Best lefty off all time? My vote is Koufax!

[BobC]

Originally Posted by Mark17

All of this is just listing many variables involved, and I'm sure many more can be added. The obvious difficulties that remain are:

1. How do these variables play together? Are they additive, multiplicative, subtractive, and to what degree. How do you combine and weigh them?

2. How do you value them, with respect to specific players?

For example, let's say you are comparing 2 pitchers who both have a right fielder with a .985 fielding average. But one has a weak throwing arm and the other is Clemente. How much does having Clemente help, with his reputation discouraging runners taking an extra base?

First you'd need to give a weight to the variable - what impact does the right fielder's reputation play? Second, you have to value Clemente.

Suppose there are two catchers with equal fielding percentages, and throw out equal percentages of baserunners. But one is a very astute signal caller and the other is a dolt. Take Grove having Cochrane for example. First, how much can a smart, observant catcher help a pitcher? Second, what value do you assign to Cochrane (or Roseboro?)

All you have done is thrown out a bunch of factors to consider. The real trick would be to come up with an algorithm that can effectively combine and weigh the variables, and then, there's the (sometimes subjective - like the brains of a catcher) value you assign to each specific player involved.

In short, the above is not anywhere close to an actual predictive model.

Mark,

I think you bring up some great points, along the same lines I was alluding to, that there are going to be so many variables to factor into answering a question like this that it is virtually (and likely literally) impossible to effectively factor them all into any statistical equation or formula. You can attempt to do it, but at the of end of the day you'll only end up with what a statistician thinks is the right answer. And who elected them to decide that their opinions and points of view speak for all of us, or automatically overide what everyone else may think. I understand they can create these great statistical equations and algorithims to come up with a predictive formula to help decide who MAY be the best at something, but how can one be so certain of the outcome of such an equation or formula until they've actually created it and been able to show and prove it works. I thought in science that is what is known as a theory, which is unproven, and remains as such till someone can actually prove it is true and works. I don't seem to remember any true scientists ignoring questions about their work in regards to such theories, and simply telling people to trust and believe them because they have neither the time nor the inclination to fully explain their position. Nor to claim they know the answer to a question based on such a formula, when that formula has yet to be created, tested, and proven.

And that goes for key assumptions that are part of such theories and thinking, like the making of a blanket statement that ballplayers from 60-70-80+ years ago are much weaker players than they are today. How, why, what empirical data is there to factually prove that? You can pull up all the numbers and speculate and manipulate them all you want. And I understand about the increases in the population and how that factors in and, and yadda-yadda-yadda. But has the human male really evolved and changed that much physically in that last 100 or so years, or is it more so from advances in science, training, nutrition, medicine, education, even economics playing a huge part, and on and on. Heck, I've even heard somewhere that overall male testerone levels have been dropping generation by generation over the last century or so, which would initially make you think that earlier male generations may have actually been considered more masculine (and by extension athletic) than they are today. So maybe those differences in how players played back then are more due to all the other cultural and outside influences that were affecting them than most people (especially statisticians) would think. And how, unless you took players (not pitchers) from today and had them grow up to play 60-70-80+ years ago, and then likewise had the players (again not pitchers) from back then grow up to be playing today, could you really even begin to tell which era's players were stronger or weaker. Now according to blanket statements and assumptions by some statistically minded people, by switching the players like this we would expect to have the transplanted players to back then hitting tons of home runs, while probably striking out more, but overall crushing the pitchers from back then. In fact, the way these statisticians may talk and seem to think, you'd expect that all of Ruth's home run records would have been easily eclipsed way back 60-70-80+ years ago, and as a result he might not be carried anywhere close to the esteem he is today. And as for the transplanted players from back then now playing today, following some statisticians thinking you'd expect them to be completely overwhelmed and effectively having their collective asses handed to them on a daily basis by today's pitcher's, and not even have the league as a whole batting even close to .200.

But somehow I don't think all that would happen. Because humans are affected by and react, change and evolve to fit the situations and circumstances that surround and are constantly changing around them. No one can say with any meaningful certainty how a Grove or Spahn would pitch today. No statistician can honestly measure a person's drive, ambition, competitiveness, and any other intangibles that truly make them the player/athlete they are. And in demeaning and putting down an entire era or generation's ballplayers, without at least trying to factor in all the potential contextual differences between players from different times/eras, is simply insulting to those players. Especially since there is no truly effective way to account for, measure, and quantify all of the infinite number of cultural, contextual, and human differences (in addition to the differences in the game of baseball itself) that would need to be included in such a comparative and predictive formula. But a statistician can get away with saying they can in fact create such a formula or equation to accurately say who or what era/generation was better than another, even though they can't actually or empiracally prove they're right, because they know you or I, for the exact same reasons, can't definitively prove them wrong either.

To illustrate how times and context can be be ignored in statical analysis, the greatest ever left handed pitcher could have been someone born in the 20's who ended up dying in WWII and never even got to play in the majors. Or, they were born in the 20's, but got hurt coming out of high school when there was no Tommy John surgery back then yet, so they never got to play in the majors either. Or what about the time Randy Johnson spent on the injured list, what if he was pitching 100 years ago and got injured, but the medical knowledge back then couldn't completely cure him and he never pitched again, or at least never pitched anyhere near as well as he could have? Or here's a good one, Johnson's in college in the early '50's, and we know from his actual career it would would take him a few years to get his pitching act together. Back in the early '50's, ballplayers didn't get the kind of money they got later on when Johnson actually played. Since he's what '6"10 - '6"11, who is to say the school's basketball coach approaches him about playing BBall, so he does and ends up good enough to make the NBA because of his natural height, and never even goes to pitch in the majors. So how does a statistician ever account for and measure any of these instances in their formulas and equations? They don't, because it doesn't fit into their formulas and equations, but these examples do illustrate how in trying to look at a particular player and how well they may perform in a different time or era, the context of playing in that other time/era could result in a dramatic change to how their career would look or end up.

One last example, though a different sport. Tom Brady graduates from college and ended up being drafted in the 6th round, with what was it, 32 teams in the league then. So what if Brady had actually graduated 40 years earlier, and with a lot fewer teams in the NFL, he never gets drafted and becomes the GOAT. Different time, different context, totally different career outcome.

Statisticians create statistical formulas and equations to predict current game outcomes for gambling purposes. And after doing so, they see what the actual outcome of their game is, and can then tweak and improve their formulas as needed. The main thing is, they can actually test and prove it by looking at how well they did gambling. So they think they have these formulas and equations down and can use them to now try and determine something else like who was a better player, looking at multiple players playing in different times and eras. The problem is, you don't have any actual game or competition that will occur to tell you who won, like you do when you bet on a ball game and their is an actual winner. So there is no way to actually test that type of statistical formula or equation in picking who's the greatest at something all time, and thus be able to prove if that statistical formula or equation is in fact right or wrong. Statisticians will tell you that their statistics are all that can be accurately used to make such decisions, but since they can't ever be proven right or wrong for this type of question, statistics in this regard are nothing more than talking points, no more and no less. Something to maybe talk about, but certainly not the final answer!