Whenever we talk about polls these days, we have to include a huge caveat - that caveat being that the polling has been really godawful in recent races. Just unmitigated garbage. So feel free to throw all this out and consider it not at all.
That having been said, there are now two polls that show that Cruz is creating distance between himself and Rubio in Michigan, and is setting up a battle for second place with John Kasich. The first of these is a Fox 2/Mitchell Poll that shows Cruz with 20%, Rubio with 9%. The second is a Monmouth poll that shows Cruz with 23% and Rubio with 13%.
There's reason to believe that, if anything, Cruz will overperform these polls and Rubio will underperform. The first is that Cruz has pretty consistently beaten his polling due to having a better campaign organization. I don't know if that's true in Michigan, but it has been in states past. The second is that Saturday was a very good day for Cruz and a bad one for Rubio, and these polls have not really had time to reflect the momentum factor (in fact, the majority of the Monmouth surveys were recorded before the Saturday results).
So if... and we stress the if here... but if Cruz doubles up Rubio in Michigan, that is another piece of evidence that Rubio's goose is well and cooked. The argument for Rubio has always been that he could perform better than Cruz in non-Southern states with more moderate Republican voters - states like Michigan. If Cruz is trouncing him in these states, it's over.
And before you Kasich supporters get excited about possibly being in second place in Michigan, realize that your guy basically skipped Super Tuesday in order to camp there, and it looks like he might lose to the candidate in the field who is least ideologically suited to the Michigan electorate. The only difference between Rubio and Kasich is that Rubio at least might possibly be reasonable about his chances and Kasich will not, because he's deluded and/or actively helping Trump.