This is a "secret" formula used and developed exclusive by Bill James of STATS, Inc. which is used to determine what a Minor League player would have hit had he been a player in the Major Leagues. It is not used a tool for future predictions but rather determines the player's level of performance in the past and adjusts for the difficult level in the big leagues to estimate performance on a current Major
this is what I keep coming across, and its driving me nuts. Does anybody ...
A) Know the formula or
B) Know some mathematical formula similar, so that I can look at a AAA players stats, and equate it to MLB. It would need to take into account park effects I guess, but I can deal with that later...
Ideally this is going to be used in excell, but I don't think that matters, right now.
I would love to be able to figure out what a players Major League Equivalency was for a given season, so that I can use it to project.
Bill James' aggregate translation factor (82% of AA or AAA stats) makes for an exact projection of the Major League Equivalency (MLE) for every hitter.
Major League Equivalency - The major league performance that is equivalent to a given performance in the minor leagues. Bill James discovered that by making appropriate adjustments for league and park, you could figure an MLE for a player's minor league performance that is as good for predicting future performance as prior major league data.
So it looks like he takes AAA stats and multiplies by 82% then adjusts for the league and the park. It is much more complicated than that but :
The more Bill James that I read, the more I believe that he is full of crap. As a purely academic exercise it's fun to look at stats - then he comes up with things like this:
SBR Stolen Base Runs SBR = .3*SB-.6*CS or SBR = .3*(SB-2*CS). Total Baseball's approach to quantifiying base-stealing. Numerous statistical studies show that the breakeven success rate for steals (the rate at which attempting to steal is neither helping nor hurting the team in terms of total runs scored) is about 67% -- below that you are costing your team runs. Each successful steal adds about .3 runs to a team's total -- far less than in generally believed. SBR estimates the impact of base-stealers, which, other than the elite base-stealers, rarely amounts to more than a few runs per year.
This does not take into consideration when the base is stolen, what affect it has on the pitcher, how the man at the plate performs when the pitcher is trying to hold a man on base etc...
Speed and base-stealing ability is a tool that should be used like all others (hitting behind, hit and run, sac etc) where the situation calls for it.
Mookie4ever wrote:This does not take into consideration when the base is stolen, what affect it has on the pitcher, how the man at the plate performs when the pitcher is trying to hold a man on base etc...
How would expect a mathematical algorithm to take that into consideration? It could but first we'd have to figure out how to quantify it. I'm not sure how that makes James full of crap.
Maine has a good swing for a pitcher but on anything that moves, he has no chance. And if it's a fastball, it has to be up in the zone. Basically, the pitcher has to hit his bat. - Mike Pelfrey
I think he is saying that there are so many variables, many of which cannot not be easily quantified if at all, that it is foolish to use a small set of statistical categories to make precise calculations about the value of certain strategies like stealing a base.
stumpak wrote:I think he is saying that there are so many variables, many of which cannot not be easily quantified if at all, that it is foolish to use a small set of statistical categories to make precise calculations about the value of certain strategies like stealing a base.
Actually they are using a broad group of statistical categories and 100+ years of game data. It's not quite the guessing game that you're trying to suggest it is.
Maine has a good swing for a pitcher but on anything that moves, he has no chance. And if it's a fastball, it has to be up in the zone. Basically, the pitcher has to hit his bat. - Mike Pelfrey
stumpak wrote:I think he is saying that there are so many variables, many of which cannot not be easily quantified if at all, that it is foolish to use a small set of statistical categories to make precise calculations about the value of certain strategies like stealing a base.
Actually they are using a broad group of statistical categories and 100+ years of game data. It's not quite the guessing game that you're trying to suggest it is.
Exactly, he's saying that (on average) a stolen base adds 0.3 runs to the total scored. That's an average based on statistics throughout history. He's taken those numbers and plugged them into the equations, and guess that? They work. They're based on historical evidence. They're based on that actual situations in REAL life ballgames over the last century and a half. That means that the equation (while impossible to quantify) actually DOES take into account the "SB Effect" and all the other intangibles about stealing a base in a live game situation.
Throughout history, a stolen base generates approxamately 0.3 runs per game. That takes EVERYTHING into account -- including psychological effects (unless you'd like to argue that today's pitchers are more psychologically susceptible to the stolen base). All James did was give it a name and a fancy equation. He just produced a number (0.3) that could quantify the value of a stolen base.
[size=10]"Men are apt to mistake the strength of their feeling for the strength of their argument." [/size]
I understand that all that is embedded, but I am agreeing with Mookie's assertion that an SB, like a lot of other things, is a situational tool. Because it is an aggregation, the James formula assumes a lot of things, like an average basestealer, an pitcher who is average at holding runners, a catcher who has an average arm and ecompasses a wide range of of situations as to at which base the runner is on, how many outs there are, etc.
In real life, these "average" situations are almost never the case and it does little good to make blanket statesment like a good base stealer gets you three additional runs over the season or whatever James said. If your situation is two outs, an 90% successful basestealer on first and a .250 hitter with no power at the plate, it obviously makes sense to steal; if you have no outs, a 60% successful basestealer at first and a .300 hitter with power at the plate, it obviously doesn't. I guess I am saying that SBs are almost always tactical--the decision-making on whether to steal or not can be marginally informed by aggregations (i.e., err on the conservative side of stealing according to James) but statistical aggregations in reality are never the fundamental driver of tactical decisions.
stumpak wrote:I understand that all that is embedded, but I am agreeing with Mookie's assertion that an SB, like a lot of other things, is a situational tool. Because it is an aggregation, the James formula assumes a lot of things, like an average basestealer, an pitcher who is average at holding runners, a catcher who has an average arm and ecompasses a wide range of of situations as to at which base the runner is on, how many outs there are, etc.
In real life, these "average" situations are almost never the case and it does little good to make blanket statesment like a good base stealer gets you three additional runs over the season or whatever James said. If your situation is two outs, an 90% successful basestealer on first and a .250 hitter with no power at the plate, it obviously makes sense to steal; if you have no outs, a 60% successful basestealer at first and a .300 hitter with power at the plate, it obviously doesn't. I guess I am saying that SBs are almost always tactical--the decision-making on whether to steal or not can be marginally informed by aggregations (i.e., err on the conservative side of stealing according to James) but statistical aggregations in reality are never the fundamental driver of tactical decisions.
James would agree with you. If you have a 90% success rate, then the stolen base is a valuable tool, but only a very select number of base stealers have a percentage higher than 65%. If you can guarantee that you will succeed at least 65% of the time, then it makes sense to steal. James isn't saying that all stolen base attempts are a bad thing, but he is saying that in most situations, the chances of getting thrown out outweigh the runs created.
[size=10]"Men are apt to mistake the strength of their feeling for the strength of their argument." [/size]
Cafe Home
Fantasy Football
Fantasy Basketball
Fantasy Hockey
by wrveres » Mon Oct 18, 2004 4:35 am 
