Members: Login | Register | Feedback
 
   
2007 Positional Adjustments
Posted: 27 March 2007 12:40 AM   [ Ignore ]

I haven’t been in a pool in a while really, and I’m helping a friend.  I’ve thrown some numbers together but I’m looking for some positional adjustments, ideally from 2006 data, that I could use as a multiplier.  The last time I did this was about 3 or 4 years ago and I think 2b got a 11% bump or something.

  Anyone have anything?

TM

Posted: 28 March 2007 12:37 AM   [ Ignore ]   [ # 1 ]

I don’t recall where or how I ran across this thread - but I found it a few weeks ago and have been studying it ever since….  good stuff from Tango!

It’s more focused on pure player value, but around post ~60 there begins some positional adjustment numbers….

Posted: 28 March 2007 02:26 AM   [ Ignore ]   [ # 2 ]

Thanks zonk, in the end I played with the numbers I had and decided that positional adjustments based on the stats that were important for this league (Actually, I think these are called the standard 5x5 stats, but I’m used to more detailed types) was the best way to go.  It’s interesting though, this league groups all OFs together, so initially the numbers indicated that all OFs should get an 8% boost to put them on even ground with the 1Bs.  However, looking at the actual players I realised that I wasn’t the best way to go, the bottom of rung of OFs was much worse than the bottom rung of 1Bs.  In other words, there were 2 ways to generate multipliers:

1) Take all players at a position, compare them to all other players
2) Take the top X players at each position, compare them to the top X players at other positions (where X is relevant to the size of the league)

  After much hemming and hawing, and a few phone calls, I decided method 2 was the way to go.  The multipliers based on position were much much higer than original, but it pushed up some catchers and 2Bs up the list and that made sense to me.

  I’m curious as to whether anyone has any comments on this.  I hesitant to share the actual numbers as supposedly there are a few members of the pool who lurk here, but I may do that after the draft to see what the response is.

P

Posted: 28 March 2007 01:20 PM   [ Ignore ]   [ # 3 ]

I do agree with the results—I’m probably going to keep Josh Bard at $11 in my 5X5 because there are so few valuable catchers in the NL, so I definitely buy into premium players at scarce positions getting value bumps.

The X factor, though, with any valuation system - especially in an auction setting - is that once the bids start flying, things can get mighty skewed.  Josh Willingham - who, granted, was hardly a deep sleeper heading into last year - went for $17 in our auction… and I came thisclose to going to $18 (I had him pegged at $12 beforehand).  I actually paid $17 for Corey Hart (final U spot scramble amongst 3 owners who saved for end-of-draft bargains that didn’t materialize).  Even Dan Uggla went for something like $7.

I toyed with a more analytical approach a few seasons back (not just position scarcity, but also factoring in skill weights, i.e., projected SBs, HRs, etc)—and promptly finished 5th with it (my only season “out of the money”).  That’s not to say I don’t walk into the auction with a list of values/value ranges for every player available - you certainly don’t want to be pulling numbers out of the sky - but I think any system needs to be dynamic to react to the changing realities of draft day.  Given that most owners in my league also end up with anywhere from $50 to $150 tied up in keepers before the auction even starts - estimated values end up becoming even more a reference point.

Nowadays, I lean towards a simpler tiered approach (the advent of the laptop has made it so much more effective)—cream of the crop, underated, sleepers, and ‘dump’ players (players I want no part of but bring up in bidding just to tie up someone else’s roster and $$$).

What I plan to play with a bit tonight is some sort of macro or equation that might be able to react to the changing turf… if we suddenly get a run on closers - and even someone like Armando Benitez commands a $30 price tag, Lidge another $30 - and bidding wars erupt around David Weathers - I think this changes the relative value of everyone else (especially if you buy into “never punt a category”).