 |
View Old Forum Thread
Old Forum Index » Sim Forums » Baseball Game & Site Questions » Enigmas
Frunobulax
I was wondering - can Enigmas have different curves in one season depending on majors and minors?
Consider this guy: [url=http://simdynasty.com/player.jsp?id=12344777&statsorimps=imps]Mike Young[/url]
After the first season I immediately thought bust in minors, gem in majors. But his conversions in the majors in the second year were horrible. So I guess he is an enigma after all, that for some reason converted like hell in his first year in the majors but didn't convert anything at all in the minors?
Best, F.
Enigmas
January 16, 2017 at 11:52AM View BBCode
Hi,I was wondering - can Enigmas have different curves in one season depending on majors and minors?
Consider this guy: [url=http://simdynasty.com/player.jsp?id=12344777&statsorimps=imps]Mike Young[/url]
After the first season I immediately thought bust in minors, gem in majors. But his conversions in the majors in the second year were horrible. So I guess he is an enigma after all, that for some reason converted like hell in his first year in the majors but didn't convert anything at all in the minors?
Best, F.
timemry
January 18, 2017 at 04:36AM View BBCode
You should never have a pitcher this young in the Majors. You haven't had him in the minors (where he belongs) long enough to determine any type of curve. He is probably ruined at this point.
gannable
2 out 28 in the minors is not unusual
you need to give him a chance
January 19, 2017 at 03:12AM View BBCode
you overplayed your hand2 out 28 in the minors is not unusual
you need to give him a chance
Frunobulax
But you both don't answer my question.
January 19, 2017 at 06:42AM View BBCode
I might have put him too quickly in the majors, true. He converted badly (30% of expected) and I figured I would try him in the majors, there he converted well (180%) so I figured bust/gem with high probability. And if he would have been bust/gem, then he would have been better off in the majors than with another season in the minors (at low CPs in system 5). In VarDev you always have to play the percentages, if you hit it 90% of the time then you're doing well.But you both don't answer my question.
WillyD
No
January 19, 2017 at 01:51PM View BBCode
Originally posted by Frunobulax
I was wondering - can Enigmas have different curves in one season depending on majors and minors?
No
allenciox
Here is what it shows for Mike Young:
In 2168 (his rookie year), in the minors, he should have converted his ctrl/vel at a 24.5% rate. There is only a 1.5% chance that a normal player is going to convert as poorly as he did.
That same year in the majors, he should have converted at an 18.7% rate. He actually converted at exactly the predicted rate, making his conversion essentially average. But since it is a fairly small sample size, he could have been a bust that got "lucky" (note, of course, the lower conversion rate in majors than minors for this age).
The next year is the kicker: the probability that a normal player would have converted as poorly as he did in the majors in 2169 is less than 1%: about 0.85%. The following year, he slighly underperformed in the minors (about a third of a chance a normal player would do that poorly) and the following year, he did even worse (about a 6% chance that a normal player would do so poorly).
So he only has one result that meets an average result, and many that are quite a bit worse than average. He has not hit his stride yet. Either he will always be a low performer (a full bust), or his good years are yet to come.
Though I can't be sure, I do think it is more likely that he is a Eureka or late bloomer than a pure bust: though I cannot be sure, most full busts I have seen are REAL busts: for example, a minors bust, majors gem usually often shows probabilities below 1% for most minor years and above 98% for most major years. But Eurekas, Enigmas, late bloomers, and young major leaguers, just from my experience; are generally harder to pin down because their off years are not off by as much.
February 20, 2017 at 10:49PM View BBCode
I hate to tell you this, but I think Mike Young is either a bust or a late bloomer (or he could be an Eureka who hasn't hit his good year(s) yet). I wrote a program whereby I can paste in a player card and it uses empirically derived age/rating conversion probabilities plus combinatorics to figure out the probability that a 'normal' player would get the results for each of the rows on the improvements matrix.Here is what it shows for Mike Young:
In 2168 (his rookie year), in the minors, he should have converted his ctrl/vel at a 24.5% rate. There is only a 1.5% chance that a normal player is going to convert as poorly as he did.
That same year in the majors, he should have converted at an 18.7% rate. He actually converted at exactly the predicted rate, making his conversion essentially average. But since it is a fairly small sample size, he could have been a bust that got "lucky" (note, of course, the lower conversion rate in majors than minors for this age).
The next year is the kicker: the probability that a normal player would have converted as poorly as he did in the majors in 2169 is less than 1%: about 0.85%. The following year, he slighly underperformed in the minors (about a third of a chance a normal player would do that poorly) and the following year, he did even worse (about a 6% chance that a normal player would do so poorly).
So he only has one result that meets an average result, and many that are quite a bit worse than average. He has not hit his stride yet. Either he will always be a low performer (a full bust), or his good years are yet to come.
Though I can't be sure, I do think it is more likely that he is a Eureka or late bloomer than a pure bust: though I cannot be sure, most full busts I have seen are REAL busts: for example, a minors bust, majors gem usually often shows probabilities below 1% for most minor years and above 98% for most major years. But Eurekas, Enigmas, late bloomers, and young major leaguers, just from my experience; are generally harder to pin down because their off years are not off by as much.
Frunobulax
I also have a little program to calculate conversions vs. some values that I expect. He's currently at 47 velocity, 62 control.
In the first year I expected him to convert 6.8 ICs in the minors, he managed 2 (30% of the expected conversions). He did convert 3 in the majors, I expected 1.6 (183%).
The next year he was in the majors and converted 17% (1, expected 5.7)
The year after that in the minors he started well, but finished weak, converting 7 while the average player would have converted 9 (78%).
The current year he's 28% in the minors (1, expected 3.5).
He looks to be either a crazy Enigma (I found that there are plenty of Enigmas converting on average either much better or much worse than the regular players, this might be an example of much worse) or an Eureka that has good years at OS18 (before he was drafted) and maybe 1-2 good years at OS23-OS25.
I would rule out bust (too irregular conversions) and late bloomer (who convert normally in the minors).
February 21, 2017 at 08:45AM View BBCode
I came to the same conclusion, and as long as I'm rebuildung I'll carry him in the majors, where it costs me no CPs to develop him. He was a top prospect once, but chances are that he lands on the waiver wire...I also have a little program to calculate conversions vs. some values that I expect. He's currently at 47 velocity, 62 control.
In the first year I expected him to convert 6.8 ICs in the minors, he managed 2 (30% of the expected conversions). He did convert 3 in the majors, I expected 1.6 (183%).
The next year he was in the majors and converted 17% (1, expected 5.7)
The year after that in the minors he started well, but finished weak, converting 7 while the average player would have converted 9 (78%).
The current year he's 28% in the minors (1, expected 3.5).
He looks to be either a crazy Enigma (I found that there are plenty of Enigmas converting on average either much better or much worse than the regular players, this might be an example of much worse) or an Eureka that has good years at OS18 (before he was drafted) and maybe 1-2 good years at OS23-OS25.
I would rule out bust (too irregular conversions) and late bloomer (who convert normally in the minors).
allenciox
February 21, 2017 at 02:44PM View BBCode
Actually, late bloomers don't convert normally in the minors in their early years. According to the game guide, late bloomers convert worse than average in minors until age 22 (in the majors, it is up through age 24). And although not everything in the game guide meets up with my observations (for example, ageless wonders usually start their growth at os_age 33 instead of os_age 34), this does.
dirtdevil
you do see that you're basing career and franchise defining decisions on snap conclusions derived from ridiculously small sample sizes, yes?
February 21, 2017 at 03:06PM View BBCode
Originally posted by Frunobulax
In the first year I expected him to convert 6.8 ICs in the minors, he managed 2 (30% of the expected conversions). He did convert 3 in the majors, I expected 1.6 (183%).
The next year he was in the majors and converted 17% (1, expected 5.7)
The year after that in the minors he started well, but finished weak, converting 7 while the average player would have converted 9 (78%).
The current year he's 28% in the minors (1, expected 3.5).
you do see that you're basing career and franchise defining decisions on snap conclusions derived from ridiculously small sample sizes, yes?
Frunobulax
Obviously. Those small sample sizes are all we have in judging development curves, especially for pitchers.
Still, it's better to act on small sample size than to ignore the data, even if we have a certain error rate.
February 21, 2017 at 03:11PM View BBCode
Originally posted by dirtdevil
you do see that you're basing career and franchise defining decisions on snap conclusions derived from ridiculously small sample sizes, yes?
Obviously. Those small sample sizes are all we have in judging development curves, especially for pitchers.
Still, it's better to act on small sample size than to ignore the data, even if we have a certain error rate.
dirtdevil
act, yes. overreact, no.
February 21, 2017 at 04:53PM View BBCode
Originally posted by Frunobulax
Still, it's better to act on small sample size than to ignore the data, even if we have a certain error rate.
act, yes. overreact, no.
allenciox
I have also thought about analyzing this from a Bayesian perspective, so that I can actually make some tentative decisions (regarding CP distribution, for example) even very early on in a players' career. I am still fine-tuning that approach.
February 21, 2017 at 05:17PM View BBCode
The method I am using actually comes out with probabilities... when you see a sufficiently large pattern in those probabilities the sample sizes are sufficient to draw conclusions.I have also thought about analyzing this from a Bayesian perspective, so that I can actually make some tentative decisions (regarding CP distribution, for example) even very early on in a players' career. I am still fine-tuning that approach.
Frunobulax
About Young, he converted 28% of the norm in the minors (1of 3.5) but 151% in the majors (8 conversions, 5.3 expected). I guess the answer to my original question is "yes", with some doubt remaining.
February 27, 2017 at 10:56AM View BBCode
I always check how players are converting in the minors about every 2 months, and adjust CPs if I see a pattern. After some time you get a pretty good idea what's a normal variation and what's significant, even though there are exceptions.About Young, he converted 28% of the norm in the minors (1of 3.5) but 151% in the majors (8 conversions, 5.3 expected). I guess the answer to my original question is "yes", with some doubt remaining.
Pages: 1