Steamer Projections is a projection system for baseball player statistics. The system was created and is maintained and operated by Jared Cross, Dash Davidson, and Peter Rosenbloom. Dash and Peter are both seniors in college and are former students of Jared – a high school science and statistics teacher. The three creators share a deep passion and interest for baseball and the statistics that are intrinsic to the game.

The project began in the Fall of 2008 at Saint Ann’s School in Brooklyn, New York. Saint Ann’s offers its students an independent research program where they may embark on collaborative projects motivated by personal interest together with a teacher. The Steamer Projections system is the result of such a project.

What began as an explorative academic pursuit soon became a passionate and time-consuming endeavor as we quickly began to expand and improve our system – and increase the requisite amount of work – as our skill and knowledge of the intricate process of statistical projection improved.

Now we are in the midst of working on the sixth iteration of our system, which has released a full set of projections for MLB players for each of the past three seasons by the end of March. We are still constantly tinkering on and tweaking Steamer Projections in the hope of making it ever more accurate and efficient. You can find our full pre-season and daily projections at www.fangraphs.com

We hope you enjoy our system.

Warm regards,

Jared, Dash, & Peter.

September 2013

jcross@steamerprojections.com

dash@steamerprojections.com

peter@steamerprojections.com

hey should i drop james loney for mark trumbo?

I looked at your projections before the 2012 season for the 1st time and enjoyed them, but the more I read about your work, the more excited I am for the 2013 season to roll around. Keep up the amazing work guys.

I’m in complete agreement … can’t wait until 2013′s are available. Perhaps the most useful projection set I’ve seen yet!

yes, yes you should…

Think your projections are great…having trouble determining what stat is being used to drive the order in which players appear on the SS…random?

Have you considered setting up a system so that those of us that value your work can donate to the cause? I don’t hesitate to pay for projections, but I still value steamer above all others on draft day. I love that you’ve decided to open source your work, but think there are many of us that would jump at the chance to throw you guys a few bucks any way.

ok, ok…. you twisted my arm, I added a donate button on the sidebar.

Seriously, though, many thanks I really appreciate the sentiment.

Hi – question that would be of infinite help…

In March, you posted this: http://steamerprojections.com/blog/the-rosenheck-factor/

How did you associated IFFB% and SwStr% to BABIP and ERA? Another way to ask the question i guess is what is your new expected ERA and BABIP formulas (if you dont mind giving something like this out)?

From an expeted BABIP perspective i assume you use Staudes’: xBABIP = 0.4*LD% – 0.6*FB%*IFFB% + 0.235 .

…but what about expected ERA? How did you include the IFFB (or xBABIP in general) and SwStr into your equation?

…Thanks Infinitely.

have you ever weighed a stat more heavily from 3 years prior than from 2 years prior to optimize a projection? this is what i’m encountering when i seek to minimize the absolute value of the projection from the actual result. i don’t know whether to assign the weights like that or if i should toss out that result as an outlier not likely to be repeated in future years.