Beating the projection systems

There is no one way to prepare for a fantasy baseball draft. Some won’t prepare very much at all, opting instead to trust the preset order the players appear in their draft room window. Others will ignore numbers and rank by reputation or what they remember reading about certain players. Others will buy a magazine and go purely by it’s rankings and in a similar vein others will take what a projection system spews out as gospel and create rankings off those numbers unaltered.

Personally, I like to use a projection system as a baseline for what I can expect a player to contribute, but I often find myself manually adjusting this batter’s home run total a little up or that pitcher’s ERA slightly down. This season what I’m going to do is see how helpful, harmful, or neutral my manual adjustments are by making 10 claims about different players’ stats now and seeing how often I was right at the end of the season.

The projection system I will challenge is CHONE, simply because it is freely available and is well-respected. If one of the players I make a prediction on misses serious time due to an injury, the prediction will be nullified.

All that’s left is to make the picks, which I will do over/under style.

The claims

One down. 23 more to go. (Icon/SMI)

1) Adam Lind will hit over his projected 24 home runs.

2) Brett Anderson will have an ERA below his projected 4.04 mark.

3) Michael Bourn will steal more than his projected 37 steals.

4) Brett Gardner will have a higher batting average than his projected 0.266 average.

5) Chone Figgins will steal less than his projected 33 steals.

6) Daric Barton will hit above his projected 0.261 mark (if he holds off Chris Carter long enough)

7) Ben Zobrist will have a higher average than his projected 0.268 one.

8) Chris Carpenter‘s ERA will be above his projected 3.10 mark.

9) Curtis Granderson will hit more than his 26 projected home runs.

10) Jimmy Rollins will hit above his projected 0.268 average.

And those are some of the changes I would make to what CHONE predicts for certain players. Obviously then, based on these claims I move certain players up or down a little in my rankings. By checking back at the end of the season on how my predictions fared, I can get a glimpse into whether I have positively or negatively altered rankings based strictly off of CHONE projections. Also feel free to make any of your own predictions in the comments and we’ll check back at the end of the season on those as well.

As a final note, here is a link to the article on the Yahoo fantasy sports homepage today offering mine and the other members of the Friends and Family Leagues’ thoughts on the draft, as a well a a working link to the full draft results.

The Incompleat Starting Pitcher
The end of the nine-inning start and how we got here.

Print This Post
Sort by:   newest | oldest | most voted

Am I missing something here, or does it seem silly to challenge a projection system on something like steals? Aren’t the accumulation of steals related to too many variables? I’d be curious to see how often a projection system is actually close to hitting the number of steals it suggests for a player.

It may not be the same, but couldn’t it be like challenging CHONE’s RBI or run projections?


“If one of the players I make a prediction on misses serious time due to an injury, the prediction will be nullified.”

CHONE doesn’t project PT, right?  To the extent it does, isn’t it regressing PT for the possibility of injury?  Seems like you’re setting yourself up for a win, particularly with respect to the “overs” on the counting stats.

Jacob Rothberg
Jacob Rothberg

Had this debate with a league mate recently, but what do you think of CHONE’s IP projection for Kershaw?


Chone seems to put a bigger factor then ever on age.  Carlos Gomez is supposed to have his highest OBP yet for example.  A number of players are projected to do the same if they are young.  I’ll take the under on Gomez’s .329.  Great post.  I agree that significant injury certainly should not be an excuse to toss out the prediction.  The concept of regression is founded partly on the fact of injury.