Halogens, except for fluorine, aren’t all that electronegative. Chlorine is just slightly less electronegative than nitrogen and bromine is even less again. Iodine almost has carbon’s electronegativity. Thus, there won’t be so much of a dipole moment created by substituting a hydrogen with a halogen. This applies even more if a hydroxy group is replaced with a halogen.
To illustrate this, take dichloromethane. Of all the chlorinated methanes it has the highest permanent dipole moment. Yet it is still almost immiscible with water but fully miscible with hexane. Note by the way that your data somewhat reinforces that: the molecule with the highest dipole moment of all your halocarbons, fluoromethane, has the lowest water solubility.
A second point is that halogens come with free lone pairs. Lone pairs are not polarised per se but can easily be polarised in London interactions. Thus, halogens actually increase molecule’s ability to interact with lipophilic, unpolar media by allowing stronger London forces.
Note also that while nitrogen is capable of forming hydrogen bonds with surrounding aquaeous media, chlorine — in spite of its similar electronegativity — cannot since its lone pairs are much more diffuse. Only fluorine is in theory able to accept hydrogen bonds. In principle, this is another consequence of the larger size of higher-period halogens.
In all this, fluorine is somewhat separate and special. But fluorine is special in itself since perfluorinated molecules are typically insoluble in anything but fluorocarbons. Fluorous extraction is actually a technique used by some green chem labs.
To sum it up, just looking at the simple variable polarity cannot explain the solubility of halocarbons adequately. Their size needs to be considered.
