2015 Update

Cardinals · Post by **Cardinals** » Thu Mar 26, 2015 4:09 pm

I'm guessing we'll get our disc second or third week of April. Once we do, I'll set up a simming schedule ala the last few years until we're caught up. I haven't heard anything out of the ordinary, and DMB will produce the ZiPS disc again to the best of my knowledge.

Still haven't seen a second build of the ZiPS spreadsheet, but that's because the ZiPS man took ZiPS requests which totaled hundreds of players, so I guess that's lots of added work.

Giants · Post by **Giants** » Thu Mar 26, 2015 6:07 pm

Bren, is hiRes or however you spell it ready just in case?

Cardinals · Post by **Cardinals** » Tue Mar 31, 2015 2:55 pm

ZiPS man is sending the disc to DMB tomorrow via clown car. It's usually about a week turnaround time to when the disc can be purchased. Hopefully we'll be simming by next weekend.

Dodgers · Post by **Dodgers** » Tue Mar 31, 2015 10:55 pm

Fyi I'm out of town 4/4-12 so they'll be in oopss sometime after that.

Cardinals · Post by **Cardinals** » Wed Apr 08, 2015 5:48 pm

The disc will be available on April 15 from now on.

Hopefully we will begin simming by the 17th or 18th.

Rockies · Post by **Rockies** » Wed Apr 08, 2015 6:21 pm

FFS that seems like an eternity.. I'm ready for some baseball.

RedSox · Post by **RedSox** » Wed Apr 08, 2015 7:01 pm

There should probably be a discussion of whether or not we use the updates during the year.

Cardinals · Post by **Cardinals** » Wed Apr 08, 2015 7:34 pm

RedSox wrote:There should probably be a discussion of whether or not we use the updates during the year.

Agreed.

I'm not in favor of it, but it's certainly a discussion that ought to happen.

Cardinals · Post by **Cardinals** » Wed Apr 08, 2015 7:42 pm

Actually, I'm not sure we can even use it. I don't think it's possible to import projections into a database. By installing a midseason or postseason one, I'm assuming we would lose our standings and statistics to date.

RedSox · Post by **RedSox** » Wed Apr 08, 2015 9:41 pm

I'm with you. The ROS ZiPS stuff on Fangraphs is just garbage. Looks like a kindergartner subtracted the YTD data and then split the difference from the projection to the end of the season.

Reds · Post by **Reds** » Wed Apr 08, 2015 11:48 pm

RedSox wrote:I'm with you. The ROS ZiPS stuff on Fangraphs is just garbage. Looks like a kindergartner subtracted the YTD data and then split the difference from the projection to the end of the season.

Agreed... A waste of time.

BlueJays · Post by **BlueJays** » Thu Apr 09, 2015 8:32 am

RedSox wrote:I'm with you. The ROS ZiPS stuff on Fangraphs is just garbage. Looks like a kindergartner subtracted the YTD data and then split the difference from the projection to the end of the season.

It seems like that's all they do, so guys in our league would be punished for anybody who had a hot start to the real season.

Perhaps it does correct itself when a player actually does have a break out season, like I'm sure Corey Kluber out performed his projection last year, and maybe it updated to recognize he's a different pitcher than he previously was.

But for every instance of that, you get way more instances of guys like Lonnie Chisenhall who has a major fluke of a first half, and FG correctly predicts him returning to earth, so then whoever in our league has him ends up with FG predicting a crappy second half when they probably didn't get to enjoy his fluke of a first half.

Mets · Post by **Mets** » Thu Apr 09, 2015 11:29 am

I would be in favor of importing newly added (new projections) guys during the year. There is a way to do it that I know of without screwing up the season that's been started. This would be like rookie call ups, etc.

Reds · Post by **Reds** » Thu Apr 09, 2015 11:32 am

Mets wrote:I would be in favor of importing newly added (new projections) guys during the year. There is a way to do it that I know of without screwing up the season that's been started. This would be like rookie call ups, etc.

Seems like a good idea to me.

Rockies · Post by **Rockies** » Thu Apr 09, 2015 11:42 am

Mets wrote:I would be in favor of importing newly added (new projections) guys during the year. There is a way to do it that I know of without screwing up the season that's been started. This would be like rookie call ups, etc.

I can get behind NEW projections, but not altering the existing ones. If this is something to consider, how would we go about announcing or finding out about what player has a new projection?

Giants · Post by **Giants** » Thu Apr 09, 2015 12:50 pm

Fangraphs ROS projections are shiite, I had much more accurate (by which I mean profitable) results using the updated full season projections. [WARNING: DATA NERD TALK COMING]: While ROS projections might be more accurate individually (meaning a smaller variance between actual and predicted results) they are biased towards regression, which means that a true breakout has lag before its recognized. For an individual player thats not such a big deal, but if you're looking at whole teams that lag compounds and makes them less accurate for projecting team performance.

Totally down with adding rookies if John knows how to do that. Appoint someone to do it regular intervals (weekly? monthly?) and let's have at it if its not too much trouble to identify the new additions.

BlueJays · Post by **BlueJays** » Thu Apr 09, 2015 12:53 pm

I guess we could leave it up to the GMs to ask for specific players to be added?

Cardinals · Post by **Cardinals** » Thu Apr 09, 2015 1:07 pm

Giants wrote: Totally down with adding rookies if John knows how to do that. Appoint someone to do it regular intervals (weekly? monthly?) and let's have at it if its not too much trouble to identify the new additions.

Well no, the update would come at the All-Star break as provided by DMB.

Athletics · Post by **Athletics** » Thu Apr 09, 2015 1:25 pm

Orioles wrote:
RedSox wrote:I'm with you. The ROS ZiPS stuff on Fangraphs is just garbage. Looks like a kindergartner subtracted the YTD data and then split the difference from the projection to the end of the season.
It seems like that's all they do, so guys in our league would be punished for anybody who had a hot start to the real season.

Perhaps it does correct itself when a player actually does have a break out season, like I'm sure Corey Kluber out performed his projection last year, and maybe it updated to recognize he's a different pitcher than he previously was.

But for every instance of that, you get way more instances of guys like Lonnie Chisenhall who has a major fluke of a first half, and FG correctly predicts him returning to earth, so then whoever in our league has him ends up with FG predicting a crappy second half when they probably didn't get to enjoy his fluke of a first half.

Really, you got to point out my guy?

BlueJays · Post by **BlueJays** » Thu Apr 09, 2015 1:56 pm

Hey, I like Lonnie, and his slash line last year was solid: .280/.343/.427

But it was also the result of a .332 first half and a .218 second half.

Orioles · Post by **Orioles** » Thu Apr 09, 2015 9:32 pm

Yeah, I don't think we want to mess with updates for existing players. However, if there were some way to import the any newly added players and it doesn't require a ton of work, that'd be great. If not, we should just ignore the update.

Dodgers · Post by **Dodgers** » Mon Apr 13, 2015 9:32 pm

I'm not in favor of updating the projections and I'm not even sure it's possible without losing stats