Last year I debuted my fantasy baseball projection system that liberally takes from a variety of expert projection sources. It was essentially a pilot year for the Special Blend projections which went quite well. The idea behind them was to combine the best advanced projection sources out there but weight them properly in each individual statistic based on which projection system performs best in that stat.
The sources that go into this Special Blend are:
Each statistic within the projection weights those projections differently but each of those sources brings something to the table in the final product here. In the debut year, the Special Blend projections provided accurate hitting projections that ranked ahead of all other sources in my analysis. However, in the pitching arena, the projections couldn’t outperform Steamer and the Steamer+Fans projections so improvements needed to be made for 2014. As a result, I made a few tweaks to my formulas and I’m hoping this year’s version of the projections will prove to be even more accurate for fantasy baseball purposes.
Browse through the projections below but I recommend going directly to the Google Doc for a closer look and to download them for your own purposes.
Updated: 03/08/2014
You may notice there that some players are not included in the projections. This will happen if one of the projection sources did not provide a projection for a player which is often the case with rookies or players coming from foreign leagues that will be making their MLB debut this year. All sources involved here must have a projection for a player in order for this special blending to work.
Greg Baumler
03/02/2014 at 6:54 PMLuke are these part of the latest spread sheets? Or will these updated ones be in your next update?
Luke
03/02/2014 at 6:59 PMActually, just half an hour ago, I uploaded new cheatsheet updates that include these (as well as updates to the MORPS projections for those interested in that)
Greg Baumler
03/02/2014 at 7:09 PMOK so would I have to redownload the cheatsheets to get these updates, correct?
Luke
03/02/2014 at 7:11 PMYup, gotta go back to the download links listed here: http://www.mrcheatsheet.com/2014/02/2014-cheatsheets-3.html
Greg Baumler
03/02/2014 at 7:13 PMAwesome thanks Luke
Mike
03/06/2014 at 5:14 PMJust a thought, but have you considered OneDrive.com for your document hosting? I am not an MS fan, but Excel is a necessary evil and I believe OneDrive is easiest to work with for sharing Excel.
Mike
03/06/2014 at 5:33 PMMasahiro Tanaka,SP appears to be missing (unless you don't think he is top 500)
Mike
03/06/2014 at 5:59 PMAlso want to point out that many players names are duped in the google document (Gattis, Span, Tommy Hunter, etc.)
Luke
03/07/2014 at 2:22 AMThanks, yeah, I need to fix that. Good catch.
Rich Upson
03/07/2014 at 6:19 AMAny input how to enter your xQS formula into excel? Upon entering the last step the formula produces a negative number.
Luke
03/07/2014 at 2:32 PMSo, yeah, it's tricky depending on how many starts they're projected for. If they're a releiver or something with 0 projected starts, you'll get funky numbers. So I usually add in an IF statement saying to ignore results lower than a certain number than have the formula nestled into the IF statement: ((IP/GS)/6.15) – (0.11*ERA))*GS
Luke
03/08/2014 at 4:27 PMFYI – Got the duplicates fixed and a few more players added in.
Ben Gellman-Chomsky
03/08/2014 at 7:06 PMLuke, any thoughts given to projecting holds? Great work here!
Luke
03/08/2014 at 7:41 PMI'd love to but unfortunately there's not a lot of those projection sources that project Holds in order for me to come up with a good mixture. Within my cheatsheets, I just use the MORPS projections for Holds no matter what the main projection source chosen is.
Ben Gellman-Chomsky
03/08/2014 at 10:51 PMFollow up question, Luke:
In terms of placing proper value on pitching rate stats in my league (K/9, K/BB, ERA, WHIP, BAA), I’m trying to weight my Z-values for those by innings pitched, so that somebody with more IP gets more credit to reflect that rate stat being used across a greater number of innings. When using Steamer projections, my formula for this has been to multiply the initial Z-value by (Individual IP/201), since 201 was the maximum IP in the Steamer projections (Wainwright). For example, a pitcher with a K/9 of 9.0 in 175 innings would be treated as though he had only a 7.835 K/9.
However, when I do this, top relievers like Kimbrel/Chapman/Jansen take massive hits in value; Kimbrel goes from being valued among the top 10 pitchers to being behind the likes of Hyun-Jin Ryu, Ian Kennedy and Matt Garza, while Jansen slips behind Scott Kazmir. I feel like this is too big a drop, but I’m not sure how to mitigate this without overbalancing. Thoughts?
Rhett Oldham
03/09/2014 at 10:40 PMLuke,
On draft day, I am considering using the WERTH value as my chief source even though experts and ADP may be different. Is that what you are doing in your drafts?
Jim P
03/11/2014 at 1:47 AMI am going to use this data and pretend that I came up with it on my own when I draft with my longstanding league.
I hope you understand.
David
03/25/2014 at 5:15 AMYou note that some players aren't in the projections and I understand why. Have you identified the relevant names that may not appear on your lists?
gperih
03/27/2014 at 2:07 AMare you planning an update to these projections prior to the start of the season?