MASTER NOTES: Fantasy's least projectible element

You know what drives me crazy? The conventional wisdom that we really can predict the future. Now, I know that baseball has a nice structure, and that there are certain elements we can track and draw conclusions about. But we play our games with hard numbers, and drawing conclusions about general tendencies really does not cut it.

And even when we nail a player's projection, we usually haven't. Even with stable commodities. After Miguel Cabrera hit 44 HRs in 2012, we all expected some regression into the high 30s. That didn't happen, but because the error was in our favor, we didn't care that the projection was still pretty far off.

The fact that we routinely miss on 67% of first round players each year is met with a shrug. Last year, these critical foundation players that went belly up had names like Braun, Kemp, Pujols, Bautista and Stanton. These players cost us titles.

But here's the thing. We are still keeping these guys fairly high up on our rankings this year—though granted, not always first round—because we do recognize that they own certain skills. These are still talented players. The reason we blew their projections last year was not because they suddenly lost the ability to perform. The reason we blew their projections is because we messed up on the playing time.

Research has shown that 70% of the grossly missed projections each year—those that are off by $10 or more in roto value—are due to miscalculations of playing time.

Think about the monumental task it is to project how many at-bats or innings a player is going to get over 162 games.

It starts with skill. A player has to have the skill to merit the playing time.

But then we have to consider team context. Is there a spot for him on the depth chart? Are there other players on the team that could encroach on his playing time?

Now think about all the things impacting that:

He could get hurt. The ability to recover from injury is different for every player.

He could slump, and it's the manager's decision how much rope to give him before he's benched.

He could surge and merit more playing time, which in turn will cost someone else. It's a zero sum game, after all.

He could get traded to a team with a completely different depth chart puzzle.

And we have to project all these variables in advance for 162 games over a six month time span.

So when you ink in Miguel Cabrera for 575 AB this year, pat yourself on the back. You're wrong.

One way to deal with this uncertainty is to attach very broad expectations to each player's opportunity. We use the Mayberry Method here when projecting playing time. Each player is projected as either a full-timer, a mid-timer or a part-timer, assigned codes of 5, 3 or 1 when calculating their Mayberry score. We acknowledge that there will be a lot of volatility within those broad ranges and that helps us account for the impact of all the variables.

There's not much more we can really do about this, at least not for our full-season leagues. The daily games nail playing time, needless to say. You know who's in the lineup on any given night. Of course, you sacrifice much of the ability to project skill for a single game.

The one-month leagues I'm running at have a slightly better time of it. At least if a player gets hurt, you can decide whether he's worth rostering for a particular month. It's a more manageable time period, but it's not perfect.

The one thing that these games do teach us is that to gain any advantage in the projectibility of playing time, you have to do that analysis over shorter time spans.

We simply can't project six months' worth of variables.


Click here to subscribe

  For more information about the terms used in this article, see our Glossary Primer.