F-Bball Statistics - Fantasy Baseball Cafe 2014 Fantasy Baseball Cafe
100% Deposit Bonus for Cafe Members!

Return to Baseball Leftovers

F-Bball Statistics

Moderator: Baseball Moderators

F-Bball Statistics

Postby Matthias » Thu May 11, 2006 1:39 pm

Just general curiosity/debate question: What is the meaning of the standard 5x5 cat's?

I always assumed that they were chosen because of the ease with which they could be compiled and that initially the Rotisserie Cafe guys were doing this with pencil and paper. But there was a bit in Fantasyland where the author said that they were chosen because one of the original Roto guys did an analysis of what statistics were correlated(?) or predictive(?) of winning and came up with the standard 5x5.

But, for the most part, the categories seem strongly correlative but rather unpredictive. I mean, e.g., wins. Yah. Of course that's correlated with team wins. Ditto with saves. Runs & RBIs, same. You score more, you're going to win more.

But they seem to be the end-product of statistics which go into winning rather than the statistics which go into actually winning if that makes any sense.

Thoughts?
Matthias
General Manager
General Manager


Posts: 4860
Joined: 16 Mar 2005
Home Cafe: Baseball

Postby Niffoc4 » Thu May 11, 2006 2:20 pm

what stats would be as correlated but more predictive? combo stats like runs produced per at bat? Fantasy stats have never been predictive... the goal is to have stats that can be combined to produce a team of players that would win more than another... why wouldn't R, HR, RBI, BA, ERA, K, WHIP, Saves and to a lesser extent SB and wins work for that?
Niffoc4
Major League Manager
Major League Manager


Posts: 1700
(Past Year: 9)
Joined: 26 Oct 2004
Home Cafe: Baseball

Postby Matthias » Thu May 11, 2006 2:44 pm

Not precisely sure, but to give a sense, OPS seems like a more predictitive stat than correlative. Also, if you believe in DIPS, then HR allowed would be more predictative than W's. To take the statement that, "A team which has a lot of pitchers with wins will win a lot of games" seems so tautological to be asinine. Saying, "A team that scores a lot of runs or doesn't let in a lot of runs will win a lot of games" is just one level better.

But to say that a team that hits for average, hits for power, plays good defense, strikes out batters, doesn't let balls get into play (if you buy into DIPS), and makes good use out of the bases which it does get (i.e. steals) will win a lot of games seens to be process-oriented.

Or put another way, you can go to a very good bakery and buy cakes and say, "A good way to get a good cake is to go to this bakery." But that doesn't tell you how you actually make a good cake: just that someone else has figured it out. And, at the heart of it, what I really believe you're trying to do with fantasy sports is postulating that if you had this grabbag of players on your team, how would they do relative to everyone else? Which is a more raw ingredients approach rather than a this is what I bought pre-made kind of thing. Don't know if that makes things clearer or more opaque but I think the point kind of comes through. Kind of.
Matthias
General Manager
General Manager


Posts: 4860
Joined: 16 Mar 2005
Home Cafe: Baseball

Postby Xtreemes » Thu May 11, 2006 2:53 pm

I think those 5x5 stats were chosen because they are obvious stats that general baseball fans understand. They know what a win is, what a homerun is, what a stolen base is. They don't want to predict anything, they just want to use general, common baseball stats.
Xtreemes
Minor League Mentor
Minor League Mentor

User avatar

Posts: 533
Joined: 29 Mar 2004
Home Cafe: Baseball

Postby Dan Lambskin » Thu May 11, 2006 3:09 pm

Xtreemes wrote:I think those 5x5 stats were chosen because they are obvious stats that general baseball fans understand. They know what a win is, what a homerun is, what a stolen base is. They don't want to predict anything, they just want to use general, common baseball stats.


plus before the internet, it was probably easiest to pick them out of a boxscore
Image
Dan Lambskin
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicCafe RankerMock(ing) DrafterEagle EyeLucky Ladders ChampionPick 3 Weekly WinnerMatchup Meltdown SurvivorTrivia Time Trial Monthly Winner
Posts: 10225
Joined: 20 Mar 2005
Home Cafe: Baseball

Postby The Loveable Losers » Thu May 11, 2006 3:15 pm

I think what you're looking for is more of a sabermetric league.

Just off the top of my head I would think numbers like runs created, ops, sb-2*cs (to penalize people that get caught a lot) would make more sense from that point of view. Likewise you could replace wins with a stat like quality starts or something along those lines and replace saves with hr/9. I would also replace k's with k/9. From a complete sabermetric perspective though you'd really want to remove era and whip entirely and have your categories be k/9, bb/9 and hr/9 as these are the only things a pitcher actually controls.

So if I were to throw a league together just off the top of my head I'd probably do something like this:
Hitters: Runs created, hr, obp, slg%, sb-2*cs (could even go with sb-3*cs if you wanted to REALLY reward good base stealers versus volume base stealers)
Pitchers: Starting pitcher innings, relief pitcher innings, k/9, bb/9 and hr/9

Those are a couple of odd categories on the pitcher end...I realize that. My reasoning was twofold. First pitchers really only control their k/9, bb/9 and hr/9. Anything beyond that is out of their hands. So those are the only 3 categories worth tracking. Second and most important though this gives us a natural counterbalance to people that might try to win k/9, bb/9 and hr/9 by pitching only a few pitchers. By having two counting categories in there you pretty much take care of everything in a roto league (since no one is going to punt 2 categories in a 5x5 just to end up with top points in 3 categories). You'd need to have a minimum ip'd in a h2h league to avoid nastiness there. On the other hand by having 3 ratio categories you really discourage pitcher churning. The SP innings and RP innings could easily be replaced with something else but I'm not really sure what you could put in there that would make sense from a sabermetrical perspective.
The Loveable Losers
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicFantasy ExpertCafe WriterCafe RankerMock(ing) DrafterWeb Supporter
Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!

Postby Matthias » Thu May 11, 2006 3:36 pm

Dan Lambskin wrote:
Xtreemes wrote:I think those 5x5 stats were chosen because they are obvious stats that general baseball fans understand. They know what a win is, what a homerun is, what a stolen base is. They don't want to predict anything, they just want to use general, common baseball stats.


plus before the internet, it was probably easiest to pick them out of a boxscore


Yah, that's what I always thought, too (see above).

But there was a bit in the book Fantasyland that contradicts this theory. Maybe I'll look it up and post an excerpt when I get home tonight but wondering if anybody else had thoughts along these lines.

Basically what it said was that the traditional 5x5 stats were sabermetric-esque stats or at least were chosen because they had the highest correlation with winning. Which flew in the face of what I always thought.
Matthias
General Manager
General Manager


Posts: 4860
Joined: 16 Mar 2005
Home Cafe: Baseball

Postby The Loveable Losers » Thu May 11, 2006 3:40 pm

Matthias wrote:
Dan Lambskin wrote:
Xtreemes wrote:I think those 5x5 stats were chosen because they are obvious stats that general baseball fans understand. They know what a win is, what a homerun is, what a stolen base is. They don't want to predict anything, they just want to use general, common baseball stats.


plus before the internet, it was probably easiest to pick them out of a boxscore


Yah, that's what I always thought, too (see above).

But there was a bit in the book Fantasyland that contradicts this theory. Maybe I'll look it up and post an excerpt when I get home tonight but wondering if anybody else had thoughts along these lines.

Basically what it said was that the traditional 5x5 stats were sabermetric-esque stats or at least were chosen because they had the highest correlation with winning. Which flew in the face of what I always thought.


The Fantasyland bit was basically that the guy who created fantasy baseball picked the stats out based on what he felt had the highest correlation to a winning baseball team. They're not sabermetrically based as much as they're based on what one guy that looked at the box scores felt it took to win baseball games.
The Loveable Losers
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicFantasy ExpertCafe WriterCafe RankerMock(ing) DrafterWeb Supporter
Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!


Return to Baseball Leftovers

Who is online

Users browsing this forum: No registered users and 11 guests

Forums Articles & Tips Sleepers Rankings Leagues


Today's Games
Wednesday, Aug. 20
(All times are EST, weather icons show forecast for game time)

Texas at Miami
(12:40 pm)
indoors
Seattle at Philadelphia
(1:05 pm)
Toronto at Milwaukee
(2:10 pm)
NY Mets at Oakland
(3:35 pm)
Houston at NY Yankees
(7:05 pm)
Atlanta at Pittsburgh
(7:05 pm)
Arizona at Washington
(7:05 pm)
Detroit at Tampa Bay
(7:10 pm)
indoors
LA Angels at Boston
(7:10 pm)
Cincinnati at St. Louis
(7:15 pm)
San Francisco at Chi Cubs
(8:05 pm)
Baltimore at Chi White Sox
(8:10 pm)
Cleveland at Minnesota
(8:10 pm)
Kansas City at Colorado
(8:40 pm)
San Diego at LA Dodgers
(10:10 pm)

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