Great major leaguers who have poor minor league stats

[ctorg]

As discussed in another thread, it's pretty common in OOTP for players who turn out to be very good to do poorly throughout their minor league careers, while poor players who are being sent down a lot tend to do pretty well down there. There seems to be a pretty low correlation between how players do in the majors and how they do in the minors. While this does happen sometimes in MLB, it's more often that good MLB players had a lot of minor league success.

The main probable cause I can think of is that the competition level of the minors, despite whatever settings are used, is too high. Maybe there are too many borderline major leaguers feasting on inexperienced minor leaguers. I know I'll often send down a guy who has a 30/80 overall rating and see him do very well against all the minor leaguers with 21/80 overall ratings. Are there too many good older players in the minors? Are AAA players not really good enough to be in AAA?

Any ideas on how this may be solvable?

[1998 Yankees]

Originally Posted by ctorg

Any ideas on how this may be solvable?

I thought this was a most astute comment in that thread:

Originally Posted by RonCo

As I said in another thread, part of the issue is that in the past there was no way to tweak the minor league's league totals. I assume they were all based on the main major league totals. V2007 allows you to modify those totals, hence allow for more realistic creation of minor league numbers. We still have an issue with overall minor league talent levels, but one step at a time.

That sounded like a solution in the works.

[Zeyes]

I can't speak to your larger question, but one thing...

Originally Posted by ctorg

I know I'll often send down a guy who has a 30/80 overall rating and see him do very well against all the minor leaguers with 21/80 overall ratings.

As far as I can tell from the OVR ratings scale, a 30/80 player is quite a bit stronger than AAA level. From what I've seen in my leagues, the prototypical 25th man on the roster will have an OVR of about 25 or 26.

It's hard to tell because it depends in part on how many levels of minors you have, but 21/80 seems pretty weak to me compared to that.

I'm not sure if this will work as I imagine it should (and I can't check because I have a sim going right now), but if you're willing to experiment: un-affiliate your AAA league, then Act As one of its teams, and you should get OVR scores relative to that league alone. I suspect a player who was formerly 30/80 with reference to the major league will be significantly better than 50/80 in AAA (probably in the 70s). Also, switching the OVR/POT calculation to "all players" instead of "by position" might be useful for this.

[Assos]

I have had some regular AA players who hit in the low .200s and I needed them to come up to fill a backup spot in the majors and would hit, mind you, in a part time role, in the .280s+. So, it does happen.

[ctorg]

Originally Posted by 1998 Yankees

I thought this was a most astute comment in that thread:

That sounded like a solution in the works.

I don't know that the league totals are what is off, though. I haven't checked. But I do know that individual performances seem to be not quite right.

[Mike Donlin]

Not really poor, but Grady Sizemore for Cleveland is posting better stats as a major leaguer than he ever really did as a minor leaguer.

http://www.thebaseballcube.com/playe...Sizemore.shtml

[spark240]

In real life, players have breakout (or collapse) seasons, where they achieve a level of ability much better than before (or the reverse), and sometimes a breakout occurs in the season of, or just following, a promotion to the big league. The key is that such a statistical break isn't any more (or less) likely to occur between a minor league season and a major league season, as it is between two minor, or two major, seasons.

Sizemore has developed more home run power lately, and has clearly continued to grow as a player generally. Given his age, I think his growth tracks pretty well from the talent that he definitely showed with the K-Tribe and Aeros (the two key levels in a fantastic Indians organization, by the way).

[ctorg]

Originally Posted by Mike Donlin

Not really poor, but Grady Sizemore for Cleveland is posting better stats as a major leaguer than he ever really did as a minor leaguer.

http://www.thebaseballcube.com/playe...Sizemore.shtml

Yeah, it definitely happens in real life, but it seems to happen more in OOTP.

Originally Posted by spark240

In real life, players have breakout (or collapse) seasons, where they achieve a level of ability much better than before (or the reverse), and sometimes a breakout occurs in the season of, or just following, a promotion to the big league. The key is that such a statistical break isn't any more (or less) likely to occur between a minor league season and a major league season, as it is between two minor, or two major, seasons.

Sizemore has developed more home run power lately, and has clearly continued to grow as a player generally. Given his age, I think his growth tracks pretty well from the talent that he definitely showed with the K-Tribe and Aeros (the two key levels in a fantastic Indians organization, by the way).

Perhaps the AI's immediate knowledge of a player who has had a substantial ratings increase leads to the AI promoting a player immediately who in real life would still have had to prove himself ready at AAA.

[Mike Donlin]

Originally Posted by ctorg

Yeah, it definitely happens in real life, but it seems to happen more in OOTP.


Perhaps the AI's immediate knowledge of a player who has had a substantial ratings increase leads to the AI promoting a player immediately who in real life would still have had to prove himself ready at AAA.

Right. Except of course that some prospects are now jumping from AA to the majors. But this concept of "proving oneself" is what is missing from OOTP minors. Players should have to prove themselves at each level in order to earn a promotion. As it stands now, someone could have a terrible year at Single A but because of a ratings bump be promoted and have a great year at AA. That wouldn't happen in real life. After a few chances, the player would be cut.

I think there ought to be some kind of rule wherein, with very few exceptions, players cannot be promoted to the next level of the minors if they do not at least hit league average or better for at least half a season at a given level. That may not work for Rookie and Short Season leagues, but it should be the case elsewhere.

[Mattingly06]

Great topic, I too would like to see changes in this area. Minor league stats is one of the biggest topics in sabermetrics (I'm only a casual reader of their work) leading to many different projection systems ie MLEs and ZIPs. They believe minor league stats have as much or more predictive value than major league stats, after all how often does a player hit with no power down there and blast 50 hr in the majors? I might be able to name a few but they're exceptions and not the norm. This is a very important topic for OOTP players who like to watch players develop and perform in the minors. It looks very awkward when my 5.00 ERA 3 k/9 AAA pitcher dominates in the bigs.

Originally Posted by ctorg

Perhaps the AI's immediate knowledge of a player who has had a substantial ratings increase leads to the AI promoting a player immediately who in real life would still have had to prove himself ready at AAA.

This might be the key to the issue. I have one question though in regards to the morale system. When a player receives a talent bump, sometimes I notice that his expectations go up and wants a shot at the majors. If OOTP 2008 delayed his callup until he actually performed in the minors, would he get mad and consequently regress?

[injury log]

Originally Posted by Mattingly06

Great topic, I too would like to see changes in this area. Minor league stats is one of the biggest topics in sabermetrics (I'm only a casual reader of their work) leading to many different projection systems ie MLEs and ZIPs. They believe minor league stats have as much or more predictive value than major league stats ...

'More' predictive value might be going too far, but I essentially agree if you're discussing the higher levels of the minors, provided age is taken into account and stats are appropriately translated. I don't think you'll find many sabermetricians who believe there is much predictive value in Rookie or short-season A stats. Nate Silver, author of the PECOTA projection system, comments on this a lot, and usually defers to scouting judgments when talking about prospects in the low minors.

[RonCo]

Sabermetricians want to understand the environment better to suggest projectability As better definition of league-wide stats and parks effects are understood, sabermetric confidence will increase. But it's only very recently that low level minor league stats have ever had any real context associated with them.

[RonCo]

In all seriousness, I've had pretty good success with molding minor league stats with the minor league League Totals. They need to be considerably different from major league totals.

[No Pepper]

ctorg, in addition to playing around with the minor league League Totals (modifiers?), you could adjust how the AI evaluates players. Putting more weight into statistics (current or past) than into ratings might be an idea. What are your current settings?

[Neags23]

Originally Posted by RonCo

In all seriousness, I've had pretty good success with molding minor league stats with the minor league League Totals. They need to be considerably different from major league totals.

What settings would you recommend for each level?

[ctorg]

Originally Posted by No Pepper

ctorg, in addition to playing around with the minor league League Totals (modifiers?), you could adjust how the AI evaluates players. Putting more weight into statistics (current or past) than into ratings might be an idea. What are your current settings?

Right now, I have everything at default. I've been hesitant to make any changes to my current league because I've built up 110 years of history that I don't want to come crashing down. I haven't minded the minor league oddities much because, to be honest, I almost never look at minor league stats. In fact, if we had the old option to not have minor league games play out, I'd probably choose it.

But I think enough people have raised the issue that I wanted to start a thread about it.

[Assos]

The jumping of AAA has another reason, some teams may have a stud in AAA that they would want to keep him there to keep his "clock" from running down, so they add a pretty good AA player who plays the same spot to keep the roster warm while the injuried MLB players is ready to come back and this way, the clock starts for the AA player who may or may have never made the big show, the AAA stud is not on the ticking clock and management is happy.

[Neags23]

Originally Posted by ctorg

Right now, I have everything at default. I've been hesitant to make any changes to my current league because I've built up 110 years of history that I don't want to come crashing down. I haven't minded the minor league oddities much because, to be honest, I almost never look at minor league stats. In fact, if we had the old option to not have minor league games play out, I'd probably choose it.

But I think enough people have raised the issue that I wanted to start a thread about it.

So at what point do you decide to promote/demote?

[RonCo]

Originally Posted by Neags23

What settings would you recommend for each level?

It's hard to call, really. I'll post the values I created using our online league, but realize that stats are created in OOTP by an interaction of the league totals, the ballparks in the league, and the talent/skills in each league...so my totals may need to be tweaked to work for your league.

That said, I created these values by gathering AAA, AA, and A stats for real life 2006. I then calculated all the rate-based values required (hr/g, r/g, bb/g, etc), and started twiddling with the totals and running test sims until the OOTP minors were operating within a stone's throw of each.

You can see below the values we use today. We tweak them a little because we spilt our minors up into 2 AAA leagues, one AA league, and 3-A level leagues, each with their own flavor. Note that I base everything off 1,000,000 AB just to make it easier to calculate changes.

	Maj 	AAA-1 	AAA-2 	AA 	A-1 	A-2 	A-3
AB 	1000000	1000000	1000000	1000000	1000000	1000000	1000000
H 	280000 	268000 	268000 	282000 	282000 	278000 	275000
D 	47000 	60000 	59000 	51000 	52000 	51000 	50500
T 	7600 	7000 	7000 	6500 	6000 	5500 	5300
HR 	29000 	21500 	21750 	22000 	15000 	12000 	13000
BB 	114000 	90000 	90000 	94000 	94000 	92000 	90000
HP 	10500 	11000 	11000 	11000 	12000 	12000 	12000
K 	197000 	165000 	165000 	120000 	110000 	115000 	117000
BABIP 	0.3 	0.316 	0.316 	0.306 	0.29 	0.315 	0.325

[ctorg]

Originally Posted by Neags23

So at what point do you decide to promote/demote?

Actually, for most of the time in my current league, I was just commish and didn't worry about it. Now I base it on ratings. I generally move at a much faster pace than most people through seasons, at least in this league. I have my minors auto-managed and bring up guys when their ratings look good enough based on my needs. I pretty much ignore minor league stats completely, which is partly born out of them not being realistic and partly born out of me being impatient.