Suggestion for Scouting

[Mike Donlin]

I think of this as the Crystal Ball Problem. And my suggestion for dealing with it is below.

I have a 30 year old center fielder who I signed to a reasonable contract extension during his final arbitration year. He has never been much more than an average hitter, but he plays very good defense. At the start of spring training, my scout reported that his power rating had dropped. (For the sake of describing this problem, I opened up the editor and checked. His rating had fallen.)

As it happens, though, he is in the midst of a career year, and of course I now (at the trade deadline) want to trade him and sell high while he still has some value.

But this strikes me as deeply unrealistic. Mainly, that a scout would insist, against evidence, that a player is washed up (or on the way to being washed up) in the midst of a career year. No scout would say that Player X is on the decline when he has just had the four best months of his career. What evidence would he point to? A hole in his swing? Slower hands? The same slow hands he's using to hit home runs at an unprecedented rate?

So here is my suggestion. Scouting for players say, 26 and older, should follow that player's on the field performance much more closely. It is one thing for a scout to say a 19-year old prospect is going to blossom even though he is struggling now, or that a different 19-year old prospect is nothing special even though he is tearing up the minors. But a scout would never say that a 30-year old center fielder who is tearing up the majors is soon going to crash to earth. He might say he can't keep it up, and will return to his normal average self. But he would never say he's going to be worse than he was.

You should have to discover his declining ability for yourself, after he fails, when your scout could then tell you, yeah, he's washed up. But a scout should not be able to see into the future, especially when the future goes so strikingly against the present.

[olivertheorem]

Two words: Jacoby Ellsbury.

He didn't magically become a great power hitting in 2011 (32 HR) and turn into a pumpkin the next year. I would bet money that Boston's scouts were telling the front office the whole time that the power spike was a fluke and not likely to stick. Career years by definition are at least a little fluky and not indicative of a player's actual ability.

That said, you do have a fair point that, at times, ratings decreases are seen prior to performance decline, but I'd guess that, at least some of the time, a team's scouts see a likely performance decline coming prior to it actually happening, though that is more likely to happen in the offseason I'm guessing.

[Mike Donlin]

Originally Posted by olivertheorem

That said, you do have a fair point that, at times, ratings decreases are seen prior to performance decline, but I'd guess that, at least some of the time, a team's scouts see a likely performance decline coming prior to it actually happening, though that is more likely to happen in the offseason I'm guessing.

Really? I can believe that scouts can look at a guy raking in the majors and conclude that he will not keep the raking up. But I have trouble believing that a scout can look at a guy raking in the majors and conclude that when he is done raking he will be worse than when he started. Are scouts really that good?

[nyy26wc]

You said the scout told you at the start of spring training that the player's power had dropped. He didn't tell you during the season, while he was putting up career year numbers.

Subsequent to that, the player has gone on to have a career year.

If he told you at the start of spring training that the player was through, that's very different from him telling you that during the player's career year.

I'm sure this is a frequent event in baseball. A scout makes an assessment of a player and the player's performance proves he was wrong. My favorite example of that was what Bobby Murcer said the Yankees told him the day he was released. He said he was being replaced by this kid Mattingly, who was told would never be able to hit major league pitching, but the Yankees sure did like his glove.

[byzeil]

Supposedly the scout isn't just evaluating a player by his in-game performance. They would see things we cannot see in-game like batting practice, bullpen sessions, workouts, etc. They would also talk to his coaches and other scouts and maybe even players. So a scout may make an educated guess that a player's skills may be on the decline even though the player may just go on to have a career year. A player can also out perform (or under perform) his ratings. Even as some rating decline he could still have a career type year but I wouldn't expect it to continue long term.

Originally Posted by Mike Donlin

At the start of spring training, my scout reported that his power rating had dropped.

Originally Posted by Mike Donlin

But this strikes me as deeply unrealistic. Mainly, that a scout would insist, against evidence, that a player is washed up (or on the way to being washed up) in the midst of a career year.

As the poster before me said, the scout reported that drop in ratings at the start of spring training BEFORE he was in the midst of a career year. The scout could have been wrong or the player (if he were real) could have taken that as added incentive. The scout may change his mind or the player may start to decline.

The scout reporting his power rating has dropped is far from saying the player is washed up. As players get older their power may decline but other skills may not. He could slowly change into a different type of hitter. Power hitters can loose power and become better contact hitters or be more patient with balls and strikes.

Originally Posted by Mike Donlin

You should have to discover his declining ability for yourself, after he fails, when your scout could then tell you, yeah, he's washed up. But a scout should not be able to see into the future, especially when the future goes so strikingly against the present.

Then why use scouts? Their job isn't to just confirm what you already know, that the player is washed up, it is to make an educated guess as to what he will do in the future. That is what scouts do. I can see he can hit now but I would rely on the help of the scout and his experience to try to predict where his career is going.

Originally Posted by Mike Donlin

But a scout would never say that a 30-year old center fielder who is tearing up the majors is soon going to crash to earth. He might say he can't keep it up, and will return to his normal average self. But he would never say he's going to be worse than he was.

A scout would say that. That is why Major League teams send scouts to look at players they are considering trading for or signing as an upcoming free agent. If you only judged on past/current performance why would you need a scout, you would just look at his stats? The scouts earn their reputations by predicting how good a young player may become and when it is time to pass on acquiring veteran players.

It sounds to me that you would rather play stats only and evaluate only on performance which is a perfectly valid option.

[Tyler87898]

I don't really have much to add to this, but I noticed that your player that you're talking about sounds a lot like the Marlins Justin Ruggiano. Last year, as a 30 year old outfielder .313/.374/.535 in 91 games, and this year he is hitting .235.

[Mike Donlin]

Originally Posted by Tyler87898

I don't really have much to add to this, but I noticed that your player that you're talking about sounds a lot like the Marlins Justin Ruggiano. Last year, as a 30 year old outfielder .313/.374/.535 in 91 games, and this year he is hitting .235.

I don't doubt that players have career years or players can outperform their ratings or anything like that. I do doubt scouts' ability to predict that a player will be WORSE than what he was before his career year.

I appreciate the defenses people are making of this, since I am trying to salvage my belief in the realism of the game. But it strikes me that people are giving scouts much more perceptive (and predictive) ability than they can possibly have.

[byzeil]

It is one thing for a scout to say a 19-year old prospect is going to blossom even though he is struggling now, or that a different 19-year old prospect is nothing special even though he is tearing up the minors. But a scout would never say that a 30-year old center fielder who is tearing up the majors is soon going to crash to earth. He might say he can't keep it up, and will return to his normal average self. But he would never say he's going to be worse than he was.

I don't get your logic here. Why can a scout predict improvement/decline of a 19yo but not a 30yo? Seems to me he has more info to go on with the 30yo. Unless your saying 'he wouldn't dare' predict a 30yo's decline. Just don't see your logic.

You don't seem to believe a scout can predict on anything other than performance then I guess you should change the 'AI Player evaluation options' to heavily favor stats and set the ratings weight to zero or some such number. Then, as best I know, the scouts would not take the ratings into consideration

[Mike Donlin]

Originally Posted by byzeil

I don't get your logic here. Why can a scout predict improvement/decline of a 19yo but not a 30yo? Seems to me he has more info to go on with the 30yo. Unless your saying 'he wouldn't dare' predict a 30yo's decline. Just don't see your logic.

You don't seem to believe a scout can predict on anything other than performance then I guess you should change the 'AI Player evaluation options' to heavily favor stats and set the ratings weight to zero or some such number. Then, as best I know, the scouts would not take the ratings into consideration

I am assuming that it's a scout's job to project how a player will perform in the major leagues. At some point, his projection would have to give way to how a player is actually performing in the major leagues.

I would change the AI Evaluation, but having ratings set to zero then gives me an advantage over the AI. I know this player is about to tank. And not even stats only would change that--there is still the written scouting report, which also changes.

[Fyrestorm3]

Originally Posted by Mike Donlin

I am assuming that it's a scout's job to project how a player will perform in the major leagues. At some point, his projection would have to give way to how a player is actually performing in the major leagues.

I would change the AI Evaluation, but having ratings set to zero then gives me an advantage over the AI. I know this player is about to tank. And not even stats only would change that--there is still the written scouting report, which also changes.

Do you? So far you've told us he's in the midst of a career year. I would like to see what happens down the road.

Scouts can be wrong in OOTP, you know - that's kind of the whole point of the scouting system.

[Mike Donlin]

Originally Posted by Fyrestorm3

Do you? So far you've told us he's in the midst of a career year. I would like to see what happens down the road.

Scouts can be wrong in OOTP, you know - that's kind of the whole point of the scouting system.

You may have missed it in the opening post, but I opened up the editor and checked his actual ratings, scout-free. His rating had dropped. A lot. And looked to drop more.

[olivertheorem]

Which means your scout is right, the guy will crash to earth sooner or later, and you may want to try selling high before that happens.

You KNOW for a fact that the scout is right. The in-game abstractions for gameplay's sake are irrelevant to the fact that you have a scout who, correctly, identified a player who had lost power. That he is not playing to that rating at the moment means you have to guess whether or not (A) his ratings came back up and you just haven't had another scouting report yet, or (B) he's just getting lucky and you move him before he starts playing like the guy your scout says he is.

[Mike Donlin]

Originally Posted by olivertheorem

Which means your scout is right, the guy will crash to earth sooner or later, and you may want to try selling high before that happens.

You KNOW for a fact that the scout is right. The in-game abstractions for gameplay's sake are irrelevant to the fact that you have a scout who, correctly, identified a player who had lost power. That he is not playing to that rating at the moment means you have to guess whether or not (A) his ratings came back up and you just haven't had another scouting report yet, or (B) he's just getting lucky and you move him before he starts playing like the guy your scout says he is.

I'm not sure what in-game abstractions you're referring to, maybe the actual ratings, which I take as something like the player's true talent level, but in any case we've come full circle. I don't think it is realistic for a scout to predict failure in the face of success.

[The Wolf]

Why not? They predict success in the face of failure all the time.

[Fyrestorm3]

Originally Posted by Mike Donlin

I'm not sure what in-game abstractions you're referring to, maybe the actual ratings, which I take as something like the player's true talent level, but in any case we've come full circle. I don't think it is realistic for a scout to predict failure in the face of success.

I'm no scout, but I've been known to look at a player and think "there's no way this success is going to last". Clay Buccholz, for example - he throws a no-hitter in his second big-league start, and then continues to do well the next year. And the entire time, I sat there adamantly stating that his success would never continue, that he was being overrated and would never be an ace.

And hey, I was wrong. But the point is, it IS possible to predict failure in the face of success. They do it all the time on sports radio, so why should scouts be any different?

[byzeil]

Originally Posted by Mike Donlin

I am assuming that it's a scout's job to project how a player will perform in the major leagues. At some point, his projection would have to give way to how a player is actually performing in the major leagues.

Scouts don't stop looking at players when the hit the Majors. MLB teams still employ major league scouts to not only prepare for upcoming opponents but also to evaluate possible trade/fa acquisitions. If the only evaluation that was used was to look at their current performance then the GM could do that over breakfast reading baseballreference.com.

Originally Posted by Mike Donlin

I would change the AI Evaluation, but having ratings set to zero then gives me an advantage over the AI. I know this player is about to tank. And not even stats only would change that--there is still the written scouting report, which also changes.

Well yea changing the AI evaluation will not help with this player since you have looked at his actual ratings. Nothing is going to help once you see the real ratings but with the change in AI evaluation for scouts on other players would be based on performance only.

[Mike Donlin]

Originally Posted by Fyrestorm3

And hey, I was wrong. But the point is, it IS possible to predict failure in the face of success. They do it all the time on sports radio, so why should scouts be any different?

Because I would like to think scouts are better than sports radio.

But, OK, you've convinced me. Or, if not quite convinced me, at least given me a cleaner conscience about selling high on this player. Pity the poor GM who buys high.

[Mike Donlin]

Originally Posted by byzeil

Well yea changing the AI evaluation will not help with this player since you have looked at his actual ratings. Nothing is going to help once you see the real ratings but with the change in AI evaluation for scouts on other players would be based on performance only.

No, I mean that if you set ratings to zero in AI Evaluation, regardless of whether you have looked at a player's actual ratings or not, I will consider ratings in my evaluation but the AI will not, and that would give me an advantage. I will never buy high. Or sell low. The AI will.

[byzeil]

Originally Posted by Mike Donlin

No, I mean that if you set ratings to zero in AI Evaluation, regardless of whether you have looked at a player's actual ratings or not, I will consider ratings in my evaluation but the AI will not, and that would give me an advantage. I will never buy high. Or sell low. The AI will.

There are a couple options to make the game more difficult for yourself and make the scouts more like how you see them. You could do some combination of:

-Set AI evaluation to not include ratings. My understanding this affects the scouts overall rating of a player.
-Set the ratings scales to 1-5 so little granularity.
-Set scouting accuracy to low or very low.
-Do not display current ratings. Your current eval would be based on performance but the scout still gives you an idea of potential (with a possible 1-5 value)
-Do not display current or potential values and have to evaluate strictly on performance.

[Mike Donlin]

Originally Posted by byzeil

There are a couple options to make the game more difficult for yourself and make the scouts more like how you see them. You could do some combination of:

-Set AI evaluation to not include ratings. My understanding this affects the scouts overall rating of a player.
-Set the ratings scales to 1-5 so little granularity.
-Set scouting accuracy to low or very low.
-Do not display current ratings. Your current eval would be based on performance but the scout still gives you an idea of potential (with a possible 1-5 value)
-Do not display current or potential values and have to evaluate strictly on performance.

Yes I think the solution is to turn ratings off and leave potential on. (I already play on a 2-8 scale.)