Anyone have a general idea how far off amateur draft scouting results could be at various budget levels? I've got a higher college than HS budget right now. There are a couple HS players who look dominant, but I'm more confident in the college ratings obviously. Was wondering if there was some consensus that if a pitcher has projected 100vR with a budget of 5, 10, or 15, that could really be a "true" projected vR of 65, 75, or 85 because of the error bands (if that makes any sense).

Biggest giveaway I've seen with pitchers are splits(37/88) and pitch 2 or 3 being higher than P1. Another is the 40 DUR/88 STM. There just aren't many pitchers built with 50 diff in splits, higher secondary pitches(although P2 happens often enough) or a DUR/STM combo that would allow them to throw 330 innings.

Note, this was before that update where they said they were making college scouting more accurate (or HS less accurate?, I don't remember which).

$20M will have a range between -10 and +10 (except you need to take an additional 3 or 4 points off of coach-able ratings, the ones that improve over the course of the season, because no player hits their ceiling in those ratings in my observations). This is based on the tracking of 10 players I did a few years back. Can't speak to the other budget levels because I didn't track those levels.

There's no magic formula to compare budget level to variance. Even 16M can have obvious scouting outliers, like 100 ratings in one category on an otherwise mediocre player, and even 4-6M can be useful once you account for those outliers and understand the stuff MikeT is talking about in his post. One thing to keep in mind is that hitting/pitching/fielding ratings are almost always overestimates wherever there is variance, but general ratings (dur/health/sta/pat/makeup) often vary in either direction.

One thing to keep in mind is that hitting/pitching/fielding ratings are almost always overestimates wherever there is variance

This hasn't been my experience, I've seen it both ways fairly consistently. I think what makes it seem this way is like I mentioned above, players almost always miss their ceilings by 2-4 points (or more, if you don't develop them properly). I know this because there used to be a bug that let you see all players' hidden ceiling ratings. Every player (across teams and across leagues) missed their ceiling by a few points. Either we're all doing something wrong during player development or there's something in the code that doesn't make it possible.

Seems like we don't really know. Oh well. I'm sure the HS pitcher I'm seeing with 95 vL and 100 vR isn't really that good. If I had 20M in there I'd be confident he's still great though. If I had 5M in HS scouting, and that could really be 65vL and 70vR, he's still solid. But if that could be as bad as 45vL and 50vR, then all bets are off.

Sounds like the consensus is that if the splits are 37/88, the 37 is probably pretty accurate and it's the 88 that is off. That is useful anyways since I agree - biggest split I've seen for a developed pitcher is somewhere around 20 points.

It's usually somewhere in the middle but on the lower end. 62.5 per the numbers but more likely 57.

@gman981981: The alternative is that if you had 20M in scouting the projected splits might not be 95/100.

@brianplath, I do agree that projections are based on optimal training, makeup, coaching and development that doesn't really exist, so guys struggle to make projections in most cases, but I have yet to see a single non-DITR that vastly outperformed my $8M HS/college projection. By vastly, let's say, projections in the 50s turning into ratings in the 70s at the end of the development cycle. Every draft I can find at least a half-dozen that are overestimated that much. As I said in the previous post, that doesn't apply to general ratings, just hitting/pitching/fielding, since especially STA/DUR for pitchers I've seen be way off in both directions.