Every year fans get excited when players surprise them in spring training with terrific performances. And whether it's by a non-roster invitee or a rookie player, fans are always surprised on Opening Day when that player doesn't make the team despite outplaying his competition. These spring surprises are facing an uphill battle: No matter how well they perform, they have next to no chance at cracking the roster.

Take Ian Happ of the Chicago Cubs -- he has been on fire, hitting .452 with three home runs. He has no chance. Even though he might be outplaying Ben Zobrist, Albert Almora Jr. and Javier Baez in Arizona, he is not making the Cubs. The plan is for him to start the year in Triple-A, and no matter what he does in the spring, those plans aren't changing.

Similarly, the Houston Astros are really impressed with left-handed reliever Framber Valdez. He pitched in Rookie and A-ball last year going 4-5 with a 3.19 ERA and 79 strikeouts in 73 innings pitched. The Astros have been so impressed with him that GM Jeff Luhnow told me this week that he won't be surprised if he pitches in the big leagues for them this year. For a player not listed on anybody's prospect list, that says a lot. However, as much as the Astros have been impressed, they're not changing their plan for him out of spring training.

The reason for this is simple: Most general managers go into spring training with a blueprint of what his roster looks like from Nos. 1-25, as well as Nos. 1-32 (preparing for the first two months of injuries). This plan is based on the players' performances the previous year. There might be some competition at a couple of positions and/or a few roster spots, but that's usually about it. For top prospects, the organization has a specific plan on where they want them to start the year. Most non-roster invitees (NRI) are there more for injury insurance than for having a serious shot to make the team.

"As far as the young players go, I always tend to lean toward having a plan for timing and ETA, then stick with it," Seattle Mariners GM Jerry Dipoto said.

"This is my sixth season as a GM and all five of the previous Opening Day rosters have included at least one NRI," Dipoto said, but even that's part of his plan. "Assuming the veteran NRI has a niche on the 25-man roster, he simply needs to show that he's healthy with applicable skills and that his track record is still relevant. Many veteran NRIs are just like having players numbered 41-plus. It's more of an administrative or flexibility issue to bring them in as invitees.

"In our current camp we have a fairly set roster, barring injury. We've had a pair of standout rookie NRIs in Andrew Moore and Max Povse, but unless there's injury ahead of them both will start the season in the minors. We have one non-roster invite, Mike Freeman, who is competing for our utility infield job. We have 40 on the roster, but if he's the best guy we will make the adjustment when necessary."

General managers don't evaluate players based on spring training performance because the quality of competition is not the same as it is in the regular season. Hitters are facing pitchers that will start the year across every level of the minors, or major league pitchers who haven't built up arm strength, or not throwing their full repertoire or who haven't found their command or release point. Some pitchers have success because they're facing only bench players or minor leaguers or because the regulars don't have their timing down yet.

"I'm very cautious and conservative with spring training evaluations as it pertains to performance," Kansas City Royals general manager Dayton Moore said. "The matchups are simply different once the regular season begins. Competition in camp is simply determined based on previous season's performance and evaluation."

So what does a player (rookie or NRI guy) need to do to change your mind and change your plan as a general manager?

"As a member of a front office, I try to stay disciplined to not over-evaluate spring training performance," Minnesota Twins GM Thad Levine said. "Intuitively, we know that 10-15 innings pitched or 40-50 plate appearances do not represent a significant amount of performance from which we should make meaningful changes.

"That being said, when a player's performance is married with work ethic, character and a clear sense that he will enhance the team's chemistry, the temptation to alter plans becomes real."

Such an instance occurred five springs ago.

"Sometimes -- Mike Trout in April '12 is a great example -- the young guy knocks the door down and you simply try not to get in the way," Dipoto said.

Then again, this is easier for the best player in the game than it is for anyone else. During my 15 years as a GM, it was really difficult for a player to force me to change my plans. The few times it happened involved relievers with sizzling heat, strong secondary pitches and improved command that had me convinced that they had figured it out.

How infrequent was it? Right-hander Hector Carrasco comes to mind. In 1993, he pitched in A-ball with the Marlins' organization with an inflated ERA of 4.11 in 28 games started. Our plan was for him to start the year in Double-A as a reliever. However, when he got to spring training he was so dominant and better than any reliever I had in camp, that I wasn't afraid for him to make the jump all the way to the big leagues. He rewarded that decision with a 2.24 ERA in 45 appearances.

The only other player I remember who changed my mind in spring training was Scott Williamson back in 1999. He had gone from rookie ball in 1997 to Double-A and Triple-A in 1998, and then made the major league team out of spring training in 1999. His stuff and command were both too good to not take him north and he finished that rookie season with 12 wins, a 2.41 ERA and 19 saves. After 15 years as a GM, those are the only two players that I remember altered my thinking during spring.

I remember when I was a GM and had an outfielder named Adam Hyzdu, who had a great spring training back in 1995 leading our spring training team in both hitting and power. I explained to my owner, Marge Schott, that we were going to cut him. She told me I couldn't cut him because he was a local boy from Cincinnati who had gone to Moeller High School and had outplayed all of our major league players based on stats and performances in spring training.

I had to explain to Schott on each of his home runs how they were almost all on 84-87 mph fastballs, most down the middle of the plate belt-high, and that when the regular season started he wouldn't see those pitches and the performance will be the complete opposite. I showed her the results he had against good velocity and breaking balls, which was backed up by a sub .200 average. She didn't quite understand, but that's another story for another time. But it's just another example of why it's so hard for a player to make the team out of spring training even if he does well.

Bottom line, while there will be a few surprises on Opening Day rosters -- because of trades, injuries, players who got older faster than expected or that rare prospect who made a key adjustment -- rosters were mostly set when the teams reported to spring training.