|
||||
|
01-09-2020, 10:42 PM | #41 |
All Star Reserve
Join Date: Apr 2010
Posts: 693
|
From what I can tell, there seems to be a semantic breakdown between how some are viewing what's happening
Some seem to be bothered that OSA's overall/potential ratings don't seem to change based on scouting accuracy settings Some seem to think that this possible issue potentially equates to 100% accuracy for OSA IF OSA's overall/potentials never change despite putting scouting accuracy at low or high or anywhere between, that doesn't necessarily mean that those OSA ratings are in fact accurate (even before the development engine is even brought into it) However, I believe Dyzalot's point is that when there's a change in scouting accuracy, that the OSA overall/potential ratings should be expected to change somewhat also (assuming they aren't) |
01-10-2020, 12:51 AM | #42 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
Not sure how much of the thread you have read but my research seems to indicate that on the overall potential rating, the OSA scout is always spot on with the result you get from 100% scouting accuracy. My research did show that all of the individual ratings change. But think about this. If you are looking at your scout on a prospect and he's rated as three stars overall but the OSA scout has him at four stars yet when you look at the individual ratings your scout has him rated as a higher potential than OSA at every rating, then it would seem as if the overall potential rating as scouted by OSA removes the "fog of war". That doesn't mean you will have accuracy to know if a guy is rated highly because he has high contact, power or other potentials, but just knowing that the true overall potential is always showing the "100% accuracy" value means you should just scout by that one metric and almost ignore everything else.
EDIT: I meant this to be in reply to Hrycaj Last edited by Dyzalot; 01-10-2020 at 12:53 AM. |
01-10-2020, 12:56 AM | #43 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
But can't one infer that from the fact that OSA seems to consistently match what a team scout shows when accuracy is set to 100%? Like even at the "very low" setting for scouting accuracy, it appears, at least in the smallish sample size I used, that at every level of scouting accuracy, OSA shows whatever will be shown when set to 100% accuracy. That seems to indicate to me, unless 100% accuracy isn't actually 100% accurate, that the overall potential rating as scouted by OSA is always spot on.
|
01-10-2020, 06:33 AM | #44 |
All Star Starter
Join Date: Mar 2007
Location: Nashville Area
Posts: 1,257
|
Also remember if it's on relative ratings it compares to the players in the league not true ratings so the overall rating being discussed isnt the actual overall rating its how the scout compares to the rest of the league. If this is going to be looked at do you have the data in how the non relative ratings look?
|
01-10-2020, 09:42 AM | #45 |
Hall Of Famer
Join Date: Sep 2008
Location: with my army of orangutans
Posts: 2,943
|
Could be that maybe something is happening weirdly in-game with real-life prospect draftees because they're not created by the game. Since that seems to be mostly what people have tested in here, I'll check in a fictional league when I get home from work.
|
01-10-2020, 09:57 AM | #46 | |
Hall Of Famer
Join Date: Mar 2018
Location: Denver, Colorado
Posts: 4,262
|
Quote:
I only play fictional these days and anecdotally it feels like scouting is working as advertised. In other words, my sense is that my very fine head scout does a better job of projecting future big leaguers than the OSA does. But this is admittedly purely an anecdotal impression, based upon no rigorous study.
__________________
The Denver Brewers of the W.P. Kinsella League-- The fun starts here(1965-1971: https://forums.ootpdevelopments.com/...d.php?t=289570 And continues here (1972-1976): https://forums.ootpdevelopments.com/...d.php?t=300500 On we go (1977- 1979): https://forums.ootpdevelopments.com/...d.php?t=314601 For ongoing and more random updates on the WPK:https://forums.ootpdevelopments.com/...d.php?t=325147, https://forums.ootpdevelopments.com/...d.php?t=330717 |
|
01-10-2020, 10:07 AM | #47 |
All Star Starter
Join Date: Oct 2010
Posts: 1,361
|
What I did was with a completely fictional league.
|
01-10-2020, 10:13 AM | #48 |
Hall Of Famer
Join Date: Sep 2008
Location: with my army of orangutans
Posts: 2,943
|
Yeah yours was the only example I saw that lined up with what I'm going for and I'm curious to expand upon that to see if the issue is potentially real-life prospects being scouted weirdly.
|
01-10-2020, 12:48 PM | #49 |
Bat Boy
Join Date: Apr 2019
Posts: 16
|
I've always been aware of this as well. OSA potential rating is always 100% accurate to the players true potential. The OSA doesnt do a good job of always quantifying the exact skills like contact, power, etc. but they will nail down the exact potential rating. I'm sure the OOTP team could fix this up. I always draft really well just by sorting out the draft pool by OSA potential, which has really hurt some of the immersion for me.
|
01-10-2020, 03:20 PM | #50 | |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
Quote:
|
|
01-10-2020, 04:02 PM | #51 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
I'm going to do a test with fictional as well. Starting a new fictional save right now based upon the standard 2019 MLB setup. Will sim once again to the date of the release of the draft class and then I'll take a look at some of the top prospects listed in that email.
|
01-10-2020, 04:17 PM | #52 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
Ok here's my first example. I set up a fictional league based upon the 2019 MLB format. Populated the teams with random fictional players and simmed to the release of the first draft class. I also edited the team's scout to have max ratings for all scouting so as to have as little variation as possible in the team scout to compare that to whatever variation there is in the OSA scout. Once again you will see three pictures for each player. First one is with scouting accuracy set to "very low", the second one will be on "normal" and the last one is shown with "100% accuracy". Here is the first one. As you can see from the first example, it appears this is affecting fictional as well. I'll do a few more though to see if we get the same pattern as we saw when I did it with real players.
|
01-10-2020, 04:26 PM | #53 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
Here is the next one. On this one the OSA overall matches the scout's overall at every scouting level. Note though that the current overall is more accurate with OSA than my scout. Also let's remember that I now am doing this with the best scout possible. So logically it would seem to me that a team scout with maxed ratings should on average scout more accurately than OSA scouting. Therefore we should see some examples where my scout is more accurate than OSA on overall potentials at lower scouting accuracy settings. I'll do a few more and see if we see any that work that way.
|
01-10-2020, 04:30 PM | #54 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
Next one...
|
01-10-2020, 04:32 PM | #55 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
OK someone can do more of these if they want a larger sample size but I've seen enough. Across five "real life" prospects and three fictional prospects, OSA has never been incorrect on what 100% accuracy scouting would show for the overall no matter what scouting accuracy is set. I have definitely seen enough myself to hope that the OOTP devs take a look at this to see if it is an actual problem.
|
01-10-2020, 05:02 PM | #56 |
OOTP Developer
Join Date: Jun 2009
Location: Here and there
Posts: 14,127
|
Thanks for everyone's work in looking into this. There's a lot of moving pieces to scouting/OSA and the ratings, it definitely appears that the OSA ratings are shifting as much relative to the accuracy settings as most normal scouts shift.
|
01-10-2020, 05:17 PM | #57 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
Are you ignoring the evidence on the overalls? OSA isn't shifting even 1%. And not to be a jerk about this but I put a lot of time into this for you to just come back like some DOJ rep telling me Epstein really did kill himself. I mean the evidence is plain as day and you make a post like that? I'm feeling pretty disrespected.
Last edited by Dyzalot; 01-10-2020 at 05:23 PM. |
01-10-2020, 07:10 PM | #58 | |
OOTP Developer
Join Date: Jun 2009
Location: Here and there
Posts: 14,127
|
Quote:
|
|
01-10-2020, 08:00 PM | #59 |
All Star Starter
Join Date: May 2008
Posts: 1,735
|
OK. I apologize if I came off a bit caustic in that last post but your previous post seemed like you had already made the determination that everything was "kosher". Thanks for looking further into this.
|
01-10-2020, 09:50 PM | #60 |
Minors (Double A)
Join Date: Feb 2019
Posts: 100
Infractions: 0/1 (1)
|
I guess a lot of us didn't notice this given we're trained to ignore the OSA and always go with our scout....after testing I too can see OSA never changing from Very Low to Normal to 100%, but my scout suddenly 100% agrees with the OSA on 100% scouting. This is happening with every player on my team.
Edit: I understand the ratings under the hood are changing, I'm speaking along the lines of overall and potential grades for the player as a whole. That's where I'm seeing the problem....the whole point of the overall and potential ratings is to give you an big-picture viewpoint of the player, (regardless of how everyone here likes to play the game, please spare me a lecture). I would expect, for instance, OSA's rating of Jason Foley (Closer,Tigers) to not stay at 22/58 at all times while his basic pitching ratings are changing as I keep switching the scouting accuracy. Makes no sense. My team scout's overall and potential ratings are definitely bending as I make accuracy changes, why isn't OSA's? JASON FOLEY - TIGERS (using him as a random example) VERY LOW ACCURACY - OSA Stuff 50/65 Movement 40/50 Control 30/30 Overall/Potential - 22/58 VERY LOW ACCURACY - TEAM SCOUT Stuff 50/65 Movement 35/45 Control 35/35 Overall/Potential - 20/36 NORMAL ACCURACY - OSA Stuff 50/65 Movement 40/50 Control 30/35 Overall/Potential - 22/58 NORMAL ACCURACY - TEAM SCOUT Stuff 50/65 Movement 40/45 Control 35/35 Overall/Potential - 20/36 100% ACCURACY - OSA Stuff 50/70 Movement 40/50 Control 35/45 Overall/Potential - 22/58 100% ACCURACY - TEAM SCOUT Stuff 50/70 Movement 40/50 Control 35/45 Overall/Potential - 22/58 How in the world could this player gain 5 potential to Stuff and 15 potential to Control, yet stay EXACTLY at 22/58 (as a closer)? Maybe with stars you could argue you wouldn't notice a change because it's such a bad rating system with only 10 outcomes, but a perfect 20-80 scale, you're telling me that didn't change his 58 potential one bit from low accuracy to 100%? Seems odd to me, it feels like the overall/potential ratings are stuck at 100% given the team scout agrees at that setting. If I'm being redundant, I apologize, but I'm worried some people would only look at the stuff/movement/control settings, not seeing the problem being presented. I know some people here like to do "stats only" or maybe only care about the "stuff" rating for relievers, but I'm someone who loves to rely on the scouting to give me a rough estimate of what a player is/could be big-picture....this possible bug means I really had better hope I don't accidentally click on OSA and see the true ratings of my players. That would KILL the immersion for me. Anyway, back to my hole....thank you Dyzalot for your efforts in this thread. Wanted to back you up. Last edited by ThePride87; 01-10-2020 at 10:41 PM. |
Bookmarks |
|
|