International prospects: what's the point?

[ultramegaOK]

Recently, I have noticed that all the international prospects that my scout has discovered are all crap. All are at a level that I just end up releasing them. It's to the point where I just want to turn the whole international discovery thing off. My whole international roster is filled with players that top out at one or two stars. It's more of a pain in the butt to go through and delete them than it's worth. What's the point? Why even have the international prospect scouting if it is all crap? It's just a waste of time because, YAY I got a prospect whose potential is .05. So basically can I just turn this once intriguing (and fun) factor off? Because now I just waste my time deleting these (prospect) pieces of crap.

[Yankee Hotel Foxtrot]

Originally Posted by ultramegaOK

Recently, I have noticed that all the international prospects thhat my scout has discovered are all crap. All are at a level that I just end up releasing them. It's to the point where I just want to turn the whole international discovery thing off. My whole international roster is filled with players that top out at one or two stars. It's more of a pain in the butt to go through and delete them than it's worth. What's the point? Why even have the international prospect scouting if it is all crap? It's just a waste of time because, YAY I got a prospect whose potential is .05. So bascially can I just turn this once intriguing (and fun) factor off? Because now I just waste my time deleting these (prospect) pieces of crap.

Not my experience. I've had several start off say 25/36 who have developed into good players, one who is now 41/73 and about to press for MLB.
Yep quite a few who don't pan out or regress by the time they've developed, but more than enough who are worthwhile continuing to invest and persist with.
I do tend to focus on those who have positive personality traits like work ethic, leadership or IQ, abd/or adaptability.

[ultramegaOK]

I go on a 1-100 basis. Because I can't wrap my head around the whole 20-80 thing. That's just me. So here I am going through the season and my scout is exclaiming that he has found an international prospect. And low and behold it's a guy who's roof is 1.5 stars. Or less. Sometimes, yay, the prospect's ceiling is a whole 2 stars. Yay. a backup player at AAA at best. What's the point? I just go through all the "prospects" and release them because none have any future. What's the point of the whole International league if all their ceilings is nothing more than a backup AA player? At least, until recently my scout would find some gems in there. But now it's all just garbage. Maybe it's the latest update or whatever. But now, it is just trash. I have simulated several seasons and the top international prospect my scout has discovered is 2 stars. Yay. A freakin double A shortstop.

[kidd_05_u2]

Those prospects are supposed to fill your DSL teams, and very few of them are supposed to be good enough to ever get above low A. Once in a while you will have a scouting discovery turn into a top player, but the best international prospects are supposed to be in the IAFA pool.

If you don't like dealing with organizational filler, just delegate the handling of the minors to your coaches.

[millhousebc]

Agree........very rarely these guys make the jump up to the majors. Help fill the gaps in your minor league systems, otherwise complaints of "not enough players on my team" will start flowing in. Like Kidd said, the great thing about OOTP, don't like an option, turn it off or delegate it.

[Sweed]

Originally Posted by kidd_05_u2

Those prospects are supposed to fill your DSL teams, and very few of them are supposed to be good enough to ever get above low A. Once in a while you will have a scouting discovery turn into a top player, but the best international prospects are supposed to be in the IAFA pool.

If you don't like dealing with organizational filler, just delegate the handling of the minors to your coaches.

Originally Posted by millhousebc

Agree........very rarely these guys make the jump up to the majors. Help fill the gaps in your minor league systems, otherwise complaints of "not enough players on my team" will start flowing in. Like Kidd said, the great thing about OOTP, don't like an option, turn it off or delegate it.

This. ^^

The first thing I do when I get a low rated discovery is figure out how far he might advance in my minor league clubs before hitting his ceiling. By using the "relative to league" feature you can ask your scout how he projects to play at each level. If his potential shows he could get to A+ or AA I have a nice minor league roster filler for a few years. Who knows in that time I might get lucky and he'll train and develop to where he goes farther.

In my game I have both age and service time limits per level. I also control minor league signings and promotions/demotions. I am always looking for "decent" players to fill slots going into each new season. International players are a good source to fill some of those needs.

[snepp]

Lottery tickets that cost you nothing to acquire.

Once per season I mass select and release any player below 30 potential, takes all of 15 seconds to do.

[BaseballATeam]

Originally Posted by ultramegaOK

Recently, I have noticed that all the international prospects that my scout has discovered are all crap. All are at a level that I just end up releasing them. It's to the point where I just want to turn the whole international discovery thing off. My whole international roster is filled with players that top out at one or two stars. It's more of a pain in the butt to go through and delete them than it's worth. What's the point? Why even have the international prospect scouting if it is all crap? It's just a waste of time because, YAY I got a prospect whose potential is .05. So basically can I just turn this once intriguing (and fun) factor off? Because now I just waste my time deleting these (prospect) pieces of crap.

I don't care for the discovery process. I turn it off and increase the size of the IFA class to the maximum. That way, you can sign 1 big target, 6 medium ones, or 12 lesser ones. The game engine usually leaves about 30% of the IFA guys unsigned, which I figure is perfectly reasonable.

Don't forget to turn on ghost players!

[moshinturtle]

I view them just as filler for my DSL and FCL leagues etc when other guys age out or are promoted. If any of them turn into anything (rare because I play with TCR lowered somewhat) then great, but they're absolutely free to acquire lol.

[PSUColonel]

The problem is, too many younger, newer OOTP players want a game...not a simulation. IT's the same with the draft class potential ratings. The need to understand this is a realistic simulation of baseball, not a game that allows them to win at every turn.

[locuspc]

I mean, it is a game. Mechanics like the dev lab are put in because they're fun to play with, not because they accurately simulate real baseball development.

[Sweed]

Originally Posted by locuspc

I mean, it is a game. Mechanics like the dev lab are put in because they're fun to play with, not because they accurately simulate real baseball development.

Optional. Turn it off and don't use it. Problem solved.

[snepp]

While I do agree the dev lab is pretty "arcadey", it does have more in common with real life baseball than a couple rounds worth of players graded as generational talents.

[Pelican]

Originally Posted by ultramegaOK

I go on a 1-100 basis. Because I can't wrap my head around the whole 20-80 thing. That's just me.

Good on you! I am precisely the same. Could never use the 20-80 approach, that supposedly mimics an old school scouting system (although you will still see it referenced today). Fortunately the game easily and readily implements the 0-100 reference points. Others have warned that the information is not any more granular, and I get that. It's just that I grew up with a decimal system, and I can't for the life of me understand why I would use this odd scale for baseball scouting. Makes no sense.

[BaseballATeam]

Originally Posted by locuspc

I mean, it is a game. Mechanics like the dev lab are put in because they're fun to play with, not because they accurately simulate real baseball development.

I mean, I'd argue they do realistically simulate real player dev. Half the league jumps in the lab every winter and half of those come out looking different--not always better, usually marginally improved, and sometimes monstrously improved. Some few even break themselves.

My only complaint is the default setting underemphasizes the role of the dev lab in irl MLB. A sub-complaint--the cpu is just not very good at sensibly selecting players to put in the lab or proper focus traits. That's the next thing to fix imo.

[BaseballATeam]

Originally Posted by Pelican

Good on you! I am precisely the same. Could never use the 20-80 approach, that supposedly mimics an old school scouting system (although you will still see it referenced today). Fortunately the game easily and readily implements the 0-100 reference points. Others have warned that the information is not any more granular, and I get that. It's just that I grew up with a decimal system, and I can't for the life of me understand why I would use this odd scale for baseball scouting. Makes no sense.

The 20-80 system is more intuitive when thinking about player traits and how they fit in the context of roster building. It reflects a normal curve, so, if you're Bayesian-trained, you should prefer the 20-80 method.

Simplistically, a 50-grade trait is MLB average. 60 is one standard deviation better than average. 70 is two SDs, 80 is three SDs. A 90-grade is technically plausible but so rare as to not be assigned, it's a 1-in-31,500 outcome for an MLB player. According to gemini, there have been 22,664 MLB players in history. A 100-grade outcome would be a 1:1.75M*.

*hence, all those PT grades are utter nonsense.

A benefit of this scale is it moves. A 50-grade player is always average, even if that means we're comparing 2025 Lawrence Butler to 1993 Doug Strange. On a 1-100 scale, if Strange was a 70, then Butler would be like a 170. i.e. if we pitted the two players against each other, Butler would run laps around Strange, and yet they were both league-average 50-grade players (2 WAR).

Of course, because we've anchored average/50 to MLB, that means most of the population is a 0 or a 10. We say the worst professional players are 20s because they're still comfortably among the top 250,000 people to ever play the sport even if they have no hope of reaching the Majors.

Again, to emphasize, the overall 20-80 is spurious. It's just a fun compilation. The 20-80 scale is really for grading traits.

[Pelican]

Originally Posted by BaseballATeam

The 20-80 system is more intuitive when thinking about player traits and how they fit in the context of roster building. It reflects a normal curve, so, if you're Bayesian-trained, you should prefer the 20-80 method.

Simplistically, a 50-grade trait is MLB average. 60 is one standard deviation better than average. 70 is two SDs, 80 is three SDs. A 90-grade is technically plausible but so rare as to not be assigned, it's a 1-in-31,500 outcome for an MLB player. According to gemini, there have been 22,664 MLB players in history. A 100-grade outcome would be a 1:1.75M*.

*hence, all those PT grades are utter nonsense.

A benefit of this scale is it moves. A 50-grade player is always average, even if that means we're comparing 2025 Lawrence Butler to 1993 Doug Strange. On a 1-100 scale, if Strange was a 70, then Butler would be like a 170. i.e. if we pitted the two players against each other, Butler would run laps around Strange, and yet they were both league-average 50-grade players (2 WAR).

Of course, because we've anchored average/50 to MLB, that means most of the population is a 0 or a 10. We say the worst professional players are 20s because they're still comfortably among the top 250,000 people to ever play the sport even if they have no hope of reaching the Majors.

Again, to emphasize, the overall 20-80 is spurious. It's just a fun compilation. The 20-80 scale is really for grading traits.

Duly noted. But can't the same thing be done on a scale of 100 data points, instead of only 60? Fifty would still be the median, if not the mean. 0 and 100 would be rare events. Contrived, but more granular. Not reflective of reality, though, in terms of actual talent.

To me the idea of a "normal curve" in baseball scouting is a fallacy. I accept the theory that we all occupy places on a giant talent bell curve for the entire population. That distribution makes sense. It comports with reality. But baseball players as athletes are all taken from the high end of that curve. If the scouts are doing their jobs, there won't be any "normal" distribution. Everybody is at least good, some will be very good or great. There's no bell curve. Rather, there's a whole lot of guys who will never make it. And a tiny number who will, somehow, someday. The curve will have a pronounced negative slope, with all those prospects clustered at the front end, down to a relative handful of MLB players.

Look at it this way. Everybody from the low minors on up is at least 99th percentile of the population for baseball skills. Out of that top 1%, there won't be any bulge of mediocre players in the middle. There will be lots and lots of prospects at the low end, 10-20. A few will progress to 40 or 50. And a few of them will progress to 60 or 70 or 80 and an MLB career. Yes, those ratings are objective, not comparative.

A rating system that posits 50 as an average (mean or median) needs just as many rating above and below 50. That won't happen on any objective basis. The vast majority will be below average (median or mean) in raw talent. Only a handful will be above 50. Why not, instead, rate talent on an objective, absolute scale, 1-100, fully anticipating the low end will be crowded?

[This is a variation of the "grade inflation" fallacy. In some schools, if you have five students, all brilliant, you as professor still must rank them, with one in the middle, two above "average", and two below. Which is pure nonsense, and does not reflect their skill sets. It's arbitrary. You can't force a "normal distribution" on a group selected for above-average skills.]

To return (finally) to the OP, that's why I have no problem with all the low ratings for draftees. It totally mirrors IRL. This isn't an arcade game. Most of these dudes have very limited skills, relative to MLB. Even their potential is limited. But remember, those initial ratings are not frozen. Guys can improve or regress over time. Development labs! There will be needles in the haystack. But the haystack is not half needles. And a rating system that relies on that false assumption is worthless. It misleads.

Of course, the beauty of OOTP is the ability to play it your way. To each his own. I'm not trying to convert anybody. I'm just defending why my approach makes sense to me. Reading well-thought-out posts that rely on the 20-80 scale is super frustrating. It becomes an apples and oranges situation. We aren't speaking the same language. But we are playing the same game!

[Reed]

This is from MLB.com
Scouting grades have been a staple of MLB.com's prospect coverage for years, and they generally match how clubs grade players as well.

Players are graded on a 20-80 scale: 20-30 is well below average, 40 is below average, 50 is average, 60 is above average and 70-80 is well above average. When discussing prospects, the most important number is the future overall grade, an all-encompassing number on the 20-80 scale that signifies what each player is projected to ultimately be in the big leagues.

A future overall grade of 65 or better is for a player who could develop into a future impact Major Leaguer, perhaps an All-Star-caliber standout. (Note: Some clubs use a 2-8 scale -- as opposed to 20-80 -- which is basically the same thing but without half-grades.)

[BaseballATeam]

Pelican, what you described is a skewed normal curve anchored to MLB average--which is what I was describing. I think maybe the language I used isn't clicking for you, but we're saying the same thing, more or less.

The 1-100 system can be used such that 50 (or whatever number) is league average rather than a representation of a specific amount of talent (i.e. the way most video games use a 1-100 score). In which case, it's not different from the 20-80 scale. We just lose that sense of Bayesian proximity to the state of "average." How does 50 compare to 70? We don't know, not without a lot of testing.

fwiw, I'm pretty sure that's how OOTP uses the 1-100 system. I've never tried it out, but I can't imagine their all-era's random mode has Babe Ruth as a 55/100 next to Judge's 96/100.

irl, 1-point differences in grade offer false specificity. Some scouts don't use 5's (especially for tools) for that very reason. They view it as unprofessional (they're sorta the purists, and there's fewer of them than in the past).

For video game purposes, I have no problem with people preferring a 1-100 system. I was merely trying to supply some background since you mentioned not understanding the 20-80 system. For a statistically trained mind, it's far more intuitive than a 1-100 scale (not that either is challenging).

[Pelican]

Originally Posted by BaseballATeam

Pelican, what you described is a skewed normal curve anchored to MLB average--which is what I was describing. I think maybe the language I used isn't clicking for you, but we're saying the same thing, more or less.

The 1-100 system can be used such that 50 (or whatever number) is league average rather than a representation of a specific amount of talent (i.e. the way most video games use a 1-100 score). In which case, it's not different from the 20-80 scale. We just lose that sense of Bayesian proximity to the state of "average." How does 50 compare to 70? We don't know, not without a lot of testing.

fwiw, I'm pretty sure that's how OOTP uses the 1-100 system. I've never tried it out, but I can't imagine their all-era's random mode has Babe Ruth as a 55/100 next to Judge's 96/100.

irl, 1-point differences in grade offer false specificity. Some scouts don't use 5's (especially for tools) for that very reason. They view it as unprofessional (they're sorta the purists, and there's fewer of them than in the past).

For video game purposes, I have no problem with people preferring a 1-100 system. I was merely trying to supply some background since you mentioned not understanding the 20-80 system. For a statistically trained mind, it's far more intuitive than a 1-100 scale (not that either is challenging).

Thanks for making the effort to explain. I'm more and more convinced that my high score on the SAT Maths many years ago was a clerical error. Or maybe I'm just dense. It's encouraging to hear we may be talking about the same thing, even if I can't quite grasp how it works.

One confusing aspect for me is that OOTP also uses, at various points, a scale of 1-600 (batter contact, power, walks, and pitcher stuff, movement, control) and 1-250 (defensive position ratings, except experience is 0-200; plus speed, stealing base running, bunting), internally. Are those absolute numbers, or Bayesian? When I fiddle around in the Editor, and then run a rescout, I am amused by the modest changes in the player ratings.

No question that the one-point variations do create "false specificity" - precision suggested, without basis. Even with highly accurate scouting (my normal choice - not a "fog of war" guy), single-digit differences are not significant. Except at 100% scouting accuracy (not there yet - probably never) 1/100 is not much.

Not only am I stubborn, but one thing I have found is that switching between sims is confusing, if my underlying rules are too varied. So for purposes of "standardization" I stick with the 1-100 ratings and high TCR and accelerated development (no autocalc) and aging, the accurate scouting reports, and who knows how many other settings - all based on my take on the collected wisdom of these boards when I started with OOTP 21. I guess it's too late to change now (old dog; new tricks)!