The dagger-board is not particularly snug in the trunk when deployed--it rattles back and forth unless there's a current holding it to one side. When under sail this doesn't matter at all, as the lateral forces against the boat hold it firmly in place. Only at the moment of transition from one heel to the other will the dagger-board move at all when underway. At night in a slip or moored just jam a boat hook down into the trunk to keep it from rattling.
As for weight counteracting heel, of course it will, and the lower the weight is in the water the greater both the inertial moment (resistance to motion countering tenderness) and leverage (righting force countering maximum heel) will be.
Lever force is easy to calculate: A horizontal lever has weight * length force, so a 10 lb. arm with a 10 lb. weight at the end exerts 100 ft. lbs. of torque (turning force) at the pivot point opposite the weight. That calculation is easy. However, presuming you are not typically running at a 90 degree heel, the more useful formula is Force=length*weight*sin(heel angle). For a 30 degree heel, 5 foot length, and 50 lb. weight, the additional righting torque would be:
[Be sure DRG or DEG mode says "DEG" in the display and not RAD or GRAD]
[3][0][sin][x][5][x][5][0][=] 125lbs.
You can do this one easily in your head because sin(30) is 0.5 -- 50% the righting torque of a 90 degree arm. 45 degrees is 70%, and 20 degrees is just about 1/3.
125 lbs. of additional righting torque is nothing to sneeze at, but do remember that your DB is now 50 lbs. heavier. Doing that same math with the board 1/4 down (1.25 feet) provides only 31 ft. lbs. of righting torque--not much at all. And at low angles of heel such as 10 degrees with the board down, the weight will provide 43 lbs. of righting force--not a whole lot, and with the board up, no significant righting force at all. In fact, if the board is higher than the center of buoyancy when pulled up in the trunk (as it will be), it will promote rather than reduce tenderness (such as when slipped). For this reason, you want to put the weight in the DB as low as you can. The lead shot + resin method is almost certainly best, which you can search out on this forum.
The overall effect will be similar to having one additional (

) fat guy on the windward gunnels to counter heel (the lever arm is shorter so the weight has to be higher).
To prevent capsize on a Mac, you need to have 1300 lbs. of righting moment at 70 degrees of heel. (this would be equal to the water ballast). This translates to a 275 lb. lead bulb on the keel, which would essentially allow you to safely run your boat (board ALL THE WAY DOWN) with 1000 lbs. less weight on board. The problem here is that if the board is up, the Mac will blow down because the righting torque is only 300 lbs. So if you did this mod, you'd gain a lighter boat at the cost of being unable to pull up the board for reduced drag. Probably the lighter boat would provide better speed, but there are so many variables involved here that an empirical test would be easier than doing the math.
So yeah, 50 lbs. of weight on the board should have a moderately positive effect on heel, and consider that the effect magnifies as the boat goes farther over, so that by 50 degrees of heel you have almost 200 lbs. of righting torque.
The dagger-board currently produces righting force by hydrodynamic lift--it's like an airplane wing through the water, creating lift (force) that opposes heel. I don't know the shape characteristics of the DB so I can't calculate the lifting force directly, but my suspicion is that it's similar to the amount of torque that added weight provides.
Mostly its the water ballast countering heel.