You dipped your Scout in my Text Sim

[Bonegavel]

There has been a lot of discussion of scouts lately in various threads, and I wanted to start one dealing with the subject exclusively. Every Sport Text Sim uses them, and I have to admit that, while understanding that they exist in the real world, I despise them in game.

Why do I despise them? Well, first off their ratings are imperical, meaning that there isn't a scout scout, and so, it must be true that if they are "POOR" in a field, they are poor, etc. This leads me to wonder why anyone would hire a scout they know is POOR. Ok, you say, he is the only scout available. Well, then how did he get to be a scout in the first place?

I'm going off the assumption that to get to a certain level of scouting, you need to be, at the very least, GOOD. Otherwise, you wouldn't be much of a scout. I realise that some scouts must be better than others, but a scout that is known to be POOR would be tossed out on his ass and go into coaching, or something. :)

OK, so, why are they in the game? After a player has been in the league for a set amount of time (?? 3-4 years-assuming lots of play time) little to no scouting is needed for raw talent as it should be generally known. So, scouts are ostensibly used to rate new (draft) and young talent from around the league. A coach has a finite amount of time to spend analysing talent, and must rely on Scouts to do his dirty work.

Fine.

But, I have found the translation to in-game scouts to be frustrating. Suppose my scout for QBs is POOR or SUCKS at rating QBs. Does this now mean that I can't trust a thing he reports to me regarding QBs? Will all his ratings be so wrong that it turns into a crap shoot? I have had this happen many times where I can only hire scouts that aren't even mediocre. I feel cheated.

This may be 6 of 1, half dozen of another, but I would much rather like to see percentages for scouts. IOW, check the scouts initial grade against what the player becomes, and assign a percentage to how right he is. If I know my QB scout is 35% accurate in his assessments, I feel much better about using (or not using, in this case) his advice than if I see he is POOR.

The final assessment that his scouting is based upon could be created when the player reaches year 4 (similar to when the potential bars in FOF4 disappear) and based upon some set of criteria, give him a grade and see if it matches what the scout guessed.

The number won't be 100% accurate, I know (injuries could have hampered a player's performance, but this can be flagged in the grading - or maybe injured players aren't used in the calc because of the potential tainting), but it is a least a baseline for judging.

Why do you like/dislike Scouts in the game?

How can it be improved?

[mckerney]

I hate it when you scout is responsible for evaluating the talent on your roster.

[Marc Vaughan]

Why do I despise them? Well, first off their ratings are imperical, meaning that there isn't a scout scout, and so, it must be true that if they are "POOR" in a field, they are poor, etc. This leads me to wonder why anyone would hire a scout they know is POOR. Ok, you say, he is the only scout available. Well, then how did he get to be a scout in the first place?

I tend to presume that such scales are on a level with other scouts rather than the 'general populace'.

This is how CM works, for instance a player with a pace of 1 isn't the slowest person in the world - but he will be one of the slowest professional footballers in the world (ie. about treble my speed ;) ).

[SunDancer]

Don't scouts have a rep in real-life though?

How would you do it?

[mrsimperless]

When I look at a scout's ability as being POOR, I assume that it is in relation to tother scouts in the league, not in general. If you hired me to do your scouting my ratings would most likely be below poor, and not even representable in the game. He is still a good scout, but compared to other scouts in the league he can be considered POOR.

[damnMikeBrown]

You dipped your text-sim in my Scout. Uh oh...here comes Atticus! Run!

[Bee]

I have no problem with the way most games handle scouts. Like everyone else is saying, the ratings are in comparison to other scouts. I also look at it as the scout being representative of a scouting department. Every team has a reputation as to the quality of their scouting department, so I think it really needs to be in games in one manner or another. I'm not sure if there's really another approach that would work.

[Ben E Lou]

Hmm....good discussion idea. I have several thoughts:

The Inherent Problem With Text Sims...
...is that I am NOT really the Head Coach or Manager. I don't see the team practice every day. I don't get to look at film. The only evaluation I can do is on a report about the player's performance, whether that is in practice, spring training or games, and no matter what, that report is not a complete replacement for the real-life experience that the Head Coach/Manager gets. Let's say it is a baseball spring training report that says Joe Schmo hit .256 in the Spring. That report doesn't show that he made 10 outs on line drives hit right at the fielder, or that he got 30 at-bats against minor league caliber pitching. Or, a preseason game report might tell me that my RB had 12 carries for 35 yards, but it might not tell me that 7 of those runs came when the starting guard was on the bench, and the backup stinks. My point is this: without gobs and gobs of statistical information about every practice and every game (wayyyyyyyyyyy more than we'd ever want to have to sift through), a text sim can't adequately reproduce actually being there at training camp, spring training, practice, and games. In order to make the experience meaningful, there HAVE to be ratings for players, and just like in real life, those ratings can't be 100% accurate. Whether it is "my" ability to evaluate players that is flawed or my "scout's" ability that is flawed, there needs to be a lack of 100% accuracy. How that is represented is up to the developer. Now, I like the direction TPF is going, in that my evaluation of my players will be more accurate than my scout's evaluation of Green Bay's players.

Stats or Scout?
In some sports (football in particular) it is quite difficult, if not impossible, to evaluate a player based on his stats alone. Someone was watching one of the Thomas Brown videos and commented, "with that offensive line, I could get 5 yards per carry," or something to that effect. Again, we need ratings, and they can't be 100% accurate.

Potential
I don't know if it is "me" or my "scout" but someone has to give a "potential rating" for players, and again there needs to be significant ability to blow it. Anyone here (besides Braves fans from the early-to-mid 80's) heard of Brad Komminsk??? No??? Why not??? He was "can't-miss", "the next Dale Murphy", "future Hall of Famer". EVERYONE in the Braves organization was talking about this kid. He had a career .218 average, .336 slugging percentage, and never got more than 301 at bats in a major league season. We could all tell stories of "can't miss" guys like this who never amounted to anything. Again, there needs to be a "potential" rating, and it needs to be imperfect.

Poor, In Relation To What?
This has been brought up before, but perhaps he's the worst professsional scout there is. I'm pretty sure Jim said he does this with his player ratings. While you and I may think of a QB who is rated 0/5 overall as your cousin who was the backup QB in high school, in reality that 0/5 guy was a starter or backup on a D-1 squad.

Final Comments
BoneGavel, I think the use of scouts is a flaw inherent in the system of text sims. I don't really see any way around it, apart from giving us mountains and mountains of data, which would likely be too much to process.

[Ben E Lou]

Dola--

How I'd Like To See "Scouts" Handled
1. The evaluation I am given of my players' "current ratings" should be MUCH more accurate than the evaluation of other teams' players, but their still should be some margin for error.
2. The more experienced a player gets in the league, the more accurate everyone's (AI teams and my team) view of his current ratings should be.
3. Potential ratings for some marginal players should have the ability to be widely varied by different scouts.
4. Potential ratings should be category-by-category, not one overall rating. (My freshman DE may be a better pass rusher now than he is a run-stopper, because he's 6'5", 245 pounds right now, but his potential may be as a better run-stopper because his tall frame could hold 280+ pounds eventually.)
5. There needs to be the occasional guy with high potential ratings across the board and high "combine" ratings (combination of all scouts) who completely doesn't pan out at all.

[Eaglesfan27]

Originally posted by SkyDog


The Inherent Problem With Text Sims...
...is that I am NOT really the Head Coach or Manager. I don't see the team practice every day. I don't get to look at film. The only evaluation I can do is on a report about the player's performance, whether that is in practice, spring training or games, and no matter what, that report is not a complete replacement for the real-life experience that the Head Coach/Manager gets. Let's say it is a baseball spring training report that says Joe Schmo hit .256 in the Spring. That report doesn't show that he made 10 outs on line drives hit right at the fielder, or that he got 30 at-bats against minor league caliber pitching. Or, a preseason game report might tell me that my RB had 12 carries for 35 yards, but it might not tell me that 7 of those runs came when the starting guard was on the bench, and the backup stinks. My point is this: [u]without gobs and gobs of statistical information about every practice and every game (wayyyyyyyyyyy more than we'd ever want to have to sift through), a text sim can't adequately reproduce actually being there at training camp, spring training, practice, and games.

I think these are excellent points. It appears TPF is attempting to go through this data with missed blocks, poorly run routes, bad throws, etc. I think more games need to summarize practices and data in this way. I'm really hopeful that TPF will be able to implement this successfully to augment our scouting data.

[Bonegavel]

Nice comments all.

Hmm... What I would like to know then, if a POOR scout is still "really good" then, I would like to know how much his opinion is wrong. I think what is bothering me is that scouts are good, but there are a few who are really good. I guess it would make more sense to me to have 90% of scouts in a game be like 75% accurate and have a few that are 90% and a select few that are 95%. (numbers off the top of my head).

Now, maybe this is what goes on in these games (or close), but, not knowing, I have to assume that my POOR scout is WAY OFF the mark. I don't want to pour through every pick for every scout and see how wrong or right their initial eval was (the game should handle this for me), so by not knowing what POOR exactly means, I think is a disservice.

I really think all scouts should be basically equal, and only a few (after some years in the league) should stand out as REALLY GOOD. That way, you have a 75% chance of getting good data (not bad odds) from the majority of scouts, but when you find that Super Scout (the game just might add a star to his name after a few years) you fight to keep him, or get him.

Having all the different levels of scouts makes me nuts. The more I think about it, the more I like that approach the best. Most scouts are equal, and eventually a few rise to the top and garner reputations for certain positions or just scouting in general. And don't get me wrong, even the Uber Scouts will be wrong on occasion, just far less than the rest.

As you can see, it is really hard for me to put into words why I dislike scouts in Text Sims, but I think as this discussion continues I will be better able to excorcise this daemon. Or at the very least, I can be convinced that they are worth the pain.

[Bonegavel]

dola

Ok. In an effort to better organise my thoughts on this, I went outside for a smoke and gathered my wits. Once back inside, I fired up FOF4 and took a look at my scout, Rusty Salisbury and he has the following stats:

QB: Good
RB: Avg
WR: Avg
OL: Fair
K: Very Good
DL: Very Good
LB: Good
Secondary: Fair
Young Talent: Very Good

The first thing that hits me is: Good Lord, that is a lot to track. When I first started playing I used to print out the Scout attributes and compare it to each position as I was looking for talent, but that quickly becomes more tedious than sitting through a work meeting at 4pm on a Friday before a Holiday.

First of all, I have to scour players that have many attributes (this part I don't mind), then I have to cross reference my Scout's attributes to the position to see how accurate he is, and then, wonder how his ratings affect what I am seeing. Doing this repeatedly makes me quickly stop worrying about my scout and I just look at the numbers at take them at face value. That defeats the purpose of the scout, and if I am not alone in this way of doing things, the scout-system is flawed.

Ok, now, suppose I am looking for a Running Back. My scout is Avg with RBs. Does this mean that I should ignore the scouting report and pick an RB blindly? Do I need to keep track of every RB I pick based upon his opinion and calculate myself how "accurate" he is? If the answer to either question is yes, I am in trouble, because that is more like work than a game.

If the difference between Poor, Fair, Avg, and Good are minimal (say, only a few percentage points), then why have them? Shorten the list to Average (majority of scouts) and Above Average (a select few). However, if the differences aren't minimal (meaning there is a big discrepancy between the ratings), then we have the problem of why would anyone hire a guy that is POOR, and, without knowing his percentage accuracy, those positions become dart-board picks.

I might be nit-picking here (between a moniker and an actual percentage for ratings) but this is a game and, as the player of said game, I would like to make decision based upon percentages and not something like POOR, or FAIR. I think it is much more fun to hire a scout who might look like this:

QB: 80%
RB: 75%
WR: 75%
OL: 60%
K: 85%
DL: 85%
LB: 80%
Secondary: 60%
Young Talent: 85%

Decisions can be made off of this information. I need to look for a QB. My scouts rating is 80% with QBs. That is a percentage that I am comfortable with and I know that there is a 20% chance that he is wrong, but, again, this is a game and I need things to help me make decisions because, as SD mentioned, I can't watch them practice or have access to information that leads to their stats. Next, I need to find an Offensive Tackle. I see that he is only 60% accurate there, so this an area that I might pay closer attention and may have to make the decision to bring in a known veteran because the scout is just too weak in that position.

Like I said, I hate to be nit picky about it, but I think I could live with a scout if I knew hard numbers as opposed to Name Categories. And, only because I can make a fairly quick decision based upon a number as opposed to thinking, "ummm... he is Fair with this position, so does that mean the numbers are good? Oh wait, is Fair better than Good? Is Good less than Average? Ahhhhh!"

[GrantDawg]

I think that scouting should be determined by budget. The fact is teams that have many scouts/higher paid scouts do a better job than teams that have a smaller scouting budget. No team has one scout, and even the head scout isn't final arbiter of how good a scouting team is.

[Honolulu Blue]

I mostly like the scouts as they are implemented in FOF. Seeing your players and their ratings through scouts' eyes is much better (and realistic) than either being given perfect information or merely having some random variation.

The one things that annoys me about scouts - and coaches - in FOF is that we seem to know perfectly what THEIR strengths and weaknesses are. I'd be happier with a system that gives out just one or two strengths and weaknesses. To use BoneGavel's example with Mr. Salisbury:

Strengths: DL, young talent
Weaknesses: OL, secondary

This, I feel, gives out more of the kind of information that I as a GM would know about a scouting team.

For offensive players in their prime I tend to trust the stats more than I do the ratings, no matter how good my scouts are. If a RB is averaging 4.5 YPC despite modest ratings, SOMEWHERE his ratings are good, and I have a good feel for adjusting for the factors around him. Scouts are best for two groups of players:

1) Young players, who have limited or no stats, and
2) Older players, who may be losing skills that are not yet captured in their stats.

[korme]

SkyDog, great post. You really opened my mind up and I agree 100%.

[Vince]

Originally posted by SkyDog
Anyone here (besides Braves fans from the early-to-mid 80's) heard of Brad Komminsk???

Alan T has :)

Anyhow, as for my opinion; BoneGavel, I think that giving a percentage is too much information. Like you said, sifting through all the reports to find out how often a scout was right or wrong on a player is too much work...but generally speaking over time I'd think you would develop a feel for your scout without having to do this.

That being said, I really appreciate SkyDog's input...his ideas, especially that of established players not needing scout variance, seem to be spot on, but I'm curious about the potential ratings being position-by-position. Isn't that what the bars do now anyways?

[Ben E Lou]

Originally posted by Vince
...but I'm curious about the potential ratings being position-by-position. Isn't that what the bars do now anyways? In FOF games, yes, but not in all text sims. I think PureSim is one example of a game that just has one "potential" rating.

[Bonegavel]

I do like the idea of allocating money to scouting, i.e., the more you throw at it the better your scouting team. However, the percentage idea works if you put the scouts percentage right above or below the players stats.

Scenario: You're looking for an RB and you open the player card. Right above the players attributes is Rusty Salisbury and his rating of 75%. You don't have to look up anything. Now you know that Rusty is right 3 out of 4 times with his analysis. Make a choice.

With the "allocate money to scouting option", do you allow them to throw as much money as they want at scouting? Granted I'm no communist, but this is a game and you don't want to skew things out of control. It seems obvious to me that being able to "unlock" a player's ratings through scouts would be one of the greatest advantages in the game. Being able to pick consistently great talent would make the game boring and unplayable by anyone with sense.

As far as established players not needing scouting, I think that is true for most players and is implemented in FOF4. However, there are those guys who have been in the league for a long time and have been in the shadows or not given much playing time, and it would be sweet to occasionaly pick-up one of these diamonds-in-the-rough late or midway into their careers when the game would normally stop letting scouts taint a player's stats.

Again, to me, the bottom line is a slippery slope. If you have a great scout, you have an advantage in picking talent. Obvious. If you have a GomerPyle Scout, you are basically entering a crap-shoot when picking talent. Not fun, and potentially frustrating.

You can't have a system where the values are empirical and absolute (that is agreed by all). You also can't have a system where picking players is a coin flip. We have proved these truths to be self evident many times here. What we do want is the ability to to make choices that fit our visions and strategies and, while not guaranteeing victory, we want to know that our choices produce the desired results, or at the very least, have some measurable impact more than chance would have allowed. We want to know that our choices have skewed the bell-curve produced by the game's random number generation; Sometimes for the good, sometimes for the bad, sometimes for zero gain. But, it better not feel like I am simply pressing the TURN button like in Masters Of Orion III.

The above may not be realistic, because in the NFL, it seems that the draft is like the Lottery (pick a stone), where first rounds picks fizzle and that guy in the 6th turns into Mega Star more often than not. But, this isn't the real NFL. It is a simulator and game where you make choices to skew results.

Sorry for the blah, blah, blah. This subject interests me greatly.

[SunDancer]

What about those established players who are like an Alvin Harper or a Tommy Maddox. They thrive with a team, but bloom with another.

What about the system from NCAA 2004 on the video game systems.

You get a player's combine ratings, and ratings in "arm strength" like "cannon" or "good", "accuracy", ect....You get an overall rating on a five-star system. I think a major effect on scouting would be including realistic college stats that would matter. For example, say you are looking hard in the third round at a Running back out of Ohio State.

He's rushed for over 1,000 yards the last two years as a starter, but he's averages just over three yards a carry. However, he runs an late 4.30's time. Is his football speed slow? Or does he have problems in his acceleration or breaking tackles? This is where your scout can come in.

[Barkeep49]

Scouting is one area where I would really love a game that allowed me more control. I think the TPF/CM approach to scouting, where you have imperfect information and (at least in CM) can control which areas you know more about is a winning formula. This to me is more important than the Poor/Good situation.

[SunDancer]

Also,
I think if you can send "scouts" to games, include a backdrop of college football season in the game, the better the ability to see a player. His experience should improve his abilities. On another note, how bout including a "track record" for scouts. Like, say he was with Cincinnati and scouted Akili Smith, and that will show up. Or say he helped New England scout Tom Brady, and that shows up. Give me his track record, where has he been and who give me a breakdown of wheather he "pushed" or opposed those picks on the team.

[Barkeep49]

Originally posted by SunDancer
Also,
I think if you can send "scouts" to games, include a backdrop of college football season in the game, the better the ability to see a player. His experience should improve his abilities. On another note, how bout including a "track record" for scouts. Like, say he was with Cincinnati and scouted Akili Smith, and that will show up. Or say he helped New England scout Tom Brady, and that shows up. Give me his track record, where has he been and who give me a breakdown of wheather he "pushed" or opposed those picks on the team.

This is exactly the sorta thing I would like to see.

[Bonegavel]

In addition to this, I think it would be nice to also have the game give you information about the player to flesh them out.

In the NFL, you know the super stars. They get the million dollar advertising deals, they are talked about on Sports Shows, they give press conferences after games. They are basically the "talk of the town."

I think FOF4 addresses this with the Popularity attribute, but that isn't much. Wouldn't it be nice if in your Email-bin, you get reports that tells you your star RB just signed a $10 million deal with a shoe company, or your QB is being panned in the media as a bungling idiot.

Of course, this information doesn't necessarily mean they are the best at their position, but most advertising agencies aren't going to give a crappy player millions of dollars to push their products.

[SunDancer]

Originally posted by BoneGavel
In addition to this, I think it would be nice to also have the game give you information about the player to flesh them out.

In the NFL, you know the super stars. They get the million dollar advertising deals, they are talked about on Sports Shows, they give press conferences after games. They are basically the "talk of the town."

I think FOF4 addresses this with the Popularity attribute, but that isn't much. Wouldn't it be nice if in your Email-bin, you get reports that tells you your star RB just signed a $10 million deal with a shoe company, or your QB is being panned in the media as a bungling idiot.

Of course, this information doesn't necessarily mean they are the best at their position, but most advertising agencies aren't going to give a crappy player millions of dollars to push their products.

While it's an good idea in theory, how would it actually have an impact on the game? You know who the "stars" are through MVP's, stats, contracts and the popularity rating in the games now, how would such a feature play a role?

[The_herd]

Originally posted by SunDancer
While it's an good idea in theory, how would it actually have an impact on the game? You know who the "stars" are through MVP's, stats, contracts and the popularity rating in the games now, how would such a feature play a role?

Anything that makes a game more realistic, and brings you "closer" to the real game is welcome. These suggestions are definately fluff, but its fluff that doesn't make the game more tedious in any way.

When you sim multiple seasons, the "stars" kinda get lost because things get blurry as the seasons pass. The more you see a players name, the more his name will stick. If you were to get an e-mail saying your star QB just signed a record contract with Nike, you're going to most likely remember the name. At the same time you could have a player hold out for more money because he has a big advertisement deal and can afford the hold out.

I think games need little things like these to bring you closer to your players. Something to make you think twice before letting your star RB go to free agency because he lost a step. I'm all for anything that will get you more attached to your players.

[SunDancer]

But in reality, most of the shoe deals, ect...happen during the off-season. I think that maybe a page from NCAA 2004, with SI covers.

[Vince]

Originally posted by BoneGavel:
they give press conferences after games

This right here I think is the key to a more CM-style media interaction. I think it wouldn't be too hard to have some of the higher tier players or even the coach have a press conference after a big game or after a key injury. This ties in to something else I'd like to see done better, team chemistry. As of now, we have the astrological symbols or cohesion ratings in FOF, but that doesn't begin to describe a lot of the cooler things in sports. Something like the whole Keyshawn-Chrebet rivalry type thing that happened a few years back when Key called Chrebet a 'flashlight,' or something like Brian Giles wanting Kendall to come join him in San Diego. Some might call it fluff, but I think that something like that would add a huge immersion factor to the game.

[GrantDawg]

As a follow-up, Arlie just laid out his scouting plan. It does not have the scouts looking at your team (that I can tell) and follows the pattern I suggested (putting money into areas gets better results). You can check out the discussion here. (http://www.400softwarestudios.com/invboard/upload/index.php?showtopic=42111&st=0)

Again, this is a great way of doing things. Can't wait to see the final product of this game.

[MrBug708]

One thing I don't get is when, in FOF4, my RB has pitiful ratings, yet manages to run for 1100 yards. If I were to convert my OL to a RB, he'd have better ratings no less

[Bonegavel]

Thought I'd revive this thread in light of Shaun Sullivan's update to PureSim (AKA Total Pro Baseball 200X). The link above from grantdawg to Arlie's discussion on TPF's handling of scouts is a good read, if you haven't already done so.

Shaun's approach is to just get rid of 'em:

snipped from shaun's update
No more scouts

As I mentioned in the previous updates they are gone. This has resulted in a much tighter game experience and as I expected an AI that is making decisions based solely on statistical performance (past and present) of players is yielding a much tougher computer opponent. The catch here is the initial draft. How can the AI make a determination on player value when there is no previous performance to evaluate? I am still on the fence here, and can go a few different ways I'd love to hear from you on which approach you prefer:




1. Keep ratings like contact, power etc. but simplify the scale (e.g. 1-10 or A, B+, C etc... This way you'd still get a general idea of a player's talent for the initial draft.

2. Run each player through the simulation engine to build up a representation of their average Major League stats. So when doing the initial draft, you'll see a 28 year old LF with .287 27 HR 31 Doubles etc... These stats represent his avg season in the pros. For amateurs/young players these stats might be post-fixed with (AAA) or (AA) or (College) meaning they represent how that young player has performed at a certain level up to now. Basically the game shifts from being ratings-based to stats based. Of course some ratings stay around (for example Speed and Velocity) because those are truly accurately measurable.

3. Keep the ratings exactly like they are. Now without the scouts this means they are a very accurate representation of how good a player is. Too accurate you wonder? Well, that is the crux of the issue. Interestingly, with a dynamic stats and player development engine variances happen "organically". So, things are certainly dynamic enough that they are not formulatic. Just because a guy has a 90 POWER rating doesn't mean he is a lock to hit 40 HR. That said, it is likely he will be one of the top power hitters.


maybe a bit of a group think can help shaun choose, or even find a new way to handle this area.

[Easy Mac]

I don't really have a problem with scouts, but more in the way they are used.

I think it should be a combination of their current use and the way the "fog of war"is used in CM.

Basically, I sholdn't know how good players are who I havne't actually scouted. I may have some info on them, and can get some rough ratings based on reputation (I know Barry Bonds can hit home runs, but I haven't sent a scout to see him).

Then I can send a scout to watch a team, and he'll give me ratings based on how good of a scout he is. If he is poor at scouting certain areas, the ratings he gives you in that area would be farter off than a better scout.

However, as soon as you watch a player play, you should get a better idea of how good the player is, dependent on your skills you choose as a manager (Have an option to say whether you are good at noticing power, young players, bunting, pitching... put it on a relative slider scale). That way, you'll have your own idea of how good players are. This would be the viewable option of players on your team or teams you've played against.

However, you also can't tell when a player gets better until you actually see them play again (although stats would obviously give you an idea).

I think scouts should be another tool in the process, not the be all, end all of the process.

And I by no means think they should be eliminated.

[Maple Leafs]

Good topic, I'm sorry I missed it the first time around.

To be honest, I'm getting tired of scouts in text sims too. In addition to the problems mentioned above, every human GM will always spend top dollars for scouts, giving them an advantage over the AI.

What I'd rather see are generic, high-quality scouts that you could assign to various tasks based on your own strategy. Consider it a "Scouting Budget" feature.

For example, you could have a screen with several options, such as:

- Scout draft eligible players
- Scout own players
- Scout next opponent
- Scout league in general
- Scout free agents
- Scout specific players
- Scout coaching tendicies
- etc.

You'd assign points, percentages, whatever to each group to determine what kind of scouting information you'd get. Every team has the same number of points to spend. You wouldn't get to "buy" the best scouting staff -- you'd need to determine your own organizations needs, and deploy your scouting resources accordingly.

Of all the text sims I've played, EHM comes closest to this sort of set up. That's really the only sim I've played where scouting was a fun feature, not a pain.

[corbes]

EHM did have the most interesting scout arrangement I've seen, although it was a bit tedious for me to eventually see every team in the league.

Seems like the quality of your ratings could depend on how many times you see a player, rather than universally the quality of your scout.

Of course, now that I'm thinking about Shaun's math, that might make things worse instead of better.

[Fido]

Originally posted by Easy Mac
However, as soon as you watch a player play, you should get a better idea of how good the player is

I don't disagree on anything you siad, I just have a question about this (in case I ever getting around to writing my own game). When scouting a player, would you see the scouting information you get based on the player's actual abilities or how they played in the games you scouted them in? If your scout travels to a game and witnesses someone having a career day, woudl the scout report that the player is phenomenal or would he see through it to the player's true mediocrity (and vice versa for players on bad days)? How would scouting multiple times come into play?

[WSUCougar]

I’ve always had minor issue with scouting recruits in TCY. An overall score that fluctuates given the scout’s quality, but except for one strength and weakness, nothing more.

I love the idea of the wide or narrow blue scouted rating ranges of the draft in FOF 2004. That could work in TCY. Alternatively, the ability to focus on X number of high school players to get a more detailed look would be ideal.

[Easy Mac]

I think a player needs to watched more and more in order to get a better read on a players ability. I don't think a scouts view can purely reflect the stats, but maybe a combination of stats and real ratings. Not sure the best way to implement it, but a scout shouldn't watch a player once and say "oh, so and so is a 10 in hitting and a 9 in power", but instead should have certain abilities appear overa few visits. For the most part, you can tell how fast a player is almost right off, or his work ethic or arm strength. However, things like his overall power, or contact or overall fielding need to come out slowly over the course of a few viewings.

Of course, a game like football, where you see a player only once a week (instead of 3 or 4 games at a time like baseball), the system would need to be adapted.

[nilodor]

A long time ago there was a project conducted on this board, maybe the old one where someone using FOF 2001 during the hire/fire staff period preceded to hire staff of varying ratings. For example he hired a scout that was very poor at QB's than went to his roster and took down the ratings. Than hired a scout with poor, than good and so on and so fourth and put together an extensive analysis of scouts and there accuracy or lack there of.

I did a quick search for this but couldn't find it. I was wondering if anyone else remembers the experiment or was willing to undertake one of their own.

[JonInMiddleGA]

Apparently lMaple Leafs had the same thought I was having --
I thought the EHM approach was the best I've seen.

[Maple Leafs]

Originally posted by JonInMiddleGA
I thought the EHM approach was the best I've seen. Yet another reason to look forward to the new version.

Um, when is that damn thing coming out, anyway?

[Karim]

Yet another reason to look forward to the new version.

Um, when is that damn thing coming out, anyway?
The only indication we've been given is the first quarter of this year. I'm guessing they coincide the release with the start of the Stanley Cup Finals. :D

[Karim]

Dola,

I really like Maple Leafs idea on scouting. TPF has this sort of breakdown, where you have a limited amount of total money to spend on various areas of scouting. In this way, it's impossible to perfectly scout anything.

In FOF2004, I try and be "realistic" when hiring a scout. Knowing that scouting costs are 45x the salary of the lead scout, I pick a number I won't go over and stick to that.

Expanding on Maple Leafs' list:

- Scout draft eligible players
- Scout own players
- Scout next opponent
- Scout league in general
- Scout free agents
- Scout specific players
- Scout coaching tendencies

- Scout other leagues
- Scout other countries
- Scout specific positions
- Scout for player with user specifications (90+ punishing hitter, etc.)

[Vince]

My problem with needing a scout to 'look' at a player before you can tell how good he is, is when you turn on SportsCenter. In the world of US pro sports, just about everyone has insane media coverage, and in the games of Baseball and Football, statistical analysis is so much more than in Hockey or Soccer that you can get a much better feel for how good a player is without ever physically seeing him. I'd think that if I were to have to 'rate' the current players in the NFL, I could probably do so on a scale of 0-100 and easily get within 25% of their 'real' rating just based upon my viewing of football and analysis of their stats.

This is where I think the idea posited earlier about not needing scouting for established players comes in. I don't think it should be gone completely (look at Stephen Davis, Tommy Maddox, etc), but it should be much less variable. The real impact of scouts is on players in their first few years, and right before they are drafted.

[QuikSand]

A discussion of scouting in text sims really isn't at all complete without some mention of Mac Howard's game Sick As A Parrott (http://www.users.bigpond.com/machoward/demo.html). The biggest difference betwee his approach and the "everybody else" approach is Mac's refusal to let things boil down into numbers. He uses estensive comparisons and sortings in his scoutings, but you never see a player boiled down into a set of numbers.

It's not for everyone, but it's a different approach (though it is "soccer") that merits attention -- certainly in a discussion like this one.

[KWhit]

I agree with the suggestions that try to make all scouts "equal" in a sense. In other words all scouts are able to see through th efog of war equally, but as a GM we have to strategically tell the scout where to look.

So use some sort of point system giving all teams an equal footing, but make them have to choose where they want to focus their scout's attention and spend their points there.

Basically, what I'm thinking is that on a player screen, I can view preliminary ratings for any player or prospect in the league. This represents the "consensus" opinion of the players skills. If I allocate some of my scouting staff's time to look more closely at a specific player (or all QBs available in the draft, for example), then I get a more accurate read on the player's skills.

Once I sign a player then I begin to get more accurate readings as the practise and the longer the stay with my team the more accurate my view of him becomes until I get to know him better than any other GM. If I trade him, I still keep that accurate view, but I may not be able to see all of his ratings jumps or drops (but I will still be in better shape than a GM who has never worked with the player).

Another thought...
You could make it even more strategic if you gave a team some sort of bonus budget to allocate to different areas. And you have two choices of how to handle this. You could give all teams an equal budget or you could customize your league and have one team like the Yankees who get a much larger bonus budget to simulate their added spending.

For baseball, you could spend your bonus dollars on any of the following:

Scouting - gives more scouting points as described above
Farm System - Adds a slight development boost to players in the minors
Marketing - Improves attendance for the year
Team Facilities - Makes free agents slightly more easy to sign
Etc...
The more I think about it the more I like this idea. What would be pretty cool is that even CPU controlled teams would develop a personality of their own based upon where they put their focus. This is one of the problems with sports sims in general for me. All the CPU teams seem the same.

[Bonegavel]

[edit - the charts i tried to make weren't cooperating both under pure Courier or use of code tags.]

KWIT, excellent way to put it. And, I think that is what TPF is trying to do with their budget system. Though, I have to say that the TPF method is a bit too abstract for me and I don't ever really know what my money is doing for me.

In addition, I think that is great for evaluating current potential. As far as future, that is more of an art/voodoo thing. I think what would help this along is a separate screen that tracked all players on a team and simply showed their development from year to year. At this point, you could keep all the categories and have a relative scale of descriptors (relative only to that player) and the scout could note in each category that the player has:

1) greatly declined
2) declined
3) neutral
4) improved
5) greatly improved

so you might get a chart like this:

NAME 2000 2001 2002 2003 2004
-----------------------------------------------------------------
J. Smith drafted IMP GIMP NEUT IMP
E. Jones NEUT NEUT DECL IMP IMP
R. Stain -- drafted NEUT GIMP GIMP
K. Diehl DEC GDEC NEUT IMP DEC

or, do it in number format for easier viewing

NAME 2000 2001 2002 2003 2004
-----------------------------------------------------------------
J. Smith D 4 5 3 4
E. Jones 3 3 2 4 4
R. Stain - D 3 5 5
K. Diehl 2 1 3 4 2

D drafted
- not in league

Then, at the end of the column you could have a trend indicator that is the mean, or have some sort of weight system where more recent years count for more or whatever. But, this chart only grades the player against himself. You can track him over his lifetime with your team.

Also, you can only track players that are on your team or that you have flagged to "watch" and you can only "watch" X number of players per year. This number could be derived by coach experience, set by the game, whatever.

Now, this number doesn't mean much in comparing player A vs player B, but you now have extra info with which to make a decision on a player when it comes time to contracts. You will still have to look at stats, you will still have to look at 40 times, you will still have to see how he does in camp.

In addition to this, I would like to see a position by position chart of players and how they are ranked on your team. E.g., you have the following player at WR:

J. Smith
E. Jones
R. Stain
K. Diehl

Without you having to go crazy and look at unrealistic number based attributes, any coach (or fan) will know which players are better than others. So, you would get this report:

J. Smith #1
E. Jones #2
K. Diehl #3
R. Stain #4

This would be their OVERALL rating by your staff. I'm sure anyone could go through their favorite team and make a relatively bang-on chart of the players and their rankings on the team by position.

This saves you the trouble of figuring which player is which on the team. If you had the above situation over 4 years and it stayed the same, you know that Smith is your man and you didn't have to look at the stat page, because that could be deceiving on its own. Just think of all the times the top receiver on the team doesn't get many catches because he is double-covered week in and week out, but you still know he is the MAN. Now, if the top guy is also the top stat-guy, that would mean even more come contract time for him.

Now, couple the Ranking List with the Progress Matrix and you now have a decent guide with which to make choices. You could have a Jerry Rice that has 5 years straight Decline on the progress matrix, but he remains #1 in his position either means that you have a bunch of really horrible guys at the position, or he was just that good to start and they didn't hurt him so much.

Then, if that could all be coupled with a League wide ranking list (not just relying on stats, but sort of a Who's Who of the league) you could really see where your players fit in with the whole universe. Rookie to 3/4 years would have a hard time fitting on that list with exceptions being super BOOMs that climb the list regardless of experience.

Still thinking out loud... thoughts?