Since I am not confident that either player will get more playing time than the other, I am going to simply average their projections... but first, a brief analysis of each player.
Kolten Wong is probably the safer bet to get more playing time. He is a near top prospect for the organization who's time has come. In a short stint last season, he did not impress. But it was not enough to base the future on, as Wong can probably muster much more offensive prowess than he displayed in 2013 at the major league level.
The consensus projection is an unimpressive 94 wRC+ but with some nice defense. Oliver thinks his offense will be rather lackluster, but thinks so highly of his defense to overcome the low offense rating (88 wRC+ bringing down his average projection to 94 probably) that they are projecting 2.4 fWAR. Steamer thinks much more lowly of his defense, pegging him just above average. His average fWAR projection is 2 fWAR, actually a higher projection than Matt Adams.
While Wong is certainly not going to tear the cover off the baseball, his defensive side shows enough promise that he can be a valuable player at second base. It will be interesting to see if the rookie can hold his own throughout a whole season, especially since Ellis is in the mix.
Wong being expected to play in 129 games by the consensus projection (using projections readily available at fangraphs.com) means that Ellis will get limited playing time. Mark Ellis is historically a great defensive player without much bat. His average projection is 105 games, so perhaps between giving off time to the rookie and defensive replacements (potentially?) he may approach that somewhat.
All of the projection systems available think that Ellis will get 85 wRC+, interestingly enough. His expected fWAR is 1.3. Averaging the two players together nets just about a 90 wRC+ player at the keystone. But with enough defense to boost it to well above 2 fWAR. I guess the oddest thing I noticed is that Oliver thinks Ellis will get 600 plate appearances in 2014. Now that would be unexpected.