2005 Major League Equivalencies for Brewers prospects.

[JoeHova]

Ok, this is some interesting stuff. Major League Equivalencies take what a player did in the minors and translate that to what the player would have done in the majors. In other words, it's not a projection for the future, but rather what a player would have done had they spent the year in the majors. MLEs have a long track record of accuracy, although there will always be outliers. Anyway, here are 2005 MLEs for people of interest to BF.net:

 

(avg./obp/slg)

 

Prince Fielder- .262/.345/.492 (this suggests he's already better than Lyle)

Rickie Weeks- .286/.383/.562 (too bad about his injury, cost him ROY)

Corey Hart- .277/.338/.466 (pretty much exactly average for a corner OF)

Nelson Cruz- .250/.331/.445 (slightly below average for a corner OF)

Mike Rivera- .257/.287/.495 (that power would look pretty nice in Mil)

Ryan Braun- .257/.285/.421 (already a .700 OPS, he really raked in R & A)

Hernan Iribarren- .210/.254/.259 (still far from the majors)

Alcides Escobar- .196/.217/.250 (even farther than Hernan)

Dave Krynzel- .227/.285/.362 (he needs to start hitting at some point)

Tony Gwynn- .230/.309/.285 (needs more time in the minors)

Lou Palmisano- .213/.258/.294 (so does he)

Brad Nelson- .237/.317/.352 (let's see about next year, it's not too late)

Enrique Cruz- .254/.297/.379 (this is probably about average for a UT)

Steve Scarborough- .228/.286/.364 (hey, it's better than Guzman)

Steve Moss- .233/.276/.330 (similar OPS to Gwynn)

Vinny Rottino- .253/.297/.333 (needs more time in the minors)

John Vanden Berg- .215/.303/.299 (not too bad, but needs improvement)

 

Unfortunately, Angel Salome isn't listed. According to the MLE's, the Brewers had 6 guys in the minors who could have slugged .400 in the majors this year: Prince, Rickie, Corey, and Nelson Cruz (the usual suspects, in other words) and Mike Rivera (he really deserves a call-up after the PCL championship series) and Ryan Braun (that is surprising to me, he's off to a heck of a start). They had 5 guys who would be about average or above for their positions (the big 4 hitters and Rivera) and two guys who would be about average as utility guys (Braun and Enrique Cruz). That doesn't seem too bad to me, especially when you consider that guys like Hernan, Alcides, Angel, and the CFs still have alot of potential to improve.

 

Thought people might be interested in this. It's pretty encouraging to me, especially as it regards Prince, Rickie, Corey, N. Cruz, and Braun. Rivera and E. Cruz were nice surprises this year and deserve to stick around the organization next year.

 

link: home.comcast.net/~briankaat/mle05.csv

 

edit: added two BF.net favorites

[molitor fan]

Thanks Hova, really interesting stuff......

[deeswan]

Joe, just curious, you mention that MLE's have a long track record of accuracy. What do you mean by that? Aren't these just projections based on minor league production? How is accuracy measured? Again, not trying to be a naysayer, I'm just curious.

[GenoSeligPrieb]

"it's not a projection for the future, but rather what a player would have done had they spent the year in the majors. MLEs have a long track record of accuracy, although there will always be outliers. "

 

Although my intense contempt for this statistic is well-documented, Joe, I understand you're simply providing a service of sorts here, so that's cool.

 

That said, if it can't be used as a projection or a predictive tool, then MLE's accuracy can't be measured honestly, right?

 

Anyone can say Corey Hart's performance at AAA therefore means he equated to as good as an average MLB corner outfielder, but what good is it if he's not dealing with MLB pitchers, ballparks, media, paychecks, crowds, teammates, pressures? The furthest I'd be willing to go would be to equate someone's minor league performance from the lower rungs of the ladder, i.e. High-A performace *1.123456789% = AA, or something similar. But as we've seen so many times, the leap to the majors is fraught with too many personal influences/issues to invoke any kind of formula.

 

I'm not calling you specifically on the carpet, JH...nothing personal. I just want someone who refers to this as gospel to prove why it's worthwhile to us skeptics.

[SheetsCyYoung04]

Just out of curiousity does that take park factors into account, so would Lou have the same projection if he had hit that way in High Desert?

[JoeHova]

yes, the MLEs take park factor into account.

 

the accuracy that I speak of is that a player's MLEs are usually a good indication of what a player will do in the future. At the same time, they aren't a projection: they don't take multiple years into account, they don't consider a player's age relative to league, etc.

[GenoSeligPrieb]

"the accuracy that I speak of is that a player's MLEs are usually a good indication of what a player will do in the future. At the same time, they aren't a projection"

 

Sorry, JH, perhaps I'm ignorant (perhaps?! http://forum.brewerfan.net/images/smilies/wink.gif ) but isn't "an indication of what a player WILL do in the FUTURE" a projection?

 

Again, it's not you here I have the issue with, it's this formula. Do you really believe in it? Exactly how accurate has it been? 20%? 60%?

 

I feel that a formula which is believed to be an accurate indicator of future performance stands alone when you can take, say, the last 3 years' sampling of numbers for a given pitcher or hitter, and see how closely it hews to current performance.

 

For example: Carlos Lee averaged 29 HRs and 97 RBIs with an .855 OPS from 2002 - 2004, all at the major league level, of course. It's September of '05, and CLee is at 30 HRs, 103 RBIs and an .840 OPS. While not every player is this consistent, this works for me...

[JoeHova]

Here is a quote on MLEs from Gary Huckaby, founder of Baseball Prospectus and a consultant with the Oakland A's:

 

Quote:

---

MLE's are just as good as major-league statistics

in predicting future performance, with the exception of strikeouts and walks.

---

 

he goes on to say that the errors for Ks and BBs are: "not consistently in either direction."

 

Sean Forman, proprietor of baseball-reference.com and a professor of mathematics at St. John's, says:

 

Quote:

---

Minor league batting statistics will predict major league batting performance with essentially the same reliability as previous major league statistics.

---

[JoeHova]

Quote:

---

Sorry, JH, perhaps I'm ignorant (perhaps?! ) but isn't "an indication of what a player WILL do in the FUTURE" a projection?

---

 

no, in the same way that what Billy Hall or JJ Hardy did this year is not a projection. It can be used as part of a projection, but it is not a projection, just a record of what they did.

[dixieflatline]

Quote:

---

Alcides Escobar- .196/.217/.250 (even farther than Hernan)

---

 

Alcides is an 18 year old who held his own in A ball this year. Hernan is a 21 year old who had a decent year in A ball. Of course Alcides is far from MLB ready but even a year at A+, then a year at AA, and a year at AAA puts him in the majors at 22. Even if he needs a repeat year along the way he makes the majors at 23 the same age as JJ. I do not understand why nobody here is even the slightest bit excited about him. I think that with four years of growth he coud put up compareable numbers to .237/.324/.372(J.J. this year). Sure he could flame out but this guy has a super high ceiling. He is a long way off but he is someone to watch IMO.

[JoeHova]

(Alcides) has a super high ceiling. He is a long way off but he is someone to watch IMO.

 

Yes, he is someone to watch, I agree. That's why I posted his MLE. At the same time, he's probably 4-5 years away at least, which is all I said. I wasn't trying to insult him at all.

[molitor fan]

Alcides is an 18 year old who held his own in A ball this year. Hernan is a 21 year old who had a decent year in A ball. Of course Alcides is far from MLB ready but even a year at A+, then a year at AA, and a year at AAA puts him in the majors at 22. Even if he needs a repeat year along the way he makes the majors at 23 the same age as JJ. I do not understand why nobody here is even the slightest bit excited about him. I think that with four years of growth he coud put up compareable numbers to .237/.324/.372(J.J. this year). Sure he could flame out but this guy has a super high ceiling. He is a long way off but he is someone to watch IMO.

 

I am very excited about him personally, though I know Toby is not....I see him as a potential Tony Fernandez Jr.

 

Errors at this level and at this age are wayyyyyyyy over-rated.

[nsimps]

Quote:

---

MLE's are just as good as major-league statistics

in predicting future performance, with the exception of strikeouts and walks.

---

 

Wouldn't that affect OBP or does it even out over the long run?

[DougJones43]

Quote:

---

Minor league batting statistics will predict major league batting performance with essentially the same reliability as previous major league statistics

---

so, according to this statement, the amount of certainty we have in a guy who is killing the ball in AA being successful in the majors at some point is equal to the amount of certainty we can have in albert pujols putting up good numbers next season?

 

i have to take issue with this statement, and this statistic. i don't see how minor league numbers could give any consistant indication of wheter or not a guy has potential at the major league level. having success at the minor league level does not seem to regularly translate into a hitter having what it takes to hit major league level pitching.

 

for example, take 2 former brewer prospects.

 

antone williamson: had a full 100+ game season at AA in which he posted an ops of .855

 

bobby hughes: posted a .963 ops in 67 AA games one season, and later posted a .864 in 89 games at AAA. i would imagine that these seasons would have translated through to be solid major league seasons using the MLE, while in reality, hughes most likely would have been blown away by major league pitching.

 

i don't know all the ins and outs of this formula, but wouldn't it have predicted some major league success for these guys? i would imagine it would have. but it became apparent that neither of them had the talent to be good major league hitters. i'm not trying to attack anyone or just point out exceptions to a stats usefulness, i just think that to say this stat is as reliable as previous major league statististics is way off the mark.

 

edit: i also don't think rickie weeks would have put up a .945 ops this season, injured or not

[endaround]

Puljos, no. The average major league player? Yes. Stats can't and won't be reliable for all palyers. All Joe is saying is that MLEs are just as unreliable as looking at major league stats going forward.

[DougJones43]

Quote:

---

Puljos, no. The average major league player? Yes. Stats can't and won't be reliable for all palyers. All Joe is saying is that MLEs are just as unreliable as looking at major league stats going forward.

---

i guess i just don't agree with that. i'll take an average major league player like you said, using previous major league stats.

 

over the past 3 seasons, damian miller has posted an avererage ops of .731. following the trend of his previous major league numbers, this season he is at .726.

 

the MLE theory will state that the chances of damian miller hitting close to his career averages will be no better than stating ryan braun would have posted a .700+ ops this season at the major league level. i just don't buy into the two being EQUALLY reliable. that's the whole thing i have issue with, is saying MLE's are equally reliable or unreliable.

 

edit: in response to "stats can't and won't be reliable for all players", this is obviously true and has been discussed a lot around here. the doesn't mean some stats aren't much more reliable than others. i just happen to believe that MLE is a stat that is far less reliable than others, although the nature of it (you can never truly say what a player would have done) makes it nearly impossible to prove this true or false.

[JoeHova]

Quote:

---

antone williamson: had a full 100+ game season at AA in which he posted an ops of .855

 

bobby hughes: posted a .963 ops in 67 AA games one season, and later posted a .864 in 89 games at AAA. i would imagine that these seasons would have translated through to be solid major league seasons using the MLE, while in reality, hughes most likely would have been blown away by major league pitching.

---

 

 

no, these would both translate to being bad major leaguers. An OPS of .855 in AA will translate to an OPS of well below .700 in the majors. An OPS of .864 in AAA (ignoring park factors because I don't know where he played) translates to an OPS of .705 in the majors. Hughes' career MLB OPS was .680. If anything, Hughes is good evidence that MLEs are pretty accurate.

 

 

 

 

Quote:

---

All Joe is saying is that MLEs are just as unreliable as looking at major league stats going forward.

---

 

well, I prefer to look at it as them being just as reliable, but the reverse is also true.

[BillScottCanRake]

All this statistic does is it converts a players 2005 minor league statistics and converts it into an equivelent ML line. It looks to me like it assumes a certain level of performance is needed to put up certain numbers in certain leagues, and it translates that performance into an equivelent ML statistic. A player who hit 25 AA home runs, if they could take those swings and ABs to Milwaukee, would have hit X number of home runs.

 

It doesn't try to project what sort of numbers a player will put up next year, because it doesn't take into account things like age, repetition of levels, return from injuries, etc. A 32 year old AAA OFer will probably have an artificially high ML Equivelent because they are really old for their league, and a 19 year old AAA player coming of elbow surgery might have an artificially low ML Equivelent.

 

What these numbers do is provide a sort of baseline from which we can adjust the numbers based on those factors not considered. This allows us to make an educated guess about next year should that year be in the majors. One could look at JJ Hardy's 2004 Major League Equivalency and see that he'd have a good shot at hitting pretty well once fully recovered from his injury and the rust were knocked off.

 

EDIT: Forgot to say thanks for posting that. It was an interesting read!

[JoeHova]

Quote:

---

i guess i just don't agree with that.

---

 

It doesn't matter if you agree with it or not, statistically it's true.

 

Quote:

---

over the past 3 seasons, damian miller has posted an avererage ops of .731. following the trend of his previous major league numbers, this season he is at .726.

 

the MLE theory will state that the chances of damian miller hitting close to his career averages will be no better than stating ryan braun would have posted a .700+ ops this season at the major league level. i just don't buy into the two being EQUALLY reliable. that's the whole thing i have issue with, is saying MLE's are equally reliable or unreliable.

---

 

no, it's not saying that. Miller has a track record, Braun has half of one season. That's why MLE's aren't projections. It will take a few years before we can become more confident that Braun didn't have a fluke year. Prince has a track record of domination in the minors, we can be fairly confident that he will be at least a good major league player, something that the MLEs backup. Braun has no track record, but is off to an excellent start.

 

So, recap:

 

MLEs=major league stats

long track record>short track record

 

Damian Miller had an 8 year track record of being a .750 OPS guy in the majors. If Braun was the same age as Damian and had an 8 year track record of being a .750 MLE guy while he was in the minors, we could say with roughly the same confidence that they would both put up an OPS around .750 in the majors. That's how it works.

[gregmag]

Joe, thank you for posting the info, and thank you for patiently, courteously taking the flak -- some of which is very reasonable and thoughtful, but some of which is just predictable beef from a predictable direction. You're a gentleman and a scholar.

 

Greg.

[Katuluu]

i guess i just don't agree with that. i'll take an average major league player like you said, using previous major league stats.

 

over the past 3 seasons, damian miller has posted an avererage ops of .731. following the trend of his previous major league numbers, this season he is at .726.

 

You do realize that by hand-selecting the player to participate in your study allows bias to play a role and thus reduces the validity of your findings and conclusions.

 

i don't see how minor league numbers could give any consistant indication of wheter or not a guy has potential at the major league level.

 

And this statement I'm almost at a loss for words. I was under the impression that statistics offered a direct correlation to show how effective a player was during a particular season at any particular level. Too completely ignore past success be it in the minors, high school, or college would be a truly unprecedented approach in baseball's history of player development. How would you even choose which players to promote? How do they apply only at the major league level, but cease to exist any where else?

[JoeHova]

Quote:

---

All this statistic does is it converts a players 2005 minor league statistics and converts it into an equivelent ML line. It looks to me like it assumes a certain level of performance is needed to put up certain numbers in certain leagues, and it translates that performance into an equivelent ML statistic.

---

 

 

yes. The way it works is similar to park effects. Think of AAA being like an even more extreme Coors Field, AA being like High Desert but with 280 foot fences, A+ being a park with 200 foot fences and where you are only allowed to use 3 infielders, etc. You would have to adjust all the raw numbers way downward in those cases because an MLB player would really dominate in those environments. That's why an .800 OPS in AA is not impressive in itself. It's important to look at the context in which statistics are produced because not every context is equal. What MLE's do is strip out all the noise of park and league effects.

[H0FFY]

You guys know I am NOT a sabr guy but I AM AM AM a stats guy. I never understand the arguments that show up around here.

 

Here we have a simple comparison of what guys did in the minors v what they did in the majors. And then other guys in the minors are not yet in the majors, but if they minor league numbers are plugged into the 'proven' formula, out the other end spits a MLB number.

 

Then we have a bunch of people who write 'nah, I dont believe that'.

 

The writer is only saying if you plug four chooks, a tomatoe and a can of beer into the portal between the minor and the majors, out the other end NORMALLY comes Albert Pujols. So, if I can collect the chooks and the tomatoe and the beer I 'likely' will see Albert stolling out of the portal.

 

Guaranteed? No.

Likely? Probably.

Unlikely? Probably Not.

 

But you cannot diss the study. It was done. (well, I am accepting that!)

[JoeHova]

well to expect a guy in AAA to come up and hit like pujols, he'd have to have about a 1.300 OPS in a neutral park. That's pretty rare, to say the least.

[deeswan]

The writer is only saying if you plug four chooks, a tomatoe and a can of beer into the portal between the minor and the majors, out the other end NORMALLY comes Albert Pujols. So, if I can collect the chooks and the tomatoe and the beer I 'likely' will see Albert stolling out of the portal.

 

 

Wow, you must be into sci-fi novels. Where can I find one of these portals? More importantly, what the heck is a chook?