[Update: I’ve reached a partner at the fund, who requested anonymity because of securities laws governing public statements while fundraising, but he says the PEHub story is bogus. “We’ve never talked to anyone about an $800 million fund, in print or verbally.” Instead, VentureBeat has learned the firm may eventually raise a fund, but the amount will be much smaller than that. I’ve made an update below, clarifying the record on DAG’s investments. ]

DAG Ventures, the venture capital firm that became the butt of jokes a few years ago for its strategy of investing in companies that had previously gotten funding from top-tier venture firms, is reportedly raising an $800 million fund to make new investments.

The news, broken by PEHub, is surprising because DAG raised a large $700 million fund as recently last last year, which followed a $325 million fund in 2006. Usually, venture firms take a few years to breathe before raising their next fund.

I’m trying to reach DAG for more information. But the Palo Alto, Calif., DAG Ventures has been on a tear, pouring money into companies after they have previously been backed by top firms like Sequoia, Kleiner Perkins, Accel and Benchmark (recent investments include Friendster recently, Pacific Biosciences, ODesk, SearchMe, Pelago). This is all after DAG started with a small $60.3 million venture fund just four years ago. In other words, it’s engaged in a huge ramp up in size, at a time when most other Silicon Valley venture firms are feeling queasy about the investment climate.

It’s easy to scoff at DAG’s strategy of following other firms (it gained the nickname “Coattail Ventures”), especially in proud Silicon Valley, where the tradition among VCs is to try to foster an image of being able to discover and then anoint promising entrepreneurs. The strategy is somewhat unique. Following means DAG has to invest in companies that are more mature, and thus it pays a higher valuation (not such a good deal for DAG). On the other hand, gradually, the firm has won some respect among its peers, in part because it has been able to access some of the best companies, and has had the appearance of a steady focus at a time when some other firms have fallen apart. And despite appearances, we’re hearing DAG has not blindly followed these other firms; it has in fact passed on several deals brought to it by the top-tier firms, in cases where the deals didn’t look attractive.

PEHub suggests the firm’s successes so far haven’t exactly been home runs.

In 2005, SanDisk acquired 3D integrated circuit maker Matrix Semiconductor in a deal valued at $250 million; Matrix had raised $175 million from eight firms, including DAG and Benchmark Capital. DAG also made out well when online video-sharing site Grouper sold to Sony Corp. for $65 million in 2006. Grouper had raised $3.75 million from DAG and another firm after raising $1.5 million in angel funding. And earlier in 2006, Verisign paid $30 million net of cash to purchase CallVision, an Internet billing and customer relationship management company that had raised $5.8 million from DAG along with three other investment firms.

Update: PEHub appears to have missed some others: DAG invested in Zimbra, which was sold for a considerable profit to Yahoo for $350 million, after only $30.5 million in. It also made money on the following deals: Oakley Networks, which was bought by Raytheon for $193 million; Plaxo, which was bought by Comcast for a reported $150-$170 million; and Trapeze, which was bought by Belden for $133 million (DAG made money because it invested preferred capital, with a multiple liquidation preference, meaning it got its money back, and then some, even though other, earlier investors lost money). In addition, keep in mind that the firm only started investing four years ago. Any good venture firm avoids trying to sell its best companies early, since those are likely to emerge into large profitable deals where waiting five or six years is best. Google, for example, is a good example of a company — albeit not a DAG investment — that was held back, but which eventually produced an enormous exit.)