I didn't see the Cat 4 race at Dawg Days, but a glance at the results and Siggy's pictures provides insight into the benefits of racing for a large team.
First is the large proportion of riders from two teams: NCVC and DVR. Both DVR and NCVC had 9 riders, totaling 18 of the 43 riders listed in the results.
Second is the proportion of riders from those two teams at the top of the results.
It's clear from the spreadsheet that the blue and red sit toward the top of the results list.
Consider the chart below. The horizontal axis shows placing for the 43 races; the vertical axis shows the rider (in this case, both NCVC and DVR have 9 riders). NCVC won 1st and 2nd; thus, their line rises for places 1 and 2. DVR's first rider was third, and its line rises. It rises again from places 5-8 and again 11-12.
If you don't follow all that, just think: shallow line=good. Random, evenly-distributed results would simply bisect the rectangle, as the black line (labelled "Linear Poisson") does.
In sum, the chart shows both NCVC and DVR's team effect. That is, their linear results run at a shallower angle than those of a mere random (or Linear Poisson) distribution (shown in the black line in the chart below).
This illustrates a simple point: riders on large teams did better than riders with small teams or no team presence. I think this is generally the case, especially on courses like Dawg Days.
Of course, Levi Leipheimer just won the Tour of Utah without teammates, and strong individual riders inevitably post results befitting their strength. However, the Tour of Utah is a climber's race, and teams provide less benefit in vertical races than they do in flat ones. This is true of MABRA races as well; compare the dominance of Harley in local crits vs. their less dominant performances in vertical road races.
If we wanted to get all "Freakonomical" we could look at the team effect for various teams and see which teams provide the biggest team effect for their riders over the course of a season. That effect would be, essentially, the slop of the line shown in the chart above--the flatter the line, the bigger the team effect. We could also weight podium places so that the lower placing of teammates who sacrifice themselves for results would be rewarded.
Why bother? Well, to me, the team effect is the best part of racing. It's what gives you that gloating feeling when you pass stronger riders who make poor choices and lack teammates. It's the element of smarts that makes bike races collective action games rather than simply chemical reactions and genetic exhibitions.