Scouting & Drafting

[Alessandro]

By the way I work as a scout too...maybe in the future we might discuss something interesting regarding scouting tasks and so on...

[Ogilthorpe]

Along those lines, something to think about is how OOTP (Out of the Park Baseball) handles scouting. In that, you hire a scouting director and then set a scouting budget. The quality of your scouting reports is based on the director's skill and the amount of money you spend.

So you are able to get routine scouting reports on the next opponent as well as on upcoming drafts without having to continually assign and re-assign guys. I kind of like it.

That has to be pretty close to how real teams handle scouting, at least it sounds realistic and I think it's a great idea.

For those of you who play OOTP.....how would you compare that type of scouting system to that of the NHL EHM series?
Is there anything that OOTP can't do?

[empach]

Along those lines, something to think about is how OOTP (Out of the Park Baseball) handles scouting. In that, you hire a scouting director and then set a scouting budget. The quality of your scouting reports is based on the director's skill and the amount of money you spend.

So you are able to get routine scouting reports on the next opponent as well as on upcoming drafts without having to continually assign and re-assign guys. I kind of like it.

I kind of like this idea too, but I think it would need a few options to concentrate on given areas. Say I want to look at older Europeans before the off season for potential signings. Or the option to scout specific players.
I think combining this director idea with many of the options we have in EHM might work pretty well.

[laskey 16]

I agree, ultimately the GM should be able to say "hey, I want you to look for this type of player specifically/give me a report on this guy". The scouting director thing could be very useful though... you could maybe ask him to "prepare for the Entry Draft" and then it will automatically set all your amateur scouts to 'draft mode' scouting assignments.

[alexob18]

Yes, I like the idea of being able to draft by need. There should be more of a boom/bust type of thing, where a guy can look decent and have amazing potential, but can be seen as a high risk guy to pick, like Pulkkinen. Has talent, awesome potential, but could be a major bust too.

[ArtVandelay]

Yes, I like the idea of being able to draft by need. There should be more of a boom/bust type of thing, where a guy can look decent and have amazing potential, but can be seen as a high risk guy to pick, like Pulkkinen. Has talent, awesome potential, but could be a major bust too.

I think what makes these guys bust are:

Lack of consistency
Inability to adjust to NHL speed as a result of lacking instincts
Not working hard enough to adjust to NHL

A good scout might help predict those things. An average scout could miss them. A bad scout will be like, "he's got mad skeelz - draft him".

[laskey 16]

Add to that list:

Lack of size or failure to grow into a lanky frame
Inability/unwillingness to adjust to North American lifestyle off the ice
Bad attitude or discipline

This is in real life, I'm not saying all of these could/should be in the game. But maybe for these things the combine interviews could let scouts 'find' these problems.

[alexob18]

Add to that list:

Lack of size or failure to grow into a lanky frameInability/unwillingness to adjust to North American lifestyle off the ice
Bad attitude or discipline

This is in real life, I'm not saying all of these could/should be in the game. But maybe for these things the combine interviews could let scouts 'find' these problems.

I really like these ideas. Especially the size one. Make size relevant in this game, as it was somewhat useless in EHM (I've been told it does effect something, I just don't know what).

[ArtVandelay]

Add to that list:

Lack of size or failure to grow into a lanky frame
Inability/unwillingness to adjust to North American lifestyle off the ice
Bad attitude or discipline

This is in real life, I'm not saying all of these could/should be in the game. But maybe for these things the combine interviews could let scouts 'find' these problems.

Agreed

[laskey 16]

Thanks guys. Yeah I wonder if maybe strength attributes could be relative to size. So if a player who is 5 foot 8 has an 18 in strength and a 6 foot 5 player has the same attribute, the bigger player is actually still stronger. I don't know if that's possible to code, or if it's even a good way to achieve it, but size should play a bigger role than it did in EHM I think.

[CeeBee]

Add to that list:

Lack of size or failure to grow into a lanky frame
Inability/unwillingness to adjust to North American lifestyle off the ice
Bad attitude or discipline

This is in real life, I'm not saying all of these could/should be in the game. But maybe for these things the combine interviews could let scouts 'find' these problems.

Also maybe being a more coachable player would have a better chance to develop and a less coachable player might have a lesser chance

[bruins72]

Good idea, CeeBee!

Also, I would like to see a hidden attribute called "development". This could help determine how quickly a player reaches his potential. If it's 10, it's a normal development timeline. If it's less than 10, it's slower. So a player with a 5 in development might be a late bloomer. A player with a number higher than 10 would have an accelerated development. That doesn't mean that this player is going to be a star though. It could mean that some players peak at 16 or 17 years old while playing in Junior and never get any better. But you could have a star like Crosby who is "the man" right away when he enters the NHL and he could have a development rating of something like 17 or so. With enough scouting, your scouts could give you an idea of what his development speed might be like.

[dabo]

Good idea, CeeBee!

Also, I would like to see a hidden attribute called "development". This could help determine how quickly a player reaches his potential. If it's 10, it's a normal development timeline. If it's less than 10, it's slower. So a player with a 5 in development might be a late bloomer. A player with a number higher than 10 would have an accelerated development. That doesn't mean that this player is going to be a star though. It could mean that some players peak at 16 or 17 years old while playing in Junior and never get any better. But you could have a star like Crosby who is "the man" right away when he enters the NHL and he could have a development rating of something like 17 or so. With enough scouting, your scouts could give you an idea of what his development speed might be like.

This is definitely something I will look to implement, if done properly, this could be one of the nicest features I think.

As you say, on one end we have Crosby and on the other players like St. Louis/Zetterberg (perhaps there are better examples?)

[philou21]

That would be really cool Bruins good idea!

[laskey 16]

Great idea! Scouting reports could hint at this too - "player has a lot of room for improvement" meaning that they could be a late bloomer, and "player is almost at his most NHL-ready" meaning the player is a fast developer.

[Canuckk]

MOD NOTE: I moved this to the Scouting and Drafting thread because it belonged here more than the Development thread. Also moved Empach's reply to this. - B72

I think rather then to rely more on a "guessing game" of sorts for the draft with the scouts giving reports that could be inaccurate and so on, it would be better to simply have a broader, more general rating the less a player is scouted. So if a poor scout scouts a player for a short amount of time, the ratings should be very broad, like on a 1-5 scale or even just a basic "his shot is good, his skating needs work" kind of report. But if a very good scout scouts a player for a short time, the scale might become 1-10, to make things more accurate, and the longer the player is scouted for, the more detailed the scale becomes, like 1-20, and begins to reveal more about the individual attributes rather then the general categories like skating, scoring, physical ability, etc. Of course, there should still be some inaccuracies with the scouts, but this should have more to do with the ability of the scout then anything else, and shouldn't range terribly - even the worst of scouts can tell if a player can't skate or is unwilling to get involved physically!

So just as a few examples, here's what I would see as a solid implementation of this idea. Say we're scouting player X, who's a scoring RW with decent size, a great shot and explosive skating ability. The scout reports might look like this:

Poor scout, after a short scouting session:

Player X is a goal-scoring winger with a good shot, good skating, OK size and great puckhandling skills. Needs to work on defense and using body effectively.

Good scout, after a short scouting session:

Player X is a goal-scoring winger whose focus is more around offense but is a good talent. Here is an evaluation of his skills based on my session:

Skating: 4/5
Scoring: 4/5
Puckhandling: 5/5
Physical Abilities: 2/5
Defense: 2/5
Mental: 3/5

Good scout after a longer scouting session:

Player X is a goal-scoring winger with great offensive talent but has trouble defensively and apparently doesn't get along well with his teammates. Here's a report based on my session:

Skating: 8/10
-Speed: 9/10
-Acceleration: 7/10
-Agility: 7/10
Scoring: 9/10
-Wrist shot: 10/10
-Slap shot: 6/10
-Offensive instincts: 8/10
Puckhandling: 10/10
-Stickhandling: 10/10
-Puck control: 8/10
-Deking: 10/10
-Flair: 10/10
Physical Abilities: 5/10
-Checking: 4/10
-Size: 6/10
-Strength: 5/10
-Hitting: 4/10
Defense: 4/10
-Pokecheck: 5/10
-Defensive instincts: 3/10
-Positioning: 4/10
Mental: 3/10
-Teamwork: 2/10
-Aggresiveness: 5/10
-Determination: 4/10
-Work ethic: 2/10
etc.

Obviously this is just an example but hopefully you get the idea. I think misleading people too much with inaccurate scouting reports shouldn't be the only method used to make the draft more of a rubbish-shoot then in EHM. The EA NHL games use something like this as well - give vague ratings from scouts that can be inaccurate. A combination of the two would probably work well I think - just be sure not to overdo the inaccuracy of scouting reports, especially on things like skating, physical abilities and so on that are pretty easy to judge.

The thing that probably should be key in a scouting report to ensure that the attributes given are not inaccurate is how long they scout for, now that I think about it. Of course, with a short scouting session it's tough to tell if a player was just hot for one session or if that's really his talent level. I suppose the longer you scout a player, the more you correct for the times when he is hot and cold and come up with a good estimate of his potential. Though I think when real NHL scouts describe a player's potential, they base it a lot on the "flashes of brilliance" a player can show at times when being scouted. Or at least, that is the case for flashy offensive forwards. For defensemen, it's judged more on consistency in not making mistakes and so on.

Sorry, I'm rambling on a bit, but yeah, the quality of scout and the length of the scouting session should probably be the biggest two factors in determining a player's abilities from scouting reports. This should really only apply to prospects though, as established NHL players should generally be pretty well-known and therefore should have most of their attributes already visible. As suggested above, reputation would work brilliantly in ensuring the lesser-known players are wild cards if they aren't scouted sufficiently, whereas with the well-known players it should be easy to judge their ability fairly accurately due to all the media coverage and scouting reports from organizations like the ISS and NHL Central Scouting.

[empach]

Pretty good thoughts Canuckk. I think one thing we should look at for scouting is the common causes of players not working out and how they could be implemented in the scouting engine.

Over in the Scouting & Drafting thread they discuss some of the things that affect players not reaching their potential. I think we should make those aspects the hardest to get an accurate scouting on.
http://www.ehmtheblueline.com/forums/vi ... c&start=25

[Alessandro]

Well, but being NHL ready doesn't mean having less potential

[Alessandro]

About scouting. I was wondering if maybe after a first quick scouting one could have a report similar to my site:
http://www.russianprospect.com/vladimir-tarasenko
goalie: http://www.russianprospect.com/alexander-pechurski
with ranks and graphic i mean.
It looks nice hehe

[dabo]

I like it, looks pretty.

I could use the categories in the radar chart and put them in a horizontal bar chart (or whatever they are called).

[bruins72]

The thing that probably should be key in a scouting report to ensure that the attributes given are not inaccurate is how long they scout for, now that I think about it. Of course, with a short scouting session it's tough to tell if a player was just hot for one session or if that's really his talent level. I suppose the longer you scout a player, the more you correct for the times when he is hot and cold and come up with a good estimate of his potential. Though I think when real NHL scouts describe a player's potential, they base it a lot on the "flashes of brilliance" a player can show at times when being scouted. Or at least, that is the case for flashy offensive forwards. For defensemen, it's judged more on consistency in not making mistakes and so on.

Interesting ideas. I snipped the rest of your post because I mostly wanted to address this bit I kept quoted.

This is an important part of scouting, I think. If your scout watches a player for just one match, he could be playing the game of his life and get too high of a rating. In the same way, he could be having an off night and be judged as much worse than he really is. Now if that same scout watches the same player for 5 or 6 games, he'll get a much better idea of what the player is normally like. The more he watches a player, the more details he might reveal.

[bruins72]

I just wanted to expand on what I was talking about above. I've got some ideas for scouting. I'm not sure how feasible it is from a programming perspective but I think it could be interesting. This is working off of Canuckk's idea somewhat. Scouting reports would have different levels of detail and accuracy. You would also be able to assign your scouts in different ways. You would NOT be able to just tell them to "Scout NHL Draft". That's just too broad of a range of players to assign 1 scout to because there are so many different leagues all over the world that have players in the draft. Here is what I envision...

Let's assume you're scouting for the real life NHL 2010 draft.

Assign scout to:

Scout league for 1 months: So let's use the OHL as an example. Your scout will watch the OHL for a month. He'll plan out a schedule so that he gets to see every team play a somewhat equal amount of games in that month. He'll get a very rough idea of the players in the league. Think of it as a "getting to know you" scouting trip. He probably won't see any team more than 3 times during this month, so he won't get a very accurate read on the players but he might spot a few players that he thinks merit more interest in another trip. He'll get that low detail/inaccurate report, adjusted by the scout's skill.

Scout league for 3 months: On this type of a trip he'll probably see each OHL team 9 or 10 times. The scout's accuracy will be improved somewhat but because he's looking at the whole league in general he's not going to give you very detailed reports on each player. He'll give you a low detail/moderate accuracy report, also adjusted by the scout's skill.

Scout league for entire season and playoffs: By doing this, you're giving a scout the of just watching this OHL. That's all he does. Because he's watching everyone all season long, after the playoffs end he'll give you a moderate detail/best accuracy report, heavily adjusted by the scout's skill. Also included in this report might be clues to whether or not the player steps up his game for the playoffs or disappears. This type of scout could be very important! Many people might want to have a scout for each of the Canadian Junior leagues (OHL, QMJHL, and WHL) especially. During the course of the season, your scout will recommend certain players that you take a closer look at. You can then target those players specifically (more on that later).

Scout team for 2 weeks: On this trip your scout will watch one team, let's say the Windsor Spitfires, for a couple weeks. So during this time he might watch the team 6 or 7 times. He'll have a decent read on the team. He might be able to give a moderate detail/moderate accuracy report, accuracy adjusted by the scout's skill as usual. Of course, if a particular player is hot or cold during this 2 week stretch, it could skew his report.

Scout team for 1 month: On this trip your scout will watch this team quite a bit. The accuracy of his reports will improve because over the course of a month he'll get a much better idea of what each player's normal game is like, with a few highs and lows mixed in. Still, I wouldn't say that the individual player reports would be that detailed because he's still watching the entire team. He'll give you a moderate detail/better accuracy report, adjusted by the scout's skill.

Scout team for 2 months: This trip is the same as the above but the accuracy is even better. He'll give you a moderate detail/best accuracy report, adjusted by the scout's skill. This might be a good option to use on teams with a lot of draft eligible players. If you're watching the Spitfires, it would give you good reports on Taylor Hall, Cam Fowler, and Justin Shugg.

Scout player for 1 week: Use this trip if you want quick look at a specific player. You'll get a moderate detail/low accuracy report, adjusted by the scout's skill. You see that Cam Fowler is one of the top rated defensemen in the draft, so you want a quick look at him to see what kind of defenseman he is.

Scout player for 2 weeks: After 2 weeks of watching just one player, you'll get a much better idea of what he's like. He'll give you a better detail/moderate accuracy report, adjusted by the scout's skill. He might also give you basic low detail/low accuracy reports on other players who played in those games that stood out (both positively and negatively). Of course, you can't always trust these reports because they're probably players that had a good or bad game.

Scout player for 1 month: Same as above but you're watching this player for a whole month. Your scout would give you a best detail/best accuracy report, adjusted by the scout's skill. Use this type of assignment for the guys you might be targeting with your 1st round pick. While on this assignment, your scout might offer reports on other players. Reports on players on other teams would be low detail/low accuracy. The reports on players that were on the same team as the targeted player would be moderate detail/moderate accuracy because you're seeing these players every game and your targeted player isn't on the ice all the time. So with this report, you've targeted Taylor Hall and you've got an excellent read on his. You've also got a pretty good read on Cam Fowler and Justin Shugg, among others.

So this all sounds a little complicated and I don't know if it would be a nightmare to code. What I envision is having these scouting reports attached to the player for that season. Once July 1st hits, the slate is wiped clean (maybe keeping the most basic details such as "Offensive Defenseman" and "Scoring Winger") and you'll need to scout them again. This method would really make you think about your "shortlist" and the players you have on your draft board. If you've got the #25 overall pick, you're not going to waste your time looking closely at the top 5 guys unless you're thinking about trading up. You're going to have a handful of guys that you think might be available at #25 and maybe a couple you hope might slide.

[dabo]

Thanks for your input that certainly sounds good. I guess a scout could only be assigned one task at a time?

[BTerran]

Just to throw this out there as my two cents on scouting...in games it's always felt like busy work to me. It's just pointing your scouts to where you want them to go and then x number of turns later, you have to do it again. Most games don't alert you when a scout is free so I have to keep checking...

Something I love about Civilization V is, even though it's full of busy work, it holds your hand through it. Before you complete a turn it takes you to every city that's completed production and asks you what you want the city to do now. I'd love the same thing for scouting. When you click to finish a turn, it first takes you to a scouts page who is unassigned so you can assign him to something else. Of course in an ideal world, I'd just like to assign the scouts to auto so I never have to deal with them. There's plenty of aspects of sports management games I love, but scouting isn't one of them. I just don't see the skill or strategy in it, it feels like - in most sports games - it exists solely because real GMs have to manage their scouts so I guess we should too.

Only, I think real GMs mostly determine how many scouts they hire and where they're assigned too. I don't know that a guy really sends a scout to Russia and then the States. I think he decides, I have X amount of dollars to spend on scouts, so I'm going to spend X amount on three Russian scouts and they're always going to scout Russian talent and occasionally they'll send me worthwhile information. Should the team's circumstances change, I might reallocate the budget or I move a scout to a different region, but those are special circumstances, not something that I'd have to do every few weeks or months.

[bruins72]

Thanks for your input that certainly sounds good. I guess a scout could only be assigned one task at a time?

I'm just glad you're attempting to make this game and taking our input.

Yes, a scout could only be assigned to one task at a time. So if you've got 10 scouts, you've got to think about how you want to assign them. You get a general idea of what a player is like from a superficial scouting trip of some type and then you have to target specific players that you are interested in. I'm sure the Oilers didn't spend much time looking at McIllrath or Campbell. They surely spent a lot of time watching Hall and Seguin though. They might even have spent a little time watching a couple of those defensemen but maybe not.