It seems like it would be far easier to just make the acetone halide and run a Friedel-Crafts which requires AlCl3 instead of palladium. Interesting paper though. It might even be easier to get palladium under certain circumstances.

friedel crafts is not "shit". not by any reasonable standard, especially if you consider the newer publications. but it does have its drawbacks. the Pd route is very promising since it is far more tolerable to aromatic substituents then friedel crafts (AlCl3 will dealkylate aromatic ethers, even though with proper conditions it is avoided), and it is catalytic. it is obvious, due to the high loadings of the ligand and catalyst (4 and 1%, respectively) that this combination is not ideal. it works well, but when you find the ideal catalyst, the loadings can be 0.1-0.01% Pd. look at the newer generation Buchwald-Hartwig aminations and the "Homeopathic, Ligandless" Heck reactions.

if you can afford the ligand (or make it, but youd probably a positive nitrogen pressure box to work under inert and dry atmosphere) and the catalyst and still stay economical this route is golden!

preparation of cinnamyl chloride:

1-Chloro-3-phenylprop-2-ene.5s Concentrated aqueous HCI (50 mL)
was added over 1 min to a stirred solution of  cinnamyl alcohol (21 g) in
T H F  (50  mL), whereafter  the  solution  became  heterogeneous.  The
mixture was vigorously stirred for  15 min and was then poured into water
and was extracted with CH2CI,.  The CH2CI2 was washed twice with
water, once with aqueous NaHCO,,  with water again, and dried (MgS-
04). On removal of the solvent, a yellow oil remained which was distilled
a t  45 mm  (bp 105 "C) through a  15-cm vacuum-jacketed  Vigreux col-
umn to give 20.4  g of  pure cinnamyl  chlorid

preparation of Pd[(pi-cinnamyl)Cl]2:

Hz, 1 H) ,  1.38 (s, 3 H ) ,  1.32 (d, J  = 7 Hz, 3 H) ,  1.20 (s, 3 H) . 
Palladium  chloride  (1.77 g) and lithium
chloride (1.77 g) were dissolved in hot water (2.5 mL) and then cinnamyl
chloride (6.2 g) in ethanol (20 mL)  was added. Carbon monoxide was
bubbled through the solution a t  25 OC  and after 15 min orange crystals
began to form and after a total of  3 h, the carbon monoxide flow was
discontinued and the mixture was stirred under carbon monoxide for 20
h a t  25 OC  and for a further 24 h a t  0 OC.  The product was then filtered
and was washed  with methanol followed by ether.  The orange powder
was  recrystallized  from chloroform  by the addition ether  (2.3 g, 90%).

ref:
Asymmetric Synthesis, Asymmetric Catalytic Allylation Using Palladium  Chiral Phosphine Complexes
Pamela R. Auburn, Peter B. Mackenzie, and B. Bosnich
JACS 1985, 107, pg 2033-2046