Let's look at the normal concentrations of palladium:
|slightly less energy from this beta plus decay|
|This is still good, but cadmium doesn't like to cooperate any further, the chain stops here|
We get no further benefit, as we can see from the isotopes of cadmium chart, no further beta decays are going to be available. Cadmium 112, 114, and 116 just aren't going to be useful because they are stable, and the one isotope, 116 that isn't has a very long half life.
If we follow this chain, we get several good reactions, but it will have to stop with the production of cadmium. However, this chain starts with an isotope that makes up only 1% of the total amount of palladium. What happens if we start at some other isotope. If we start at 104 and 106, we get a shorter version of this chain.
If we start with palladium 105, which makes up a significant portion of the sample, the reaction stops with the production of silver. If palladium 105 is overrepresentative in the sample, the potential is reduced. Likewise, if palladium 102 is underrepresentative, the potential is also reduced.
Palladium 110 doesn't do much good since it will end there at cadmium, as indicated above. Very short chain. That leaves palladium 108, which is shown below:
|Back to cadmium, it the chain will end here|
In summary, 102, 104, 106, and 108 all work, but you should start with 102, which is the most rare isotope. I think that is the whole point. The other chains are too short and most of the palladium isn't all that useful. If you just start with 102, you will generate the rest of the chain in sequence. Most of it isn't very good. You can use it, but not for as much energy production as you may have wished for.
You could remove palladium 105 and 110 altogether since they don't help. This makes up almost a third of the palladium in a normal sample. Over 1/2 of the palladium gives short chains.