2017 Special Blend Projections for Fantasy Baseball

Preparing for your fantasy baseball draft requires using the best data available. If you use bad data then you’ll make bad draft decisions. It’s as simple as that. The core piece of data behind any good fantasy baseball draft prep is the type of projections that you use. I’ll do a rundown soon about which of the available projections are most accurate (I gave you a hint in my recent tweet here) but, first, I’m releasing my 2017 projections out into the world.
2017-Special-Blend-Projections

The thing that is unique about these projections is that they pull from a variety of other free projections sources that are available and strategically combine them into a super projection. Back in my early days of doing fantasy baseball research, I quickly found that averaging the available projections each year will result in a very accurate projection model. I later found that some projections were better than others at certain things. This resulted in an experiment with combining the projections but providing certain weights to the source for each individual stat, based on which projection was best at projecting it.

For instance, ZiPS has historically done the best job of projecting stolen bases and CAIRO has done the worst at it. So, for 2017, I took all of the stolen base projections out there then put the highest weight on the ZiPS stolen base projection while lower weights on Steamer and Clay Davenport’s projections and leaving out CAIRO altogether. When all is said and done, these weightings generated my optimized stolen base projections for 2017. I then moved on to the next stat and kept going until I’ve projected all of the stats for hitters and pitchers.

I’ve found that this way of combining the projections has been very accurate and much more accurate than simply combining the projections without weights. Each year, I learn a bit more about what works and what doesn’t. After four years of weighting projections in this way, I continue to adjust and improve. For instance, this year I’ve decided to start incorporating previous year stats and x-stats into weightings this year. They are used sparingly but still have value in increasing the projection accuracy at times.

In the 2017 version, when all was said and done, the following sources are used:

If those sources did not exist, my Special Blend would not exist so many thanks to them for all of their hard work in generating their initial projections.

The 2017 Projections

I’ve embedded the list of 2017 projections below but you can see a full view of them by accessing them through the Google Doc here. From the Google Doc, you can also download them and do whatever crazy things you want with them. When looking through these, keep in mind that there’s a separate tab for Pitchers and Hitters. There’s also columns for the calculated WERTH roto value of a player based on a standard 5×5 roto league format. The WERTH value is a z-score that calculates how far above or below average a player is in the main roto stats. If you want custom WERTH values for your specific league settings, check out my Excel cheatsheet where you can input your league settings.

Frequently Asked Questions To Address!

YES, these will be added into my cheatsheets very soon. The next official update of my cheatsheets will include these. I will announce on here and on Twitter when they are available.

YES, there are some players missing from here. Carlos Carassco comes to mind and many closers too. If any of the six main projection sources are missing a projection for a player then I will be too. The source that has the least amount of projections is the Fangraphs Fans (so go there and add your own projections maybe so I can have more projections!).

YES, I will be continuing to update these until Opening Day.

YES, I like piña coladas and getting caught in the rain.

Anyway, let’s get on with it!

Projections last updated on 03/28/2017

You Might Also Like

13 Comments

  • Reply
    Randy Munro
    March 13, 2017 at 8:24 am

    Just curious behind the reason ave, obp, slg and ops are rounded to the nearest 10? Has it always been this way and I just noticed now?

    Great work and cheers,

    Randy

  • Reply
    steven
    March 13, 2017 at 10:06 am

    merry baseball christmas! thanks, santa luke.

  • Reply
    Kevin
    March 14, 2017 at 6:03 pm

    Is there a way to import your special blend projections into an older version of your spreadsheet? Our draft has already started and I’d like to update the projections without having to start a new sheet. Thanks!

  • Reply
    Davith
    March 20, 2017 at 7:45 am

    Is there a place I can enter my leagues point values into your special Blend projections for 2017?

    • Reply
      Luke Gloeckner
      March 28, 2017 at 2:10 pm

      Sorry, I saw this way too late. For future reference, you could have downloaded the points league Excel spreadsheet I publish on the site to get point league values with these projections.

  • Reply
    Dan
    March 26, 2017 at 2:49 pm

    I think this projection may be missing some pitchers, including some higher performers (e.g. Carrasco, Colome, D. Robertson, Sam Dyson, Cody Allen)

    • Reply
      Luke Gloeckner
      March 26, 2017 at 2:54 pm

      Correct. I make a note about that in the article above in the FAQ part. If you use my Excel cheatsheets, I have a version in there that puts Steamer as a replacement for those cases.

  • Reply
    Brandon Ragaller
    March 28, 2017 at 1:32 pm

    Brandon Phillips not even on the Reds, hate this know i don’t trust you really do update this! Sad

    • Reply
      Luke Gloeckner
      March 28, 2017 at 1:35 pm

      Haha. Sorry, man.

    • Reply
      Randy
      March 31, 2017 at 5:20 am

      If you are serious Brandon, you seriously need to get over yourself. Luke puts so much time into this for free. You don’t like it, go elsewhere. People like you make the world a worse place.

      • Reply
        Luke Gloeckner
        March 31, 2017 at 12:03 pm

        Thanks for saying what I was thinking, Randy 🙂

  • Reply
    JC
    March 30, 2017 at 11:55 pm

    I’m trying to update the custom spreadsheet for standard draft settings. My league uses OBP instead of avg, and also uses slugging %. We use W, QS, SVHD’s, era, whip, k’s. Wondering what values to put in for obp, slug, saves+holds? I would really assume OBP, but it says invalid value every time. I just stumbled onto your site tonight, and I’m really digging it so far.

    Nice reads so far even if you never see this. Thanks.

    • Reply
      Luke Gloeckner
      March 31, 2017 at 12:04 pm

      Hmm, yeah, yous hould choose OBP, SLG, QS and Sv+Hold in the sheets to use those categories. I’ll check it out to see if I can recreate the problem. Which sheet is this? The roto draft sheet?

    Leave a Reply