Return to Baseball Leftovers

How do you rank for 5x5?

Moderator: Baseball Moderators

Postby shortsavage » Mon Jan 09, 2006 7:40 pm

thinkspin wrote:Are we talking about RANKING players or PROJECTIONS? Big difference...


Ranking. Suppose you worked for MLB.com and you were assigned to put dollar values on say the top 20 C, 1B, 2B, SS, 3B, RP top 50 OF, and top 80 SP, how would you go about putting values on players?

Or if you think dollar values are not a productive use of time, in what ways do you go about your business. I'm talking about studious approaches, not winging it.
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby thinkspin » Mon Jan 09, 2006 8:29 pm

shortsavage wrote:
thinkspin wrote:Are we talking about RANKING players or PROJECTIONS? Big difference...


Ranking. Suppose you worked for MLB.com and you were assigned to put dollar values on say the top 20 C, 1B, 2B, SS, 3B, RP top 50 OF, and top 80 SP, how would you go about putting values on players?

Or if you think dollar values are not a productive use of time, in what ways do you go about your business. I'm talking about studious approaches, not winging it.


So, you want to get as complex as it gets? You have to pick a starting point and then build from there. I suggest you start with figuring out 1 category at a time. For instance, how do you RANK HRs? Most sites take the AVG for all players, assign a 0 value to the average, and then assign a weighted value to how far above or below the average a particular player is with regard to HRs. Think standard deviation. You do the same for other similar cats. Some cats need to be weighted by ABs and some need to be negated (if you did SO for hitters).

Do this for all your cats and you have a base. There are several more steps after this, but that may deserve another thread...
thinkspin
College Coach
College Coach


Posts: 167
Joined: 9 Jan 2006
Home Cafe: Baseball

Postby davidmarver » Mon Jan 09, 2006 9:08 pm

I always give a slight boost to balance. It's a lot easier (higher flexibility) to replace a balanced player than a player that anchors three categories.
davidmarver
Hall of Fame Hero
Hall of Fame Hero

User avatar
Cafe WriterMock(ing) DrafterEagle EyePick 3 ChampionLucky Ladders Weekly WinnerTrivia Time Trial Monthly Winner
Posts: 5826
(Past Year: -328)
Joined: 11 Jan 2005
Home Cafe: Baseball
Location: 5.5 Hole

Postby shortsavage » Mon Jan 09, 2006 9:43 pm

I don't want to be taught how to rank...well that is not my intention at least.

I want to compare my approach to other approaches.
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby thinkspin » Mon Jan 09, 2006 10:02 pm

shortsavage wrote:I don't want to be taught how to rank...well that is not my intention at least.

I want to compare my approach to other approaches.


I am just explaining to you how other sites work. I thought this is what you were looking for...

You never did mention your approach. If you explain your process, I will try to help.

Like I said, it is a complex process not a simple calculation. Not sure if you can compare one to the other.
thinkspin
College Coach
College Coach


Posts: 167
Joined: 9 Jan 2006
Home Cafe: Baseball

Postby Pedantic » Mon Jan 09, 2006 10:19 pm

shortsavage wrote:I don't want to be taught how to rank...well that is not my intention at least.

I want to compare my approach to other approaches.


Perhaps you should be a little more clear in describing what exactly you're seeking. You asked how we rank players for 5x5, yet that's not what you want to know? :-? If you do a search, there have been quite a few threads about creating dollar values, if that's what you're after.
Image
Pedantic
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicEagle EyeCafe Musketeer
Posts: 6663
(Past Year: -62)
Joined: 5 Dec 2003
Home Cafe: Baseball
Location: Funny movie quote that everyone knows

Postby shortsavage » Mon Jan 09, 2006 10:21 pm

I guess I'm most interested in how many people actually crunch numbers and weigh stardard deviations opposed to how many people do less and how many go way beyond standard deviation.

I've been working on an excel sheet that calculates the average numbers of a major league starter each season (.267, 14HRs, 8 sb, 66 RBIs, and 62 runs for 2005 so far with every type of hitter except outfielders). Once I came up with this figure, I set up a function that looks at how far each player deviates from that average. Then, I calibrated points to be given for slightly below average numbers to well above average numbers, ultimately to be turned into dollar values that I will use to rank players at every position and establish a top 100 overall.

I also amend values based on age (I only subtract points from aging players) and attributes that I find noteworthy (such as top prospect and burried as a backup).

The values that I have been getting are pretty crisp (I'll post them sometime this month).

Like I said above, I'm interested in figuring out how my approach compares to other fantasy baseball players in terms of the amount of thought that goes into it and its complexity.
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby thinkspin » Mon Jan 09, 2006 10:36 pm

shortsavage wrote:I've been working on an excel sheet that calculates the average numbers of a major league starter each season (.267, 14HRs, 8 sb, 66 RBIs, and 62 runs for 2005 so far with every type of hitter except outfielders). Once I came up with this figure, I set up a function that looks at how far each player deviates from that average. Then, I calibrated points to be given for slightly below average numbers to well above average numbers, ultimately to be turned into dollar values that I will use to rank players at every position and establish a top 100 overall.
.
.
.
Like I said above, I'm interested in figuring out how my approach compares to other fantasy baseball players in terms of the amount of thought that goes into it and its complexity.


If you are doing it in excel it is a good start. I was going to explain to you that after you get your "calibrated points" you will have to do another pass or two to calculate dollar values correctly. I wanted you to get your "calibrated points" first because that is the easy part.

What I mean by another pass is that the AVERAGE you have is the AVERAGE for the entire population and not the AVERAGE for the population that will be drafted. And, in the population that will be drafted, you have to account for position. To be correct, you can't figure out the draftable population AVERAGE correctly if you are not including all the Catchers that should be drafted. Like I said, complex.

Once you get this, you can move to Dollar Values. That is why in my very first post, I said...But ranking players CORRECTLY is very complicated. If you are looking for your own method, a good place to start would be the 2005 Regular Season statistics. Rank using your method and see if it compares with what other sites/sources say.
thinkspin
College Coach
College Coach


Posts: 167
Joined: 9 Jan 2006
Home Cafe: Baseball

Postby shortsavage » Mon Jan 09, 2006 10:53 pm

Yeah, I've done all of that. I have polished rankings.

Hehe, I guess I don't know how to ask the question I'm interested in.
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby dannahann » Tue Jan 10, 2006 10:02 am

I calculate something very similar to your "calibrated points" in relation to the average (draftable) player and incorporate age/PT modifiers too. I aslo incorporate another standard tho. I run each player versus the average drafted player at his given position. (use his thinnest position if a multi guy). I found that 3X(all player score) + (position specific score) works well for me in that it installs some position scarcity without it being too influential. I then add a bit more position scarcity by applying standard 'VBD' (value based drafting) techniques to the resultant scores. VBD is a bigtime football strategy you can easily find it on Google.
This gets me a baseline points ranking with players across all positions in their relative order. To this I incorporate average draft positions to formulate where I'll need to take guys. My leagues are all keeper leagues so I also utilize modifiers to ensure I properly wieght the positions as to their 'real need' after factoring in my keepers and those of the guys I'm most worried about.
dannahann
Major League Manager
Major League Manager

User avatar
Cafe Ranker
Posts: 1993
(Past Year: 7)
Joined: 26 Jun 2002
Home Cafe: Baseball
Location: Missouri

PreviousNext

Return to Baseball Leftovers

Who is online

Users browsing this forum: No registered users and 19 guests