete purchases. The question is, am I smarter than the DRIP? DRIP means the entire portfolio grows at about 3+ percent per year, which is my average dividend yield. Were I to harvest dividends and use them to make discrete purchases, I'd need to do better than the average of my entire portfolio. Since valuations are up, there aren't many bargains out there. That might argue on the side of targeted purchase, as the DRIP program is purchasing in that environment. On the other hand, those few bargains will be companies whose stocks haven't moved with the market, forcing me to rely on my own analysis or the opinion of people I read on whether stocks that aren't moving are in fact good investments. My few non-dividend paying stocks don't see any position growth and that hasn't stopped me from holding them. Not being a momentum investor, I don't purchase more of them when they are rising, so I behave pretty differently with them than the DRIP program with dividend paying stocks.
Perhaps I should be purchasing the dividend-payers when they are down, rather than relying on the dollar cost averaging of the DRIP. I might be better at keeping average dollar cost of my purchases lower than the DRIP.
My portfolio software assigns no cost to the stocks purchased with the DRIP. This seems wrong, as If I took the cash and then purchased the stocks, it would register at the current price plus commission. On the other hand, I'm not putting new money in for these stocks, so this argues for the accounting as it currently stands. I guess it makes sense that internally generated returns should be handled differently than new cash contributions to the portfolio. I'm not quite ready to turn off all the drips. Perhaps it's better to sell a piece of the best performing stocks and reinvest in those with lower valuation and higher dividends. That would simply be a form of "rebalancing".
I think I'll go sleep on this one.