How would Babe Ruth in his prime done in the minor leagues?

Cafe Society
1

I was listening to Vin Scully this morning on a tape from the 1980s (hey! we have our own ways of amusing ourselves) rattling on, as is his wont, about how minor league batters allow pitchers to get away with 2-0 cripple pitches that their major league equivalent batters drive into the gaps.

Now, I hate Scully’s consistent lack of reasoning here–obviously, MLB hitters are just flat out better than minor league batters, that’s why they’re in the major leagues, because they’re better–while it’s pretty well unprovable that the big difference in a 2-0 count. But this got me thinking:

How much better are MLB hitters than minor leaguers? if Babe Ruth had played the 1921 season, for some bizarre reason, in the minors, would he have batted .400? .500? .600? You see all the time, young players come up to the majors and start hitting well from the git-go, and very often these were NOT players who burned up the minor leagues statistically, making me think that the difference between the two groups of hitters is, while certainly significant, much more subtle than the Vin Scullys of the world believe.

Could we design a study (or has anyone done so?) that would quantify the difference? I might suggest isolating every batter over a long period of time who has played a full minor league season and then played a full major league season, and someparing the difference in batting stats. This would probably need to be normed by then comparing the difference between the two seasons with the difference between seasons (had at the average age of the players in the first study) that are both at the MLB level and the minor league leavel, to gauge how much improvement is attributable to the difference between being 21 and 22, or 23 and 24, whatever the the average ages turned out to be in the first study. Any stats geeks who know if someone has tried to study this effect in this fashion?

I hope so, because if no one tells me YES, I may sink a week of my life into doing this study, and I ain’t got that much time free these days.

2

Any study is going to be too flaws to be useful. The only way to get a true picture is to have the player silmultaneously playing a season in both the minors and the majors – a physical impossibility.

Players are not alike in talent, and part of the minors is the learning process. So if you compare years, you are comparing how much the player has learned. Some major league superstars has comparatively poor minor league careers; others had good ones. Some borderline major leaguers were minor league superstars; others were just adequate.

It all boils down to their talent and personality.

However, it’s pretty consistent that major league players sent down to the minors for rehab stints hit or pitch fairly well during that time.

3

There are methods for converting minor league stats to “equivalent” MLB stats, but I don’t know that anyone’s ever tried going the other way with it.

You also have to figure in the fact that the minors in the early 20s were nothing like the system today (Branch Rickey was only just starting to develop the farm system in the Cardinals organization in the 1920s.)

4

Well, Bonds in his prime, then. The point is that there’s got to be some upper limit on what a great MLB player could do in the minors, and I submit that we’re not within a hundred batting points of guessing what that would be. If you’d suggest that Willie Mays in 1954 would have batted .400 in Minneapolis and I’d suggest he would have batted .600, there’s no way to tell who’s totally off the charts, which is bizarre considering that we know what the approximate limits on Mays batting in Ebbets Field or Yankee Stadium would have been if the Giants had traded him there.

5

Also, it must be remembered that not all minor leagues are equal. Is he being sent down to AAA? AA? A? One of the now-defunct classifications (B, C, and D)? This can change the answer quite a bit.

6

Just for fun I’m going to throw out my belief that Ruth wouldn’t have been amazing in today’s minor leagues. He got a hell of a lot of cheap home runs because of the then-tiny size of Yankee Stadium’s right field, he couldn’t use as heavy a bat as he did, and he was out of shape.

He’d probably make the majors, but not be an allstar - about like a Prince Fielder, I’d imagine.

7

Why is it bizarre? Not everything is quantifyable. There is no way to accurately say what he’d do, any more there is no way to accurately say what a minor leaguer would do when promoted. At best, you can come up with probabilities based on historical trends, but any individual player is going to vary from those probabilities, some in unexpected ways.

A player may be great in the minors, but have trouble hitting breaking pitches, which you see less of in the minors. Another player may have a great talent for hitting breaking pitches, but is only average for fastballs. He might do well in the majors. A player may tear up the minors, but have injury issues.

There just is no way to accurately predict. You can make guesses – either by playing with numbers or by watching the player in action – but it’s still a crapshoot.

8

Because we don’t know something, or because current methods of analysis are still being worked out, is not to say we can’t know it, or that knowledge isn;t useful. To take an obvious example, look at the work done on MLE equivalences–people used to claim (I think you’re still claiming it) that MLB performance was utterly unpredictable, but now most people accept the surprisingly narrow ranges of MLB play that derive from minor league stats, properly read.

Once we come up with the formula I’ve suggested is possible, we will know what the average fluctuation would be. If it’s, say, 50 points of BA for ALL MLB players, of course you can never state that Willie Mays would have therefore batted 50 points higher if he was left in the minors. But if it’s 50 BA points, and 99% of all players are within 30 points of that figure, plus or minus, then you can state with some very high degree of probability what you can expect Mays to have done in the minors in 1954. I would be interested in knowing, for now, what you think the average difference would be? Is it 100 points for Mays and nothing for a scrub? Is it equal for all major leaguers? Is it more for a scrub than it would be for a superstar? Is it more like 100 BA points, or 200, or 400? If you want to stick with “Nobody knows” I’ll award you the Captain Obvious medal, but how much do THINK it would be?

As to the Governor’s point about the level of minor league ball, you’ll find very few players jumping whole levels of play. If you want to make a big deal out of the various levels, we could stipulate that the comparisons are all MLB:AAA. What I’m wondering, in case my point got lost, is whether anyone has actually done such a study?

9

That that assumes baseball is a mathematical construct. It’s made up of people, and people do not behave in mathematically predictable ways.

The various “prediction” made by math analysis in baseball can only come up with a probability. Ultimately, they are no more accurate than a coach evaluating a player based solely on experience.

Mays hit .354 in 1954. If it’s 50 BA points, and 99% of all players within 30 points of that figure, then by math, you’ve shown Mays most likely would have hit between .354 and .434. A range that big is useless.

I predict that if Mays had played, he would have batted between .000 and 1.000. My prediction is 100% accurate – and just as useless as gathering the numbers.

Also, there’s also the possibility he could have hit worse in the minors. Remember, the numbers you use are average, but few players are exactly average. Last year, the National League batting average was .262. How many players hit that? (Just one: Brad Hawpe of the Rockies). So the average tells you nothing. Ah, but then you say x% were within y percentage points of that number. So? What you’re doing it just shooting at the side of a barn and then drawing the target where the bullets hit. It’s not a prediction – or, rather, it’s so vague as to be useless. (Of the 99% who improved 50 points, where does Mays fall? Top tier? Bottom tier? You can’t predict.)

You are flipping coins. You can predict the odds, but you cannot say for certain if any one flip is going to be heads or tails.

Basically, all the math does is give you a meaningless number to pick out of your hat.

10

There are certain things which we can say and a bunch of things we can’t.

First, let’s ask whether Ruth was a statistical outlier or just part of the top 1% of players all of whom were about equal in ability? The answer has to be that it’s the latter. Sure, Ruth was way ahead of everybody else in showing that home runs could be hit and play a major role in the game - nobody really even tried to do this before - but once they got the message they caught up pretty quickly. Rogers Hornsby in 1922 hit .401, with 42 home runs and 152 RBI. OPB + Slg = 1.181. 1922 was a partial year for Ruth so I’ll use 1923 for comparison. .393, 41, 131, 1.309.

Lou Gehrig was a few years younger than Ruth but their batting lines in the years they played together are interchangeable.

And Ruth never won a triple crown. Other batters kept beating him not only in average but in RBI.

Part of this was because batting numbers were inflated throughout the 1920s, for reasons that people keep arguing about, culminating in the ridiculous year of 1930, when the National League as a whole averaged .303 and the aptly named Hack Wilson hits 191 RBIs (with 56 homers and a .356 average).

Various people have proposed various corrections to the problem of inflated (and later deflated) statistics - comparing them as a percentage of the league average for the year seems to be the best simple route - but the main point is that there is never an era in baseball when one person’s overall numbers overshine everyone else’s.

So the next question is, how do the best 1% in the majors compare to the best 1% in the top minor leagues. As mentioned earlier, the modern farm team system did not arise until Branch Rickey developed it in the 1930s. Before then, some minor league teams were independent powerhouses and attendance draws on their own, especially in cities that deserved major league franchises, like Baltimore and Los Angeles. They were likely to keep their best players for themselves and not sell them to the majors unless they had to. Some very fine players had their entire careers in the minors.

Then there were the Negro Leagues. Most baseball historians say that Josh Gibson was the equal of Ruth as a batter, and make the cases for several others. The circumstances of the leagues were so different from the formal structure of the majors that I don’t know anyone who has succeeded in extrapolating the stats from one to the other.

So. Back in the heyday of Ruth, the differences between the best players in the majors and best players outside probably were fairly small. This is true probably all the way down. There were only 16 major league teams, but dozens of AAA and Negro League teams of near-equal stature. Today, the situation is totally different. There are 30 major league teams and the minors are appendages, stripped of anyone with any value as soon as they show promise.

Bill James has developed a set of projections that take into account league, park, competition and other effects to show that minor league stats today are very good indicators of performance in the majors. Even so, the projections often don’t work well because there isn’t enough of a statistical base - people are called up too quick - to base the numbers on.

What you want to do prr is find a statistical basis for something that has no statistical base. There are many cases of older players moving back down into the minors - Fred Merkle had four great years for the Rochester Red Wings after leaving the bigs - but there is nothing that can compare players at their primes. It’s all statistical guesswork off of indirect indicators.

You might be able to do it, but I suggest that before you waste a week of your time you investigate The Society for American Baseball Research (SABR) first. If a study has ever been done they’d know about it. Or about why it can’t meaningfully be done.

11

IOW, you have no idea what Mays would hit in the minors, if he’d played there in 1954. Absolutely none. That being the case, why intrude yourself in here to say as much?

Now, does anyone who’s more knowledgable about baseball want to venture a guess?

Exapno, I’m really very familiar with James’ Major League Equivalents, but I don’t know why you say that they don’t work very well. After being in use for over a decade, they’ve actually worked out quite well, and James is very satisfied with them as a predictive tool. When he began his MLEs, the baseball world was basically like Reality Chuck–can’t be done, impossible, no meaningful results, every player is different, how dare you presume, etc. Now that his MLEs work out MUCH more accurately than most eyeball guesses, it’s only the most retrograde types who reject MLEs --probably most MLB clubs use them, or a form of them, regularly, some without even acknowledging James’ service to them.

What I don’t think people generally realize is how close the majors are to the minors in quality. Often a young player will yoyo between MLB and minor league baseball, getting called up and sent down several times a year, with very comparable stats at each level. If there were a huge difference between the two levels, very few minor leaguers would be able to survive MLB pitching, but many do, and very few suffer a devastating difference in their stats.

Which is not to say that it’s an easy gap to breach, only that the statistical differences may not be very large. A .280 batter who gets called up may hit .240 in MLB and never play in the majors again, but to me that suggests that only an .040 difference exists between the two levels of play, which is I think much smaller than many fans think.

12

Do you have any data to support that? The Babe hit MORE homers on the road than he did at home (347 home, 367 road) . Cite :rolleyes:

Maybe Ruth should’ve just stuck with pitching, where in had a major league record of 94-46 with a 2.28 ERA, plus a 2-0 World Series record with a 0.87 ERA before they decided he’d be even better as an everyday player.

“The Ruth is Mighty and Shall Prevail.” - Heywood Broun

13

I think you’re misunderstanding my point. MLEs work well, but only if given enough data to work upon. For some people they don’t work, because there just isn’t enough data to start with.

14

Didja have to do the rolleyes? What’s with the pissiness?

He still hit a lot of cheap homers in Yankee Stadium.

He also hit a lot of cheap homers at the Polo Grounds.

Because of the rolleyes, you made me waste an hour at these two sites:

http://www.ballparksofbaseball.com/Past%20Ballparks.htm

http://www.baseballhalloffame.org/hofers_and_honorees/baberuth/ruth_HR_facts.htm

breaking out Ruth’s homers by distance to right-field porch.

#homers rt. field foul pole
390 < 300
141 310-320
183 > 325

I picked the divisions of < 300 (because, c’mon - less than 300 feet!?), between 300 and 325 (because 325’ is the official minimum ‘regulation’ distance), and > 325 feet.

I still say a helluva lot of cheap homers.

15

Statistical comparisons won’t work. Stan Musial hit .326 at AAA Rochester. The next year, with St. Louis, he hit .315. Obviously a drop off. Except that at age 41, Musial hit .330 – higher than his minor league average.

The minor league record for home runs was 72, set by Joe Bauman in 1954, playing for Class C Roswell, New Mexico. Roswell is 3,500 feet above sea level, and Bauman never played in the majors.

16

You can’t extrapolate off of one data point. The idea behind MLEs is that the totality of a minor league career - when corrected to account for age and other variables - is a meaningful guide to the totality of a major league career.

You can’t even use one year’s batting average in the majors to predict the next year’s batting average in the majors. But you can use the numbers to tell you who is likely to be worthwhile next year and who is likely to lose ground.

It’s not like any of this is new. Sabermetrics has been around for 30 years. It demonstrably, mathematically works. It’s like an open source computer system: everybody can look at the numbers and at the math and see whether it works and how to make it better.

What happens if you correct Bowman’s numbers for the league’s average, for the park he played in, for the quality of his opponents? The raw numbers by themselves are meaningless. He was a very good low minors player. Would he have put up those numbers in the majors? Obviously not. Would he have been good had he been given the right chance at the right time? Probably. Nobody is trying to say any more than that.

You have to understand that good predictive ability is not the same as perfect predictive ability. You have to know how to work the numbers so that they are meaningful. You have to be able to apply context. But you can’t dismiss solid math with non-mathematical arguments like yours.

17

The rolleyes were - and are - necessary because you appear to be working on the iincorrect assumption that what you say in this forum doesn’t have to be supported by evidence when that evidence is available. I showed that Ruth hit more homers on the road than he did at home, which refutes your statement to the effect that he only hit a lot of homers because he played in home parks with relatively short porches down the right field line. You countered by counting all the home runs hit in New York parks and said that they must all have been short pole home runs. Tell me, how do you know that? How do you know that Ruth didn’t hit many home runs in his home park of much greater distance that didn’t just drop into the first row of the short porch. Until and unless you can demonstrate that your opinion is baseless and deserves all the rolleyes I want to throw at you.

18

Someone mentioned that B. Ruth’s heavy bat is not allowed today – is there actually a rule to limit the length and weight of a MLB bat ? I know there is a limit for the diameter …

19

Rule 1.10
a) The bat shall be a smooth, round stick not more than 23/4 inches in diameter at the thickest part and not more than 42 inches in length. The bat shall be one piece of solid wood.

Nothing specifically about weight that I see.