Draft Discussion - Fantasy Baseball Cafe 2015 Fantasy Baseball Cafe
100% Deposit Bonus for Cafe Members!

Return to 2013 Cafeholics Pond Scum

Draft Discussion

Moderators: chris8, ayebatter

Re: Draft Discussion

Postby Pogotheostrich » Wed Mar 13, 2013 11:38 am

rjforlife wrote:ESPN projects Verlander to win 20, CBS 19, MLB.com 20, Razzball 20.


Are they all foolish for projecting this many wins for him?


I would say it is optimistic to project 20 wins for anyone.
Image
Pogotheostrich
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicFantasy ExpertCafe WriterMock(ing) DrafterEagle EyePick 3 Weekly WinnerLucky Ladders Weekly Winner
Posts: 21623
(Past Year: 272)
Joined: 9 Dec 2002
Home Cafe: Baseball
Location: St. Louis

Re: Draft Discussion

Postby Ender » Wed Mar 13, 2013 11:44 am

Yes they are foolish and all of those are pretty bad sources for projections imo. Streamer has him at 17, ZIPS at 17, James at 18, Oliver at 20 on fangraphs. I projected him at 18 wins myself. I mean he has only won 20+ games once in his career and has averaged under 18 over his 6 full seasons, why would you be comfortable assuming he was going to again this year? Winning 20 isn't easy, you need things to bounce your way even if you are a great pitcher. I think he is very capable of winning 20, but he also can only win 17 like last year and he could get some minor injury and finish with less than 17. 20 wins for him is the everything goes right number, not the expected.
Ender
Hall of Fame Hero
Hall of Fame Hero

CafeholicFantasy Expert
Posts: 7733
Joined: 30 Apr 2004
Home Cafe: Baseball

Re: Draft Discussion

Postby Fenris-77 » Wed Mar 13, 2013 12:03 pm

Ender wrote:Yes they are foolish and all of those are pretty bad sources for projections imo. Streamer has him at 17, ZIPS at 17, James at 18, Oliver at 20 on fangraphs. I projected him at 18 wins myself. I mean he has only won 20+ games once in his career and has averaged under 18 over his 6 full seasons, why would you be comfortable assuming he was going to again this year? Winning 20 isn't easy, you need things to bounce your way even if you are a great pitcher. I think he is very capable of winning 20, but he also can only win 17 like last year and he could get some minor injury and finish with less than 17. 20 wins for him is the everything goes right number, not the expected.

I think you miss the context for some of those projections. Most sites have a rubric that informs their projections, and that isn't the same for every site. Some sites are better at projecting hitters than pitchers, or vice versa. Some sites like to use 'best case' projections for some guys too. My point is that every site does things differently. I think you're putting way too much emphasis on what the actual number is from one site to another.

We all know wins are a crap shoot anyway, just like batting average, so it's seems fruitless to get bent out of shape it (IMO).
Fenris-77
Major League Manager
Major League Manager

User avatar

Posts: 1688
(Past Year: 51)
Joined: 10 Jul 2008
Home Cafe: Basketball
Location: SAGNOF

Re: Draft Discussion

Postby Ender » Wed Mar 13, 2013 12:25 pm

Crapshoot is a pretty strong word. The year to year correlation on AVG is 0.34, ERA is 0.33. But even the strong categories like HR are only 0.61 and R and RBI are only around 0.50. More important is that luck neutralized AVG has a 0.50 or so correlation. If you understand the stats you can predict AVG about as well as you can predict anything but HR/SB/K. ERA is the really tricky stat to predict because all it takes is 1 bad game to drastically alter it, unlike AVG which takes a bunch of games.

I definitely agree you want to stick to one way of doing projections, whether that is using just one system or doing your own and taking the same general theme for all projections and that is the most important part. I mean using ZIPS for half your players and ESPN for the other half would be a disaster as an example. As long as you are consistent your values make sense. However to say that those projections are 'bad' is definitely incorrect.
Ender
Hall of Fame Hero
Hall of Fame Hero

CafeholicFantasy Expert
Posts: 7733
Joined: 30 Apr 2004
Home Cafe: Baseball

Re: Draft Discussion

Postby GardnerRunsRampant » Wed Mar 13, 2013 12:48 pm

rjforlife wrote:ESPN projects Verlander to win 20, CBS 19, MLB.com 20, Razzball 20.
Are they all foolish for projecting this many wins for him?


Personally I'd throw out espn and mlb.com.

I hate putting cbs stats into my projections given that they usually rely way to much on recent history, but I generally keep them in since they include 1B, 2B, 3B, BB, SO, and CS on the batting side, and QS, CG, and BS on the pitching side. These items become relevant in a couple other leagues of mine, and are hard to find in projections. That said, I know to tread lightly on their projections.

First year Grey Albright is putting projections out there that I know, but if you've read his blog he certainly is an addict :-D I included his stats in my report, but certainly he only includes the main categories for those in detailed leagues.

That said, make your own composite. That's what I do. The more trusted sources you can gather the less risk you'll have built into your system. You can also compare max/min/average's. See which player's you can count on, and which players have the biggest boom/bust potential.

FWIW: I've got Verlander at an average of: 18-7 with 228IP, 73ER, 226K
GardnerRunsRampant
Minor League Mentor
Minor League Mentor

User avatar

Posts: 362
(Past Year: 55)
Joined: 16 Feb 2011
Home Cafe: Baseball

Re: Draft Discussion

Postby jackie hayes » Wed Mar 13, 2013 12:49 pm

rjforlife wrote:Every single pitcher that pitched a full year last season is likely to pitch 10-20 less IP this season? That seems outlandish to me.


That's not true, but any individual pitcher that pitched a full season has some chance of having injuries, having a down year, etc. Whether you're looking at mean or median projections, that downside weighs heavier than the 250-inning, 2.10 era, 260k upside -- it's simply more likely. A full probability model would certainly say that someone will have a crazy great season, it just won't be able to say whether it's Verlander, Kershaw, Felix, Strasburg...Medlen... :-b
jackie hayes
Major League Manager
Major League Manager

Cafeholic
Posts: 2071
(Past Year: 65)
Joined: 19 Apr 2008
Home Cafe: Baseball

Re: Draft Discussion

Postby jfg » Wed Mar 13, 2013 1:06 pm

Funny, the projections I used only have one team over 262 homeruns. Edit: Oh never mind, I forgot to save the last two rounds.
Image
jfg
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicGolden Eagle EyeMatchup Meltdown SurvivorLucky Ladders Weekly WinnerTrivia Time Trial Monthly Winner
Posts: 6078
(Past Year: 82)
Joined: 4 Nov 2006
Home Cafe: Baseball

Re: Draft Discussion

Postby Fenris-77 » Wed Mar 13, 2013 1:40 pm

Ender wrote:Crapshoot is a pretty strong word. The year to year correlation on AVG is 0.34, ERA is 0.33. But even the strong categories like HR are only 0.61 and R and RBI are only around 0.50. More important is that luck neutralized AVG has a 0.50 or so correlation. If you understand the stats you can predict AVG about as well as you can predict anything but HR/SB/K. ERA is the really tricky stat to predict because all it takes is 1 bad game to drastically alter it, unlike AVG which takes a bunch of games.
I was really just indexing the way that AVG (and even moreso ERA) have those occasional big swings for some guys, whether due to injury or small sample size. Guy who hit homers will still launch a bunch of bombs even if their average goes into the toilet (Adam Dunn anyway). We're essentially agreeing I think. I do like luck neutralized AVG (or the idea pf it anyway) but it's not a stat I've spent a long time looking at in any detail. What sites do you gather data from when you're looking at it? I really only use Baseball Reference and fangraphs, and even then there are so many stats it's tough sort out the useful form the not, especially for fantasy purposes.

I definitely agree you want to stick to one way of doing projections, whether that is using just one system or doing your own and taking the same general theme for all projections and that is the most important part. I mean using ZIPS for half your players and ESPN for the other half would be a disaster as an example. As long as you are consistent your values make sense. However to say that those projections are 'bad' is definitely incorrect.[/quote]I don't recall saying the projections were bad, but perhaps you are talking about a different comment. :-? I did call them a crap shoot, but that wasn't to imply they were bad somehow, just less reliable than the counting stats (as you pointed out).

@Gardner - Grey's been doing projections for a while now (at least four or five years AFAIK)
Fenris-77
Major League Manager
Major League Manager

User avatar

Posts: 1688
(Past Year: 51)
Joined: 10 Jul 2008
Home Cafe: Basketball
Location: SAGNOF

Re: Draft Discussion

Postby stevethumb » Wed Mar 13, 2013 2:13 pm

http://www.baseballhq.com/special-repor ... ed-secrets

interesting coming from a guy who has made a living on selling #s
Image
stevethumb
General Manager
General Manager

User avatar
CafeholicEagle Eye
Posts: 4041
Joined: 2 Dec 2004
Home Cafe: Baseball
Location: cloud 9

Re: Draft Discussion

Postby J35J » Wed Mar 13, 2013 2:27 pm

stevethumb wrote:http://www.baseballhq.com/special-report-ron-shandlers-unpublished-secrets

interesting coming from a guy who has made a living on selling #s


LOL. ;-D

It's pretty much true though...well, no...it is true.
J35J
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicFantasy ExpertCafe WriterCafe RankerMock(ing) DrafterPick 3 Weekly Winner
Posts: 10467
(Past Year: 361)
Joined: 26 Jan 2004
Home Cafe: Baseball

PreviousNext

Return to 2013 Cafeholics Pond Scum

Who is online

Users browsing this forum: No registered users and 0 guests

cron
Forums Articles & Tips Sleepers Rankings Leagues


Get Ready...
The 2015 MLB season starts in 2:42 hours
(and 96 days)

  • Fantasy Baseball
  • Article Submissions
  • Privacy Statement
  • Site Survey 
  • Contact