Trump might think, however long to odds, Texas represents and opportunity to drive a stake through Cruz's heart.
Judging by the polling I've seen, which ain't been a whole lot, it is a long shot. What can change that? A blowout in Nevada where Cruz is a distant third. Another gaffe on the part of the Cruz campaign. A monumentally bad debate performance.
It could happen. I ain't betting against Cruz in Texas, but neither am I conceding a loss.
There are two pools of delegates. One is district by district, with 108 delegates. The other is statewide with 47.
The 108 is the pool that goes 2:1 in favor of a winner above 20% but under 50%. So, for talking purposes, give Cruz 2/3rds of 108, and Trump 1/3rd. Cruz 76, Trump 36.
The 47 goes proportional to ALL candidates getting more than 20%. Using the UT poll, Cruz gets 56% of the delegates (26), and Trump gets 44% (21).
Add 'em up, and one gets 102:57.
So, that rough cut 100:50 margin is close enough for talking purposes, but I think it misses the nature of the battle. Most of the difference between delegate awards comes from the per district pool, which swings 2:1 in favor of Trump for any district that he wins by as little as one vote. If Cruz and Trump "split the districts," they also "split the state."
Rubio has to get above 20% to get any share, unless there is only one statewide candidate above 20% and Rubio is 2nd in that (statewide) contest.