There are a linty of reasons why you cannot just divide the probability of victory in the general election by the probability of victory in the primary and get the conditional probability of candidate winning the general elections on their nomination. First, these values are both imprecise. Second, things change conditional on nomination (i.e., there is a complicated relationship between these two estimates). Third, candidates do not technically need to be the party nominee to win the general (although it is nearly 0% that any candidate wins the general should they not be a major party nominee). I could go on, but this simple rubric is suggestive of something.
As the likely party nominee, it is unsurprising that Hillary Clinton’s naive conditional probability of victory has increased slightly as the Democratic nominee’s (whomever that is) probability of victory has increased slowly over the last few weeks. As the last remaining establishment candidate and most moderate of the field, it is unsurprising that John Kasich’s value has been generally higher than the other Republicans. But, this is poorly defined, because for most of this period his probability of winning the nomination or general has been too low to precisely identify.
The most interesting thing is that Trump’s naive conditional probability of victory has been consistently higher than Cruz. Between these two leaders in the primary battle, the markets estimate that Trump is more electable. This far away from the election we know that hypothetical head-to-head polling (e.g., who would you vote for in a Clinton v. Trump match-up) is extremely worthless. Most voters are still not that invested and a couple billions dollars are yet to be spent! What the markets say is imprecise, but they are sending a very consistent message.