Thanks to DZ, Dave, Fred, James, and MR for their great questions and commentary over the last 3 weeks. Let's get into it:
BaseRuns and Linear Weights
Q:
Part 1 - I'd like to figure out if I had 8 players and I knew their LWTS, how many runs would a team of these 8 players score for the year, assuming we are playing in today's environment (2003)? it probably makes more sense to use an average team in the national league for 2003 as the run environment.
Part 2 - Sounds like if you compute the average # of runs for that environment you would just add each player's LWTS to that # to come up with the total # of runs created for that player. correct?
A:
Part 1 -
To do this, you do the following:
1 - give each of the 9 guys the same number of PAs
2 - Use BaseRuns on this team
That'll give you team runs scored.
Part 2 -
Now, once you have that, maybe you want to know: "ok, who's responsible for some of that?" You need LWTS. What LWTS formula? The best one to use would be the one that you would generate from BaseRuns. How do you generate one from BaseRuns? Well, you start off with your "team of 9" stats that you have. BsR will tell you how many runs that team will score. THIS IS YOUR BASELINE.
Then, add 1 H and 1 AB to that team. How many runs would this team score? What's the difference from your baseline? Congratulations, you just calculated the linear weight value of a single (around .47). Now, start with your baseline, and add 1 H, 1 AB, 1 2B. Do the same process for every hitting event, including the out! You have just generated custom LWTS values.
Data
Q:
Where do you get your data? I'm hoping to get it fairly cheaply.
A:
www.retrosheet.org and www.astrosdaily.com/ass and www.baseball1.com (the Lahman DB) all for free
Q:
What statistical applications do you use? Do I need to be fluent in SAS, or does excel tend to be powerful enough?
A:
I use Excel, and Access.... sometimes I write programs, like for a sim or Markov.
Q:
What sites and publications contain the bulk of the research in this field? I have a bunch of ideas but don't want to replicate old research. Where are the good discussion groups?
A:
http://www.baseballprimer.com
http://www.baseballprimer.com/studies
http://insiders2.ezboard.com/fbaseballfrm8
Leveraged Index (LI)
Q:
Never came across a formula for calculating LI. Did you post that anywhere? Sounds like you need play-by-play data as well. anyway around that?
A:
My LI calculations are based on my own engine, which I have not published.
Q:
How do evaluate relievers would be if I was say a GM, hypothetcially. Right now I basically use Wolverton's ARP and look at other factors like qDIPS era, strikeout rate, etc.
A:
Forget APR, LI and everything else. To evaluate a reliever, you need to know about his talent. ARP gives the player extra credit/debit for how he performs with men on base. That's not fair forward-looking (but it is backward-looking). Same deal with LI.
If you have 2 guys with 80 IP, 20 BB, 70 K, 50 H, then it's irrelevant if one guy did it with bases empty mostly, and the other guy always came in with men on base.
If you want to say that the pitcher performs far differently with men on base, or in close situations, for some mechanical or psychological reason, that's fine. But ARP won't help you here. You'd use LI to establish the boundaries of what is high-leverage or not, but I wouldn't use it beyond that. Even then, I don't think you'll get the statistical significance to say with much confidence that "yeah, Armando can't handle the pressue".
Win Expectancy
Q:
What I need is the park adjusted WE table but I don't have a simulator to do this
A:
What you actually need is not only a park-adjusted WE, but a WE for that type of hitting team. For example, say you have a great hitting team. Right away you're starting with a much higher WE, and therefore, any pitching value will have limited impact. So, what you need to do is establish the RUN ENVIRONMENT WE (based on hitting/opposing pitching and pitching/opposing
hitting).
The hitch here is that if you've got Pedro on an average hitting team, with average opponents, your initial WE for Boston is like .700. Anything positive that Boston does will be limited, since they've already got a great chance to win. However, this means that Pedro himself can't add much, since his presence already establishes a much higher WE.
So, what you need to do is that after you've established the run environment, you need to allocate that +.200 (.700-.500) to your 50 players (with essentially Pedro getting +.200, Manny getting +.004 or something, etc, so that it all balances out).
Then, on a PA-by-PA basis you need to recalculate the WE. The problem is that when Manny comes to bat, he brings up the WE before he even swings the bat, because that's the expectation. So, just like you redistributed that +.200 at the start of the game, you have to do this at the start of EVERY SINGLE PA.
Now, that is a tremendous amount of work, which is why I don't do it. Which is why I just assume a .500 playing field for every PA.
So, instead of killing yourself to get the park-based WE (which is really only one step in the whole process), do a park-adjustment at the end. If you figure that your pitching should be +.010 wins / PA in a neutral park, but they perform at -.020 wins / PA in a hitter's park, add +.030 wins / PA after the fact.
This may be wrong, but the whole process is wrong anyway, unless you do that tremendous amount of work.
Win Shares
Q:
In Bill James win shares system ...Tony Phillips at .262, 4 HR's and 47 RBI is as valuable to the 1989 Oakland A's as Dennis Eckersley with 33 saves and a 1.56 era. Have I been fooled ...into believing that "good pitching beats good hitting?" Are runs created more valuable than runs allowed in terms of valuing individual contributions to wins?
A:
Bill James is wrong on this issue, which means it'll take at least 20 years to correct. The best way to measure it is through a process called "Win Probability Added". You can check out the site here, as I have a link there that talks about it, as well as a link to Drinen's site that gets into more detail.
Q:
What do you have against Win Shares?
A:
Well, it falls short for pitchers mostly. If you go to my site, I show how it breaks down for pitchers. I wrote 3 articles on Win Shares: http://www.geocities.com/tmasc
I actually don't have much more to say other than to repeat myself.
My objective is not to try to convince anyone either, other than to present information in as a logical manner as I can, and let the reader decide.
If there is a particular passage or topic that you would like me to expand upon, I'd be happy to do that.