Scouts seem utterly worthless? - Page 3

Elendil · 03-25-2007, 03:43 PM

Quote:

Originally Posted by Erithtotl

These are reports that were just submitted. You are correct, it does seem that the average report is more accurate. But it seems it is only more accurate because my two best scouts consistently overesitmate, while my bad scouts are much closer. This should be the opposite!

Maybe the scouting reports on those scouts are wrong.

Erithtotl · 03-25-2007, 03:46 PM

Quote:

Originally Posted by Raidergoo

Scouting error is working as designed. This was looked at. Here's some data.

Code:

Error avg		0.129190882	13% is the error average for SISA		
Max error		29			
					
				    Years Pro	   Error average/rating	       Error %	      Players
For 0 Pro years players: 		0		0.179071141		18%		138  
For 1 Pro years players:		1		0.144380262		15%		138
For 2 Pro years players:		2		0.159432522		16%		160
For 3 Pro years players:		3		0.113848504		11%		185
For 4 Pro years players:		4		0.137739045		13%		168
For 5 Pro years players:		5		0.114065694		11%		169
For 6 Pro years players:		6		0.140923622		11%		63
For 7 Pro years players:		7		0.116287087		11%		47
For 8 Pro years players:		8		0.099825567	 	9%		35
For 9 Pro years players:		9		0.054448421	 	5%		25
For 10 Pro years players:		10		0.042398696	 	4%		12
For 10+ Pro years players:		10+		0.029110581	 	3%		59

Code:

		SISA		100 Scout	75 Scout	50 Scout
Error Avg	3.867132867	1.251748252	2.839160839	5.958041958
Error Max	27		6		21		29

Here are some operating theories behind scouting. The older, more established players will have more accurate evaluations. The higher rated scouts will be more accurate. Never trust a single scouting report; cross check everything and everyone, then do it again. Scouts will rarely underestimate talent, but will overestimate talent.

Interesting chart, but I'm not sure I follow it. So with a scout with a 94 scouting hitting potential, how far off should he be on average and max, on a 0-year prospect? Assuming a 50 for any given rating, what's the high and low? Because right now I have two scouts, one a 94 one a 95, who are averaging a %50 error across 5 different players (as detailed above).

KCRoyalsFan · 03-25-2007, 03:47 PM

Quote:

Originally Posted by Erithtotl

These are reports that were just submitted. You are correct, it does seem that the average report is more accurate. But it seems it is only more accurate because my two best scouts consistently overesitmate, while my bad scouts are much closer. This should be the opposite!

I have noticed this as well. On my star prospect my head scout overestimated by about 5 points on the 1-20 scale. All the other scouts were really close. Thank goodness the average report helps you realize when one of your scouts may be on crack.

Maybe with more testing we can see if it is consistent with the fact that the head scout is wrong a lot of the time. If that is the case, it could probably be fixed in a patch. But, you can't really say its a bug based on 4 players. We must do more simulations

KCRoyalsFan · 03-25-2007, 03:49 PM

Quote:

Originally Posted by Raidergoo

Scouting error is working as designed. This was looked at. Here's some data.

Code:

Error avg		0.129190882	13% is the error average for SISA		
Max error		29			
					
				    Years Pro	   Error average/rating	       Error %	      Players
For 0 Pro years players: 		0		0.179071141		18%		138  
For 1 Pro years players:		1		0.144380262		15%		138
For 2 Pro years players:		2		0.159432522		16%		160
For 3 Pro years players:		3		0.113848504		11%		185
For 4 Pro years players:		4		0.137739045		13%		168
For 5 Pro years players:		5		0.114065694		11%		169
For 6 Pro years players:		6		0.140923622		11%		63
For 7 Pro years players:		7		0.116287087		11%		47
For 8 Pro years players:		8		0.099825567	 	9%		35
For 9 Pro years players:		9		0.054448421	 	5%		25
For 10 Pro years players:		10		0.042398696	 	4%		12
For 10+ Pro years players:		10+		0.029110581	 	3%		59

Code:

		SISA		100 Scout	75 Scout	50 Scout
Error Avg	3.867132867	1.251748252	2.839160839	5.958041958
Error Max	27		6		21		29

Here are some operating theories behind scouting. The older, more established players will have more accurate evaluations. The higher rated scouts will be more accurate. Never trust a single scouting report; cross check everything and everyone, then do it again. Scouts will rarely underestimate talent, but will overestimate talent.

Very impressive data. So is this all based on the 1-100 scale? When you say a 100 scout you mean a 200-200 scout and a 75 would be a 150-200 scout?

Erithtotl · 03-25-2007, 03:52 PM

Quote:

Originally Posted by KCRoyalsFan

I have noticed this as well. On my star prospect my head scout overestimated by about 5 points on the 1-20 scale. All the other scouts were really close. Thank goodness the average report helps you realize when one of your scouts may be on crack.

Maybe with more testing we can see if it is consistent with the fact that the head scout is wrong a lot of the time. If that is the case, it could probably be fixed in a patch. But, you can't really say its a bug based on 4 players. We must do more simulations

Yeah and the real problem I'm having is that my best two scouts seem to agree consistently across players, while my 4 bad scouts are actually the ones who are correct. If it were working as designed, then the chances of the top two scouts both being wrong in exactly the same way and the bottom 4 scouts being right would be almost impossible. Odds are that one of the top two scouts would be in line with the correct numbers when the other was off.

KCRoyalsFan · 03-25-2007, 03:56 PM

Quote:

Originally Posted by Erithtotl

Yeah and the real problem I'm having is that my best two scouts seem to agree consistently across players, while my 4 bad scouts are actually the ones who are correct. If it were working as designed, then the chances of the top two scouts both being wrong in exactly the same way and the bottom 4 scouts being right would be almost impossible. Odds are that one of the top two scouts would be in line with the correct numbers when the other was off.

lol, you got me so interested in the rating/scouting system it is already 3pm and I haven't hardly played the game today

Time to take a break from this stuff for me. At least I now know to scout top players with all my scouts before coming to a decision.

One question I have for you guys is what do you do with your scouts during the season? I usually have my top scout scout the league, should I just set all my scouts to scout the league? Sometimes I'll have at least one of my scouts scout my team so that I know when players are developing.

ravinhood · 03-25-2007, 04:23 PM

Hey how do you scout the "amateur draft pool" before the draft begins a month later???

KCRoyalsFan · 03-25-2007, 04:31 PM

you have go to the the league transactions screen and it is a tab towards the top left.

ravinhood · 03-25-2007, 04:39 PM

thanks

Sven Draconian · 03-25-2007, 04:54 PM

Why in the world is it designed to not underrate players?

Tim Hudson springs immediately to mind, Scouts considered him to small. What about the entire book "Moneyball" where Michael Lewis talks about how often scouts dismiss a product based on apperance (In OOTP terms, underrates them).

What about players like Marcus Thames and Craig Monroe from the Tigers, who were MLFA and Waiver Wire pickups, I would say that qualifies as underrated by someone.

Id agreed that Scouts tend to be overly optimistic, but underrating players should not be "rare".

Erithtotl · 03-25-2007, 04:55 PM

Quote:

Originally Posted by Sven Draconian

Why in the world is it designed to not underrate players?

Tim Hudson springs immediately to mind, Scouts considered him to small. What about the entire book "Moneyball" where Michael Lewis talks about how often scouts dismiss a product based on apperance (In OOTP terms, underrates them).

What about players like Marcus Thames and Craig Monroe from the Tigers, who were MLFA and Waiver Wire pickups, I would say that qualifies as underrated by someone.

Id agreed that Scouts tend to be overly optimistic, but underrating players should not be "rare".

Yeah I agree with this as well, though I guess the question is how big of a split is there between overrate and underrate?

ravinhood · 03-25-2007, 05:00 PM

Quote:

Originally Posted by Sven Draconian

Why in the world is it designed to not underrate players?

Tim Hudson springs immediately to mind, Scouts considered him to small. What about the entire book "Moneyball" where Michael Lewis talks about how often scouts dismiss a product based on apperance (In OOTP terms, underrates them).

What about players like Marcus Thames and Craig Monroe from the Tigers, who were MLFA and Waiver Wire pickups, I would say that qualifies as underrated by someone.

Id agreed that Scouts tend to be overly optimistic, but underrating players should not be "rare".

I agree as well with this. This brings to mind that player I was talking about that had all 60's in three major batting catagories who never batted above low 200's, horrible against rightys and just never performed up to what the scouts rated him. I love this about the game. It's not something that happens often, but, "SHOULD" happen I think on both sides of the fense, over-rated and under-rated.

thbroman · 03-25-2007, 05:15 PM

I can think of one possible reason why the bias is toward overrating prospects, and that is if the bias was neutral or toward underrating it would be too easy to build your team. In such a case you would always just go for the best guys your scouts can identify and sometimes you would get a REALLY nice surprise. With a bias toward overrating, you are more likely to invest valuable draft picks on guys who turn out to be a bust.

KCRoyalsFan · 03-25-2007, 05:16 PM

I think the game's way of simulating Tim Hudson is just having their talents sky rocket. The game doesn't want you to be able to see who is underrated, they want it to be a surprise.

thbroman · 03-25-2007, 05:18 PM

Quote:

Originally Posted by KCRoyalsFan

I think the game's way of simulating Tim Hudson is just having their talents sky rocket. The game doesn't want you to be able to see who is underrated, they want it to be a surprise.

Precisely.

Killebrew · 03-25-2007, 05:27 PM

Quote:

Originally Posted by Raidergoo

....
Scouts will rarely underestimate talent, but will overestimate talent.

Ouch.

Sven Draconian · 03-25-2007, 05:43 PM

Quote:

Originally Posted by KCRoyalsFan

I think the game's way of simulating Tim Hudson is just having their talents sky rocket. The game doesn't want you to be able to see who is underrated, they want it to be a surprise.

Too me that sounds more like Mike Piazza than Tim Hudson. Hudson put up numbers everywhere, had the stuff, but was just never highly regarded. He didn't come from nowhere like a Piazza, he was just underrated.

Same with a guy like Albert Pujols. Only one scout liked him (St. Louis) the rest underrated him.

That is different from a guy Piazza who just appeared out of nowhere. Those players had huge talent jumps after being drafted. Guys that are underrated are just that.

I think the split should be a heavy lean of 85% in the favor of overrated, with a solid 15% underrated. Maybe even a bit more, like 88-12 or 90-10, but it should be there.

Elendil · 03-25-2007, 06:13 PM

The trouble is that Tim Hudson is an outlier: he became a successful major leaguer. For every Tim Hudson, there are hundreds of Jeff Judens who are slavered over by scouts but never make it.

Mets Man · 03-25-2007, 07:41 PM

Quote:

Originally Posted by Sven Draconian

Why in the world is it designed to not underrate players?

Tim Hudson springs immediately to mind, Scouts considered him to small. What about the entire book "Moneyball" where Michael Lewis talks about how often scouts dismiss a product based on apperance (In OOTP terms, underrates them).

What about players like Marcus Thames and Craig Monroe from the Tigers, who were MLFA and Waiver Wire pickups, I would say that qualifies as underrated by someone.

Id agreed that Scouts tend to be overly optimistic, but underrating players should not be "rare".

I too agree that the game should allow scouts to underrate talent more frequently that how it is in the game right now. I think people are right when they say that scouts tend to overrate talent more, but I don't think underrating talent should be a "rare" thing. I think we've all heard of alot of real life examples in all sports where a player who was vastly underrated becomes a superstar. For football, Tom Brady comes to mind. I don't think you can entirely attribute that to a suddent "talent explosion". I will agree that he obviously got better and probably developed better skills, but at the same time, I believe that he was always very good and had plenty of talent, but the scouts just got it wrong.

Mets Man · 03-25-2007, 07:45 PM

Furthermore, I think having scouts underrate talent adds an extra element to gameplay.

Having to wonder whether that player you were thinking of picking up might possibly be better than the scouts think he is. You check his stats and realize that he's putting up decent numbers. Maybe you should pick him up, because the scout may have just underrated him. It's the whole idea of picking up that diamond in the rough.

03-25-2007, 05:15 PM	#53
thbroman All Star Reserve Join Date: Jun 2002 Location: Madison, WI Posts: 867	I can think of one possible reason why the bias is toward overrating prospects, and that is if the bias was neutral or toward underrating it would be too easy to build your team. In such a case you would always just go for the best guys your scouts can identify and sometimes you would get a REALLY nice surprise. With a bias toward overrating, you are more likely to invest valuable draft picks on guys who turn out to be a bust. Last edited by thbroman; 03-25-2007 at 05:19 PM.

03-25-2007, 06:13 PM	#58
Elendil Hall Of Famer Join Date: Dec 2003 Location: the dynasty forum Posts: 2,318	The trouble is that Tim Hudson is an outlier: he became a successful major leaguer. For every Tim Hudson, there are hundreds of Jeff Judens who are slavered over by scouts but never make it. __________________ Heaven is kicking back with a double Talisker and a churchwarden stuffed with latakia.

03-25-2007, 04:23 PM	#47
ravinhood Major Leagues Join Date: Apr 2004 Posts: 374 Infractions: 0/1 (1)	Hey how do you scout the "amateur draft pool" before the draft begins a month later???

03-25-2007, 04:31 PM	#48
KCRoyalsFan All Star Reserve Join Date: Mar 2003 Posts: 599	you have go to the the league transactions screen and it is a tab towards the top left.

03-25-2007, 04:39 PM	#49
ravinhood Major Leagues Join Date: Apr 2004 Posts: 374 Infractions: 0/1 (1)	thanks

03-25-2007, 04:54 PM	#50
Sven Draconian Major Leagues Join Date: Jul 2005 Posts: 432	Why in the world is it designed to not underrate players? Tim Hudson springs immediately to mind, Scouts considered him to small. What about the entire book "Moneyball" where Michael Lewis talks about how often scouts dismiss a product based on apperance (In OOTP terms, underrates them). What about players like Marcus Thames and Craig Monroe from the Tigers, who were MLFA and Waiver Wire pickups, I would say that qualifies as underrated by someone. Id agreed that Scouts tend to be overly optimistic, but underrating players should not be "rare".

03-25-2007, 05:16 PM	#54
KCRoyalsFan All Star Reserve Join Date: Mar 2003 Posts: 599	I think the game's way of simulating Tim Hudson is just having their talents sky rocket. The game doesn't want you to be able to see who is underrated, they want it to be a surprise.

03-25-2007, 07:45 PM	#60
Mets Man Major Leagues Join Date: Jan 2002 Posts: 336	Furthermore, I think having scouts underrate talent adds an extra element to gameplay. Having to wonder whether that player you were thinking of picking up might possibly be better than the scouts think he is. You check his stats and realize that he's putting up decent numbers. Maybe you should pick him up, because the scout may have just underrated him. It's the whole idea of picking up that diamond in the rough.