Spring training is meaningful. Jobs are won and lost as a result. They should be.

Statistics from spring training exhibition games have no meaning. They should not be kept, reported or used in support of roster making decisions. I hope the Twins don't keep them or use them. Not only is the data in a sample too small to be meaningful but the level of competition varies.

It is critical the Twins get the roster right though. They need to rely on the skill of the coaching staff and other staff in spring training to get the best 25. They need to be able to assess whether Milone, Pelfrey and Nolasco are healthy and throwing the ball like they did when they were more successful. Suzuki can help assess Milone also. They have to make the correct assessment on these three and have plenty of time and plenty of eyes to get it right. They need to assess whether Hicks has fundamentally changed his approach and possibly swing. They need to assess whether Schafer's numbers the last two months were result of a fundamental change or simply variation due to sample.

Last year's staff should have seen a fundamental change in Kubel's bat speed. They should have seen Bartlett's inability to play outfield. They should have seen the need for Pinto to catch every day in the minors. They should have seen enough from Worley to retain him in AAA once he cleared waivers and was off the 40.

The Twins can't afford to miss again this year. They need the skill to correctly assess the pitching and bring the best 12 north. They need to be right on Hicks and Pinto and put them in a role where they will best help the team win games.

Spring training is critical to the Twins this year. Let's just hope they ignore the numbers and rely on their eyes. Let's hope the new Twins staff is more skilled than last year's staff.

Spring training is meaningful. Spring training stats are not. Let's not use them.