A currency strength meter ranks the major currencies by their recent performance across all their pairs, and the strategy buys the strongest against the weakest, on the theory that the pair combining them has the clearest trend.
The claim is that relative strength persists, so that today's strongest currency will still be strong tomorrow.
We tested currency strength rankings across 17 years of data with real trading costs, and found no edge. Rankings computed correctly, using only the data available at the time, did not predict the next period's relative performance. Versions of the idea that appeared to work turned out to depend on information the trader would not have had at the moment of the trade.
The reason is that currency returns over days and weeks have very little persistence, so a ranking of the last week says little about the next.
If you trade it anyway, be certain the meter is computed from closed data only, and count the results over at least a year.