Mixed Opinions - Fantasy Baseball Cafe 2014 Fantasy Baseball Cafe
100% Deposit Bonus for Cafe Members!

Return to Baseball Leftovers

Mixed Opinions

Moderator: Baseball Moderators

Re: Mixed Opinions

Postby kab21 » Wed Dec 31, 2008 6:00 pm

GotowarMissAgnes wrote:
noseeum wrote:
I disagree. The method to the calculation for playing time is arbitrary. Take Liriano. We know why he missed time the last two years. Is 92 innings a reasonable projection for this year? I don't think so, based on what we know of Tommy John surgery. Other projection systems take into account the injury proneness of players as well, but they also look at the individual circumstances.

Marcel's method falls flat for players that have significant injuries that keep them out for a long time.


And the problem is that those looks at individual circumstances may be just as biased as any other approach.

The advantage of Marcel is that you KNOW there is essentially no bias from the guy doing the projections.

And, I would argue that 92 IP is certainly a reasonable guess. 10-20 percent of guys returning from TJ surgey still flame out completely, so we have to factor that into the estimate. And Liriano has had other significant injuries, missing time in both 2002 and 2003. And, you never know what other factors might pop up.

Last year, for example, James predicted Liriano would pitch 152 IP. Marcel predicted 57 IP. He pitched 76 IP in MLB (200 overall), because the Twins kept him in AAA longer than expected.


The problem with the projected playing time is that Marcel isn't making a projection using TJ recovery data or anything else useful. It's projected playing based on his 3 previous seasons of data. If Liriano re-injured his elbow in the last game of the season and was going to miss 2009 Marcel's non-biased equation would still project 92IP. I think these projections are somewhat good for rate stats, but fall woefully short when projecting playing time. They are nearly worthless in that respect imo.

I think the bigger question in your original post is how do the rate stats compare between the different systems.
kab21
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicCafe WriterCafe Ranker
Posts: 5340
(Past Year: 170)
Joined: 24 Nov 2007
Home Cafe: Baseball

Re: Mixed Opinions

Postby GotowarMissAgnes » Wed Dec 31, 2008 7:11 pm

noseeum wrote:

If someone put out a projection system that integrated MLEs with Marcel, it would be better than Marcel, and it would prove to be so statistically, because MLEs have been proven to correlate highly with major league performance. James first introduced this in his 1985 abstract:
http://baseballanalysts.com/archives/20 ... rom_20.php

You can accept this or reject this, but it's still true no matter how you feel about it.


And I would suggest you go to Tango's site and read why he doesn't like MLEs.

I know all about the relationship between minor league and major league performance,and I have sited i tmany times myself.

But, the question is whether it can be appropriately added into projection systems and improve upon Marcel's simplistic approach. It's possible to take good data and use it in poor ways. And all I am saying is that it is an empirical question best tested by data, not by assertion.
"I don't want to play golf. When I hit a ball, I want someone else to chase it."
GotowarMissAgnes
Hall of Fame Hero
Hall of Fame Hero

User avatar
Fantasy Expert
Posts: 5516
Joined: 12 Dec 2003
Home Cafe: Baseball
Location: Happy Valley

Re: Mixed Opinions

Postby noseeum » Wed Dec 31, 2008 7:19 pm

Here's how PECOTA stacks up against other systems when they differ by more than .50 in their forecasts:
http://www.insidethebook.com/ee/index.p ... ations/#22

It bests Marcel about 60% of the time. CHONE comes closest to matching PECOTA.
noseeum
Major League Manager
Major League Manager


Posts: 1697
Joined: 1 Mar 2005
Home Cafe: Baseball

Re: Mixed Opinions

Postby noseeum » Fri Jan 09, 2009 1:24 pm

More info to consider when comparing projection systems. The best comparison is a player's performance against league average because they all project different levels of total offense for the entire league:
http://www.fangraphs.com/blogs/index.ph ... rojections
noseeum
Major League Manager
Major League Manager


Posts: 1697
Joined: 1 Mar 2005
Home Cafe: Baseball

Re: Mixed Opinions

Postby GotowarMissAgnes » Fri Jan 09, 2009 6:48 pm

Look carefully. Focusing on the pitchers where the systems disagreed the most, out of EIGHT projection systems, a simple stupid Marcel system bested FIVE, losing only to PECOTA and CHONE. Focusing on all players, that simple stupid Marcel bested SIX, losing only to PECOTA, and ONLY by a hair's breadth.

No fancy playing time adjustments.
No fancy system of dealing with rookies or using minor league data.

The problem is, as these results illustrate, many of those fancier projection systems have biases built into the assumptions made by the guy(s) doing the projections and the methods they choose to do that. And those biases are just as likely--maybe MORE likely--to lead to a bad assumption, a bad model, and a bad projection. Genius modeler Joe may thinks he knows how to adjust playing time for TJ recovery, or translate minor league performance into major league performance, but his assumption is built on a faulty interpretation of bad data. And as a result, a system like Marcel that just plugs and chugs data beats Joe.

More information is only better if it's GOOD information and if it's used WISELY. And be careful about assuming that you are the guy with both good information and wisdom.

So, tell me again which system is "nearly worthless"
"I don't want to play golf. When I hit a ball, I want someone else to chase it."
GotowarMissAgnes
Hall of Fame Hero
Hall of Fame Hero

User avatar
Fantasy Expert
Posts: 5516
Joined: 12 Dec 2003
Home Cafe: Baseball
Location: Happy Valley

Re: Mixed Opinions

Postby noseeum » Fri Jan 09, 2009 6:53 pm

GotowarMissAgnes wrote:Look carefully. Focusing on the pitchers where the systems disagreed the most, out of EIGHT projection systems, a simple stupid Marcel system bested FIVE, losing only to PECOTA and CHONE. Focusing on all players, that simple stupid Marcel bested SIX, losing only to PECOTA, and ONLY by a hair's breadth.

No fancy playing time adjustments.
No fancy system of dealing with rookies or using minor league data.

The problem is, as these results illustrate, many of those fancier projection systems have biases built into the assumptions made by the guy(s) doing the projections and the methods they choose to do that. And those biases are just as likely--maybe MORE likely--to lead to a bad assumption, a bad model, and a bad projection. Genius modeler Joe may thinks he knows how to adjust playing time for TJ recovery, or translate minor league performance into major league performance, but his assumption is built on a faulty interpretation of bad data. And as a result, a system like Marcel that just plugs and chugs data beats Joe.

More information is only better if it's GOOD information and if it's used WISELY. And be careful about assuming that you are the guy with both good information and wisdom.

So, tell me again which system is "nearly worthless"


You should have noticed from all of my posts already. Which system? ALL of them.
noseeum
Major League Manager
Major League Manager


Posts: 1697
Joined: 1 Mar 2005
Home Cafe: Baseball

Re: Mixed Opinions

Postby GotowarMissAgnes » Fri Jan 09, 2009 11:09 pm

noseeum wrote:You should have noticed from all of my posts already. Which system? ALL of them.


You need to go back and read your own posts.


"If someone put out a projection system that integrated MLEs with Marcel, it would be better than Marcel, and it would prove to be so statistically, because MLEs have been proven to correlate highly with major league performance."

Demonstrably demonstrated to be false by the data you posted. Marcel beats a majority of the systems that try to integrate minor league and other modeler data into their projections. Marcel is better, because even if you you have decent data, your system has to incorporate it into your projection in the right way.

Bad data + bad system loses to Marcel
Bad data + good system loses to Marcel
Good data + bad system loses to Marcel
Good data + good system is the only one that be beats Marcel. And Pecota may be the only one that meets that criteria...maybe CHONE.
B
"I don't want to play golf. When I hit a ball, I want someone else to chase it."
GotowarMissAgnes
Hall of Fame Hero
Hall of Fame Hero

User avatar
Fantasy Expert
Posts: 5516
Joined: 12 Dec 2003
Home Cafe: Baseball
Location: Happy Valley

Re: Mixed Opinions

Postby noseeum » Fri Jan 09, 2009 11:27 pm

GotowarMissAgnes wrote:
noseeum wrote:You should have noticed from all of my posts already. Which system? ALL of them.


You need to go back and read your own posts.


"If someone put out a projection system that integrated MLEs with Marcel, it would be better than Marcel, and it would prove to be so statistically, because MLEs have been proven to correlate highly with major league performance."

Demonstrably demonstrated to be false by the data you posted. Marcel beats a majority of the systems that try to integrate minor league and other modeler data into their projections. Marcel is better, because even if you you have decent data, your system has to incorporate it into your projection in the right way.

Bad data + bad system loses to Marcel
Bad data + good system loses to Marcel
Good data + bad system loses to Marcel
Good data + good system is the only one that be beats Marcel. And Pecota may be the only one that meets that criteria...maybe CHONE.
B


I said it earlier. The purpose of Marcel is to create a baseline. It has no intelligence built in. It has such obvious flaws in it, and that's its purpose. It's like the Socrates of projection systems. The advantage it has is that it knows that it knows nothing, and it's up front about that. The message of Marcel is, "if a system can't beat this consistently, it's a joke."

Projection systems are interesting intellectual exercises, and they're a good starting off point, but last time I checked, scoring a 60 in math class counts as an F. None of them can be depended on. Saying Marcel is better than another is not a compliment to Marcel. It's an insult to the other system.

And BTW, it's a good thing they can't be depended on or fantasy baseball would be boring.
noseeum
Major League Manager
Major League Manager


Posts: 1697
Joined: 1 Mar 2005
Home Cafe: Baseball

Re: Mixed Opinions

Postby GotowarMissAgnes » Sat Jan 10, 2009 9:54 am

noseeum wrote:
And BTW, it's a good thing they can't be depended on or fantasy baseball would be boring.


I don't think it's that they can't be depended on, I think it's that people misunderstand or misuse them. Projection systems are no good for telling you how an individual player will do. They are very good for telling you how your overall team will do.
"I don't want to play golf. When I hit a ball, I want someone else to chase it."
GotowarMissAgnes
Hall of Fame Hero
Hall of Fame Hero

User avatar
Fantasy Expert
Posts: 5516
Joined: 12 Dec 2003
Home Cafe: Baseball
Location: Happy Valley

Previous

Return to Baseball Leftovers

Who is online

Users browsing this forum: No registered users and 8 guests

Forums Articles & Tips Sleepers Rankings Leagues


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