This debate again . . . . . the folks who say it flat out does not work or is a myth still dont understand how it is actually used in the real world.
Lets start by agreeing. If you have the option to get one anchor (of a good design suited to the particular bottom) which is the combined weight of the proposed tandom . . . the one anchor will win the majority of the time. (as thinwater alluded to) The major exception is on a flatish rock/coral bottom where no anchors are going to actually 'set' but two anchors have more chance of catching on something.
However, that is really not the real world circumstance where randoms are considered/used. Out there we have the anchors we have. We are not running to the store to get a bigger one. So we have let's say 4 anchors on board, and we have a
forecast for
hurricane force winds. The practical question is then how best to deploy the assets you have. I personally have found (and see on other boats) some value in the tandom
concept in two particular circumstances. One is when you are shore tied or
med moored, especially in a narrowish harbour (like in Chile). In that case, I would sometimes put my best all chain rode all out plus some
rope (so like 400' of rode when I am in only like 15' of water) with tandoms on the end with a good distance (perhaps 50' of chain) between the tandoms. You are hopefully shore tied against the direction of the strongest wind but if the low might go north of you, this might save the boat. In the other case, in say a pretty big Labradorean or
Iceland harbor, with bald land and essentially zero wind protection and no shore tie possibilities, I would choose to put my 4 anchors out on two tandom rodes, again with huge (as big as possible scope) in a V against the
forecast wind direction.
In both those situations I found the tandoms better than the alternative which is to put out the multiple anchors on separate rodes. My experience with that was you often pull on one and then the other and almost never get a balanced pull on them both.
The other thing to realize is that the wind is very very infrequently steady/constant. It usually has significant variation, gusts and lulls, so theory or tests using a steady load/pull are not really a very good indication of actual effect. The variation tends to favor the tandom over the alternative, giving the two anchors more opportunity to engage. A tandom is never going to be worse than the furthermost out anchor, at the very very worst the closer anchor will act as a high drag chum. (edit: all this does assume it is done in the correct way).
Again, none of this is as good as the even/ever bigger
single anchor (and yes, several of our anchors were quite sizable) . . . . but that is not the question. The question is what to do with the anchors you actually have on board when the forecast comes thru.