Archived

This topic is now archived and is closed to further replies.

Sports RPG?

This topic is 5638 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I''ve wanted to write a sports game (specifically a basketball sim) for a long, long time. I actually advanced somewhat on a DOS-based incarnation around 1996/97, but that''s all history now. I used the term "sim" loosely then as well; now I''m asking what your reactions would be to an actual sports simulation - not a management sim, and not an arcade "sim". Sort of a Sports RPG. At the most basic level would be a "generic" sports simulation that allowed you to play the game as various characters or controlling an entire team, much like what we have today. Of course I''d make improvements in control and AI, but the real revolution would be in the RPG element. You''d be able to play through a professional athlete''s career, starting from draft day. The draft could be randomly generated, possibly even resulting in your going undrafted and having to play in NFL Europe for the Rhine Fire or for FIBA teams in Turkey. (Please note that this is representative of the American professional sports climate; the rest of the world "drafts" its pro players from development leagues and junior clubsides). Once drafted, you''d need to perform on a regular basis to earn your "minutes" and gain public recognition. As your name brand value rises, you''ll be approached with shoe and product endorsements and the like. You''ll eventually become a free agent and need to negotiate contracts with teams, ultimately seeking to position yourself with a team that has real chances of "winning it all" (it''s all about the hardware, baby!) You won''t be required to be the team leader, but can play a role (be like Derek Fisher or Robert Horry instead of Mike). You''ll also contribute to the team dynamic through your post-game comments and statements in huddles or during time-outs. As far as actually "playing" the games, you can either jump right in and control the action (but only as one player - like current "player lock" features) or leave your character to respond according to his/her skill levels. Skill can be improved by training - gym workouts, aerobics for flexibility, shooting/batting/kicking practice, running/bikes for cardio, etc - and reflexes improved through timing sub-games. Perhaps you should also be able to improve your character''s work ethic, which would cause him/her to work out independently (leaving you to focus on more important things). Inspired by a column in July''s PC Gamer, I''ve considered it worthwhile to include an MMO aspect of the game, allowing several individuals to play as teams - coaches and players - for playoff positions and championships. You can also take your developed player online for "streetball" games (roller hockey, anyone?), one-on-one challenges (winner holds the court), half-court 3-on-3s (basketball), etc. Eventually, you''ll grow too old to run the floor/grass/pitch/diamond and retire, but then you can become an assistant coach/coach/GM (depending on your attributes while you were a player) and manage or direct the team, calling plays, negotiating and trading for players, working around salary cap limitations, etc. You''d need to hire scouts and scan reports for information on emerging talents, and keep your eyes peeled for opportunities to invite them to join your organization. In the MMO version, these may be actual, real people on the other side of the globe. This is where your player''s work ethic and skill levels become important: if there''s a scheduled game that you (the gamer) can''t make, your avatar will still be able to play at a level commensurate to the time and effort you''ve put into the game. In reality this is like blending two games into one (Fast Break Basketball with NBA2K1 , for example, or Championship Manager with FIFA 2001 ), with a little bit of something extra. What do you think? Overambitious? Infeasible? Overly broad? All critiques welcome.

Share this post


Link to post
Share on other sites
I like it.

I''ve been wondering why there still is no real MMO sportsgame out there where players can compete against eachother while building a career.

Seems to me that there are enough players out there who take pleasure in proving themselves better than others. Why not create a real competition for them?

Share this post


Link to post
Share on other sites
I think the reason why has been answered by the sheer lack of replies to this suggestion. Sports are fairly big and popular, but nowhere near the gaming market prowess of RPGs (can you actually count the number of RPGs released in the last 2-3 years? Including expansion packs and add-ons [but not patches]) and FPSes (even worse). Publishers are so busy pushing "quick buck" titles - some of them remarkably good, but blaze trailers nevertheless - out the door that a game like this would have to be independently developed, but would probably end up a major sleeper hit.

Well, I've got my work cut out for me. If anyone ever has any suggestions or comments, let me know.

[Edit: outdated email address.]

[edited by - Oluseyi on April 17, 2002 6:27:14 PM]

Share this post


Link to post
Share on other sites
Oluseyi, I really like the idea even though I''m not a sports fan. Maybe something like this would get even ME to play sports.

I lived with two roomates who were both RPG players and sports nuts. I don''t imagine that you wouldn''t have an audience, though the cross section overlapping both RPG and sports players would probably automatically make it smaller. I''d imagine you''d have more success as a console game, as that seems to be the widest fan base for sports titles.

The MMO aspect sounds especially promising. Again, I don''t know much about sports, but this would seem to be a potentially lucrative area. (Maybe you''d tap into all those fantasy baseball / basketball players out there...?)



As for the lack of replies... well, not to rib anyone on the board, but perhaps if you added a medieval component? Say, undead, magick, and lots of characters whose names ended in -ia or -undrel?

Then you''ll get six pages of replies and lots of discussion, I promise (tongue firmly in cheek, that is).

--------------------
Just waiting for the mothership...

Share this post


Link to post
Share on other sites
I like the idea, but how would you handle "bench-time"? Say your character sucks, you won''t get a whole lot of minutes. So will your character just sit there or will the game automatically cut to the times(if any) your player is on the floor?



--I don''t judge, I just observe

Stuck in the Bush''s, Florida

Share this post


Link to post
Share on other sites
Have the ability to skip to the next time the player is active, but if nothing is pressed, play through computer vs. computer.

Share this post


Link to post
Share on other sites
quote:
Original post by Dynamite
I like the idea, but how would you handle "bench-time"? Say your character sucks, you won''t get a whole lot of minutes. So will your character just sit there or will the game automatically cut to the times(if any) your player is on the floor?


As Thrump said, there would be an option to either watch the game even though your character was out of it or jump ahead to the next time you got playing time (imagine if that was 3 games later!)

I think that players who aren''t terribly skilled or knowledgeable about these sports would opt to use preconfigured, relatively "good" characters until they get a feeling for the game. What that means is that in developing your character, you can choose to go with an already "developed" player, but you wont be able to reach the same pinnacles of achievement as if you developed him from scratch (he may have picked up some bad habits that will be almost impossible to overcome, etc. Besides, there needs to be an incentive to develop a "fresh" character and a potential reward for people who do). Also, the control system will be worked out to give good variability - accomodating players of different skill levels (remember that when we talk of sports games the player is the person playing the game, aka the gamer, while the avatar is the "character" or "athlete").

The control system I''ve been thinking of has a huge number of toggles and switches that can be individually enabled or disabled to allow the player control different aspects of the game. So to immediately be competitive even though you don''t know the sport, set most things to automatic (this comes in handy when you can''t be around, because essentially everything is set to automatic - the framework is already there). The balance in this is that it uses semi-deterministic AI routines, and as such can never match up to the potential level of a human player. Say you set shot selection and shot release to auto, and your athlete is defined as "explosive" and "aggressive". In a basetball game, such a player might try to dunk over opponents (and possibly commit a charge) when the "smart" thing to do is to "stop n'' pop". Or the athlete may only be able to release on hs jumper at the peak of his jump, whereas a human opponent can release at any time - off balance, falling into the crowd for the Jordan-esque fadeaway...

These are some of the balance issues that I''ve considered to give both sporting "newbies" and hardcore tele-athletes roughly equal advantage. I hadn''t been checking this thread since I only got two replies after the first week, so I''m sorry for the delay. Thanks, guys. I see that some people''s interest is at least piqued.

Wavinator: Actually, I''ve been considering the console space essential to the development of an audience for this game, but wondering how to incorporate all the RPG elements (and control variations). As for names ending with -ia, we need only look to Turkish soccer . Besides, how many gmaes have a character named Dikembe Mutombo Mpolondo Mukamba Jean Jacque Wamutombo (Philadelphia 76ers)?

Share this post


Link to post
Share on other sites
If we DO opt for an online sports game with rpg elements... I''d say let''s just go for it: create an online league.

You''d have your ''rookies of rookies'' league where all players have to start their new characters. When they start winning games, they''ll be able to move up on the league ladder and enter tougher competitions.

There''d have to be a feedback system in place to leave feedback to other players. Say you play with 8 players (rotating) per team (basketball). After the game, win or lose, you''d leave feedback on the players you particularly liked or disliked.

Of course the goal of the game could be to
a) enter the top of the leagues where with victory comes ultimate glory
b) become one of the best liked players
c) become one of the best skilled players
d) just have fun competing
e) create the best team
f) create the best duo
g) etc

I think with sports comes competition. So you''d have to give the players a place where they can really compete. With that, comes the absolute necessarity of disabling any options of cheating. Which would mean a lot of server side information

How to prevent other players from ignoring you during play (selfish bastards!)? Those players that will not pass the ball even if you''re a better shooter than they are, even if you''re wide open, even if they''re covered by three players...

Feedback.

Players that continue to play selfish will get a very low team-rating. Other players will not want to include them in their teams.

About feedback. I''ve always had big problems with the way Ebay''s feedback system works (my wife does a lot of Ebay shopping). It''s just a ''I leave negative feedback on you, so you leave negative feedback on me'' type of feedback. It really doesn''t mean anything.

I think that to establish an effective feedback system (which I really think is necessary to make online sports fun) you HAVE to create a system like the following:

If I leave POSITIVE feedback on another player, that pretty much means I accept his views and style. So whoever HE leaves positive feedback to, should also be on my positive list. If he leaves negative feedback to someone, that someone should show up with a negative number on my list.

It''ll work somewhat like the old ''seven degrees of separation'' idea. Each player that play, will somehow be linked to you.

All the system would have to do is show you the shortest route of positive/negative feedback to the player you''re investigating.

Example:

You''re trying to see if you want player X on your team.
You look at feedback and see
Player X-->positive feedback about player X by Player Y-->positive feedback about player Y by player Z-->positive feedback about player Z by self.

X-->+ Y-->+ Z-->+ Self

You can make the system work even better by assigning ''trust'' values.

If you ABSOLUTELY trust player Z (maybe he''s a close friend) you assign the number 10 to him. If you somewhat trust player A (someone you played with yesterday), you would assign number 1 to him.

Now, when you ask for feedback on player X, the system will find the shortest route, with the highest total ''trust'' value.

Sounds a little complicated, and maybe this is thinking too far ahead (let''s first design this thing) but I just think that a feedback system is needed to make players play together. Otherwise you''d just end up with a bunch of selfish ''I''m the best and I''ll prove it'' players. Nothing wrong with that, but the very reason why I want to play online with others is that I want to play as a team and compete... I don''t want to compete WITH my teammembers.

Share this post


Link to post
Share on other sites
This feedback element is very interesting, and seemingly quite unique. I do see some shortcomings, though. The ''chain'' of feedback means every single player on a server needs to have a rating (or memory allocated to hold a rating) on every other player, resulting in an N * (N - 1) table of feedback values which must be constantly updated and rapidly searched. the problem it attempts to address, however, is a very real and legitimate one (heck, it turns up in real life pickup games!). Here''s my original take on it:

Players may choose to play pickup games, sign up to teams of people they don''t know and have never met, or collectively form a team (with friends). As in real life pickup games, you need to play with people or observe them play to get a feel for their game - how selfish they are, their skill level, their court awareness and so on.

(On a self-actualizing side note, all the cats I''ve been balling with recently seem to lack court vision; I always pass out of double teams and set them up for wide open Js, wrapping the ball around defenders with, frankly, amazing skill to find my man. They wont even feed me in a post! And they never look to see me wide open, unless I scream for it - I find them, even off the ball).

After observing, one may choose to play or not to play (playground rules apply in pickup games: someone has "next" and gets to pick his team for the next run out of the available guys around when he becomes next. He can''t choose players who have played since others who haven''t played arrived). It would then be recommended that one not elect to join a team until one had either observed their game (there will be spectator capability), played with them or gotten to know them in real life (the friends scenario). The other alternative would be to be "drafted" to any team that is looking for a player with your skills (there will always be at least one computer-controlled team to ensure everyone who wants to play, can. Such a team may not make it to the playoffs/tournament/finals, but neither will everybody else). In tournament/organized games - ie, non-pickup - players will be rated on their performances automatically and this data made available to coaches. A smart (human) coach will bench a player who''s too selfish with the rock, or may encourage another to go to the hole more. A computer coach will automatically bench a selfish player - possibly pulling him out of the game - and issue "standard" encouragements/reprimands, etc.

This brings me back to the question of what happens when your player (say [s]he sucks) is benched. If you''re playing conputer-controlled opponents, the game can be simulated in fast forward till your next insertion into the lineup. If you''re playing online with other humans, however, you''ll have to sit out - which should provide incentive to improve your personal performance and/or your relationships with teammates/coach. Team sizes will be limited to the current minimum (the real-world minimum will be our maximum) of 8 (for basketball), otherwise too many players may not see any game. Maybe I shouldn''t actually limit, but rather should recommend no more than 8.

This all introduces very interesting social interaction aspects. A team manager (a possible role) may spend time observing playground pickup games and inviting players to "tryout" for or join their team (it would be cool if managers could create "highlight reels" of their team in action to sho prospective signers). This would introduce the hiring, firing and trading aspect - and possible tensions between real people. I guess managers should be able to keep computer players as reserves on the roster, in case a real-world member has a season-ending injury. All this means that it would be possible to play the game from a number of different perspectives, with varying amounts of necessitated committment - manager (you could build a team from scratch, recruiting players and juggling egos), coach (develop a playbook, fulfill your long-time desire to lead your boys and "win it all", deal with team discipline, micromanage the game down to your team''s use of the clock) or player. And, of course, you could just watch.

I think these are more than enough aspects already to consider. I''ve (finally!) received some excellent feedback and contributions. Thanks - keep ''em coming!

Share this post


Link to post
Share on other sites
I don''t know how well it''ll work online because then you will need all the players online to have a game, or forfit if there are not enough players. And what about disconnects? It''s an excellent idea but there are too many "must haves" just to play 1 game. Streetball online may work because the stakes aren''t so high if someone is shooting and gets cut off. I do think it can work single player though. The first game I ever made was a basketball game and I considered this too, but a little different. I didn''t really stick with the idea, but you make an excellent argument.



--I don''t judge, I just observe

Stuck in the Bush''s, Florida

Share this post


Link to post
Share on other sites
Actually, there is a mechanism for handling insufficient numbers of players which could be extended to handle disconnects (by considering them temporary lacks of players). Your avatar can be "represented" by AI dependent on the skill levels you have built up so that it plays somewhat commensurate to your real level of play. If there aren''t enough players, computer-controlled teammates and opponents are generated (technically they are "bench" players). If a player gets disconnected, an AI is substituted for his input or his avatar may be subbed out at the next ball out, depending on the coach/coaching parameters.

Most of this has been covered in my earlier (lengthy) posts.

Share this post


Link to post
Share on other sites
Time to awaken the beast.

Based on a suggestion in the MMO as a paradigm? thread and a bunch of recent ideas, suggestions and eavesdroppings, I''d like to propose some changes and see how people react to them.

First, there are now several "levels" of play. The previous posts have dealt with the athletic/sports level of play, and really don''t think there''s any need to change anything. Moving up one we have the ownership level, at which point ticket sales (little sim people from the city will purchase tickets depending on whether they are die-hard fan and/or if the team is doing well), merchandise (ditto), etc will become factors in determining profits. This will then give "owners" incentive to improve their teams (we''ll give them a break by having drafts regularly for the organized leagues).

Just above the ownership level is what I think is the most interesting layer - the trade level. At this point, owners can enter into deals with other owners - even in other sports - to consolidate and co-fund ventures. They can sell out/be bought out, relocate and all the other things that real-world owners can to increase profitability/cut losses.

At heart the idea is to create an MMO infrastructure into which several different types of games can be plugged, with the possibility for the various games to interact given a uniform protocol or a "contractual agreement" between the specific games for a private means of communication. I find this to be a fascinating idea and wonder what severe obstacles stand in the way. I''m not interested in the monthly subscription model, so I won''t have to worry about servers and bandwidth; the software will always come with both client and server (or the server will be downloadable) so the onus for hosting is on the gamers, much like many current FPSes.

Comments, please.

[ GDNet Start Here | GDNet Search Tool | GDNet FAQ | MS RTFM [MSDN] | SGI STL Docs | Google! | Asking Smart Questions | Internet Acronyms ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
Hmm, I don''t know if you could really make it so massive with that idea. I like the idea, but I don''t think it will support more than 300 on a server. (not that that''s necessarily bad, it does support stronger community interaction.) Also, it''s more tournament based, and it doesnt have much of the cool customization things you can do with your characters in MMORPGs, except maybe sneakers. People mostly like to be/look unique. It might be cool if players could have their own homes, and such. It seems kinda like a better candidate for a quake server system though to me.

-=Lohrno

Share this post


Link to post
Share on other sites
quote:
Original post by Lohrno
Hmm, I don''t know if you could really make it so massive with that idea. I like the idea, but I don''t think it will support more than 300 on a server. (not that that''s necessarily bad, it does support stronger community interaction.)

I just described something I''m hoping will develop over the next few years in the MMO-as-a-paridgm thread: distributed processing. Advances in distributed processing could make it possible for several independently owned servers to collectively host a single coherent game world. That would make the above concept possible.

quote:

Also, it''s more tournament based, and it doesnt have much of the cool customization things you can do with your characters in MMORPGs, except maybe sneakers. People mostly like to be/look unique.

Headbands (single? double? criss-crossed?), armbands (and whether to wear them CBA- or NBA-style), wristbands, therapeutic leg socks, armsocks, finger bands (monogrammed), hairstyles, and - the big one - tattoos.

Eat your heart out, RPGs!

quote:

It seems kinda like a better candidate for a quake server system though to me.

Yeah, me too. But it was just an idea.

[ GDNet Start Here | GDNet Search Tool | GDNet FAQ | MS RTFM [MSDN] | SGI STL Docs | Google! | Asking Smart Questions | Internet Acronyms ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
This sounds crazily ambitious. Ambition is good, but only in moderation.

First, if you want people to actually play the sport you need a working sports engine game. That is an *entire* project in itself! People can and so sell games that are just that, while your project would include that just as a small part...

So, you''ve already got a full real game''s amount of work right there. Even if you don''t want to create an engine, if you want the AI players to play a real looking game you will have to end up doing much of the same amount of work. (Go play "smallball" to see how frustrating it is when the game doesn''t play like the real game. In smallball you can train players and they play on their own, problem is the game they play looks very little like actual baseball)

Now, you want to add on playing as a single character, building up stats...ok, that seems reasonable, I could see that fitting into many games. But then you want shoe endorsement, half-time huddles, etc? And you want to be able to manage, be president of operations? And you want MMO? That''s about 3 games in one.

And you really have to do each one well. If you can play as a president or GM or whatever you have to make sure the computer AI for trades and such is good, and that the interface is good. Try playing most sports games in "franchise mode" to see what I mean. Only a purist really cares about franchise modes, but those modes always bite from a purist perspective. On top of that, the actual sports part of the game has got to play like a good sports game.

There are some Japaneses sport-rpg games, I believe what they do is not actually have an actual engine at all. The games are just cut scenes and menus and such. Which makes sense. Otherwise what you are proposing is taking an entire game and adding on a bunch more stuff, which doesn''t seem very feasible.

Share this post


Link to post
Share on other sites
quote:
Original post by AnonPoster
First, if you want people to actually play the sport you need a working sports engine game. That is an *entire* project in itself! People can and so sell games that are just that, while your project would include that just as a small part...

You''re not paying close attention. Observe...

These are supposed to be separate applications - completely independent projects that happen to communicate certain information to each other. What part of the sports game is needed in the ownership simulation? Attendance. Scores and statsheets. Hmm, maybe highligts (stills) if we''re feeling really ambitious. So we''re essentially just sending text transmissions between applications. Level 1.

Level 2: Inter-ownership simulations. Again, it''s all about trades and fiscal transactions - text and figures. Simple. Straightforward. Stuff.

Large? Definitely, but the plan isn''t to release one package containing all three levels. The plan is to design an infrastructure (preferrably extensible) and release the bottom level components first (the sports games). Based on the defined protocol, the ownership games can then be released and "integrate" with the existing sports games. And I don''t plan to do this all alone. An open protocol so that I can do my sports RPG while the author of Fast Break Basketball can integrate his management game with the sports game.

I hope you see what I''m getting at? I''m thinking in totally non-proprietary terms.

quote:

So, you''ve already got a full real game''s amount of work right there. Even if you don''t want to create an engine, if you want the AI players to play a real looking game you will have to end up doing much of the same amount of work. (Go play "smallball" to see how frustrating it is when the game doesn''t play like the real game. In smallball you can train players and they play on their own, problem is the game they play looks very little like actual baseball).

I blame the programmers and the gamers. I doubt they know the game through and through. There is NO current commercial sports game from a big name developer/publisher that has a collegiate-level athlete on the development team. Go figure.

[ GDNet Start Here | GDNet Search Tool | GDNet FAQ | MS RTFM [MSDN] | SGI STL Docs | Google! | Asking Smart Questions | Internet Acronyms ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
I like it... it''s kind of like you''re putting an interface layer on top of your game, such that any application could configure and run instances of your game for whatever reason it likes. Then another layer to receive output from the game as results.

One thing I always see people mentioning in amateur game development forums is how cool it would be to multi-tier games. You know, like where some players are the grunts in a FPS, other players are generals controlling them in a whole other interface. I swear, variations on that theme come up ALL the time. But I can''t think of any game that''s ever actually done that! It seems so obvious, it''s strange... I guess the challenge of making not 1 but multiple games is too much at this time.

Share this post


Link to post
Share on other sites
As I said, what you are proposing is the work of 3 games. However if you have the time and are going to have other people do other parts or just release one at a time that could work. In that case, the hard part becomes planning such that the framework works, and that the first releases do the right thing to work with later releases. There is a bit of crystal ball gazing there, although you can always patch.

[about smallball and sims not playing correctly]

quote:

I blame the programmers and the gamers. I doubt they know the game through and through. There is NO current commercial sports game from a big name developer/publisher that has a collegiate-level athlete on the development team. Go figure.



I don''t think that is really the problem. I think the problem is that the problem is just really hard!

I mean, say you want to take the MLB roster and put the teams in a game. You rate players based on speed, strength, hitting ability, whatever. You then need a system so that the players play like their real-life counterparts. (Baseball is maybe an easier example than basketball because personal style matters less) That''s a simple test of a system but not easy to do at all.

Then in basketball you get things like personal styles of play, signature moves and such. In baseball you go up to bat and there are relatively few choices. You can hit for power, pull or hit oppsite way, tend to hit grounders or flies...that''s about it. In basketball things are a lot more open-ended. Do you dribble, shoot a 3, pentrate, pass? In baseball you make a decision about once every 5 seconds, in basketball you are making decisions all the time.

What I am saying is taking some stats and from that creating a realisitic gameplay is not easy at all. Also the idea that an AI player could sub for your teammate is weak. Sure, their stats might be the same. But will they pass the same, and at the same time? Will they take the same sorts of shots? Will they rebound or go for a fast break? Once again in baseball you have less problems like that but still major ones. What pitches do they throw and when? What pitches do they swing at? When do they decide to steal, go for a diving catch, etc.

The fact that the basic stats of the AI player are going to be the same doesn''t mean they will play the same at all. Unless stats are a really major factor that dwarfs actual playing ability, in which case why bother playing at all? If a team of "level 10" guys can easily beat a team of "level 1" guys just because their stats are that much better it seems pretty pointless to even play the games. So assuming that stats won''t be that far apart and won''t be the determing factor, the factor becomes player skill, which AI can''t replicate. (You can talk about learning patterns and such, good luck!) That is one thing that all MMO games must deal with: how important are stats, and how important are actual player abilities?

I understand why you would want and need AI players. But I would strongly suggest looking for ways to minimize their importance. If you are playing Quake and your friend leaves you probably don''t feel the same if you replace them with a bot. The same thing applies. One basic rule might be something like if you are playing against another human team, each team must be at least 66% real players. Something like that. That would allow you to deal with disconnects and such while avoiding a lame situation where you are playing by yourself with a bunch of AI bots against another team of mostly AI bots.

Also I would keep in mind that roster numbers should be expanded. A real basketball team is what 12 players? Given that you want at least 5 available to play online it probably makes sense to raise that to 20 or more. Maybe not as the active roster but as the roster pool. So only 12 players could be active to participate in any given game, but those 12 could come from 20 or more total team mates. So you find out who is online, if you meet the minimum human requirement you can form your roster and go play against some people, with AI bots or other humans filling in when people leave.

A final thing to keep in mind is that the higher up the chain you go the less people can play the game. In the minor leagues you have tons of players, then in the major leagues not so many. And MLB players outnumber owners 30-1. At any given time very few people are in the ownership position. At some point the value added for including those higher level items really decreases. I mean, making a MMO "Commissioner of Baseball" game probably is a bad idea since there is only 1 commisioner.

------------------

In terms of rating players, that part should be the easiest! Every sport has a wealth of numerical rating systems available. If you want to evaluate a player, just do it the same way you do in real life. Look at points per minutes, rebounds per minutes, etc. If you can do that and look at the character ratings (speed, strength, etc) you should be able to get a good idea of how good the player is and how much potential they have. If a guy is averaging 10 rebounds a game and his rebounding skill is only 5/10 you would figure the player is doing great things with their character. If you are looking to build over the long haul, that would be a guy to get. If you are looking for a quick fix a guy who gets 12 rebounds a game with a skill of 8/10 might be a better choice, although their player skill may limit their ceiling more.

The only other thing you need to deal with is stuff like how nice are they, how often are they available, etc. Which could actually be a big issue. How much money do you spend to get Jordan, if Jordan in a player who is only free weekends from 4-7?

Share this post


Link to post
Share on other sites
quote:
Original post by AnonPoster
I don''t think that is really the problem. I think the problem is that the problem is just really hard!

...

Then in basketball you get things like personal styles of play, signature moves and such... In basketball things are a lot more open-ended. Do you dribble, shoot a 3, pentrate, pass? In baseball you make a decision about once every 5 seconds, in basketball you are making decisions all the time.

The problem isn''t that hard, it''s that since the developers don''t know the game or play at a high level, they don''t understand the discriminants in making decisions. What a play should do is very easy to determine once you understand the game, as evidenced by the statements of commentators.

With the game the way it is now, New Jersey really needs to get Kidd back on the floor. His court awareness really spreads the Detroit defense, and he can thread those passes right past Wallace and Williamson to find Kittles and Van Horn with open looks.

Now what made him say that? A lot of selection in basketball is simple physics: keep your body between the defender and the ball. If the defender is to your right, go left. If you need to go left, get the defender on your right (shake, spin). Create as clear a path to the basket as possible using screens - set either by the ball handler (NJ, Boston) or by off players. Use big name players to draw defenders in then make the pass (Shaq draws a double team then kicks out to the shooter or the cutting player). Make the extra pass and get a free look at the basket (Milwaukee). Realistic gameplay has nothing to do with stats. This is all physics.

quote:

Also the idea that an AI player could sub for your teammate is weak. Sure, their stats might be the same. But will they pass the same, and at the same time? Will they take the same sorts of shots? Will they rebound or go for a fast break?

It only seems weak because you''re thinking in terms of current sports AI - which is pathetic. I played Live last night, and they flat out cheat to make the AI seem tough. Big deal. I beat the Lakers with Orlando by 1. I was smart; I fouled early, took time off the clock and had McGrady sink back-toback three pointers to give us the lead with 2.9 left on the clock.

There are a lot of things that current games don''t do well, or not at all. Take blocking shots (and allow me to reiterate that motionsets are evil); most games have you go up with one hand raised such that the offensive player can shoot right beside your outstretched hand. They make it impossible to come from behind and swat a layup, for example. I propose a "morph target-esque" animation system which allows for much more flexible motion and therefore more realistic play.

quote:

That is one thing that all MMO games must deal with: how important are stats, and how important are actual player abilities?

Gamer stats are largely irrelevant to play; they only limit athlete capability (using a Lindsay Hunter would make posting up on Rasheed Wallace illogical at best and flat out impossible at worst). AI stats are used as selection values - speed, quickness, etc will determine whether they choose to stop n'' pop, drive to the hole or dish off, as will the gameplan.

quote:

I understand why you would want and need AI players. But I would strongly suggest looking for ways to minimize their importance.

Absolutely. The idea is that if you''re playing and suddenly get dropped, the AI kicks in and notifies the coach who calls timeout as soon as possible so that another human player can get into the game. Also, if you can''t find enough human players to make a game then you can opt - and I stress opt - to play a game with AI players making up the rest of your roster.

quote:

Also I would keep in mind that roster numbers should be expanded. A real basketball team is what 12 players? Given that you want at least 5 available to play online it probably makes sense to raise that to 20 or more. Maybe not as the active roster but as the roster pool. So only 12 players could be active to participate in any given game, but those 12 could come from 20 or more total team mates. So you find out who is online, if you meet the minimum human requirement you can form your roster and go play against some people, with AI bots or other humans filling in when people leave.

This is cool. I''ll keep it in mind and see how people respond to it.

quote:

A final thing to keep in mind is that the higher up the chain you go the less people can play the game. In the minor leagues you have tons of players, then in the major leagues not so many. And MLB players outnumber owners 30-1. At any given time very few people are in the ownership position. At some point the value added for including those higher level items really decreases. I mean, making a MMO "Commissioner of Baseball" game probably is a bad idea since there is only 1 commisioner.

If you''re only talking about the US. Even within the US, there''s alternative leages for a number of sports - the recently-disbanded CBA immediately comes to mind. Then take it to an international scale.

This isn''t to be a "brand name" advertisement game; I don''t care for NBA endorsements (and I can''t afford them anyway). The idea is to provide a modifable platform upon which players can create their own teams, leagues, commissions - hell, even countries. Gamers all over Europe could recreate the FIBA league using this game - and that''s a huge number of teams!

I think your ideas on ratings are excellent, by the way. However, I think the "ratings" discussed earlier was in terms of suitability for a team. A guy might have a very high percentage, but that might be because he plays around the basket - dunks a ton, for instance - because he plays against defensively soft opponents. It might also be because he doesn''t give up the ball to his teammates and takes shots as soon as he gets it. Yes, he makes them, but is he doing his team good? Will his team go far were we to have a playoffs right now and they faced up against one of the top teams? Those are questions that can''t be answered by looking at the statsheet alone.

Also, if you only focus on the statsheet you might miss those diamonds in the rough - those guys that do all the little things like setting hard picks, grabbing rebounds, hustling, putting a body on tough players... guys that could really blossom into big-time players given the right environment. Kinda like Ben Wallace coming from Orlando to Detroit and now (2 seasons later) leading the league in rebounds and a major blocked shots contender. How about international players who would have been glossed over had GMs been college stats obssessed? Pau Gasol (MEM, Rookie of the Year, no doubt), Wang Zhi-Zhi (DAL), Dirk Nowitzki (DAL), Eduardo Najera (DAL), Zeljko Rebraca (DET), Peja Stojakovic (SAC, All-Star Three-Point Shootout Champion), Hyadet Turkoglu (SAC), Vlade Divac (SAC), Toni Kukoc (RET?) - guys who have all made huge contributions either this season or in past seasons.

That''s why we have the Scout role. Guys who spend time watching games and players trying to identify real talent. Scouts can trade information however they deem fair or post findings publicly and confer over picks. I''ll try to flesh out that role and make it as social as possible.

Okay, I''ve said enough. Just stop telling me I can''t do what I can do. This isn''t rocket science; it''s a basketball game.

[ GDNet Start Here | GDNet Search Tool | GDNet FAQ | MS RTFM [MSDN] | SGI STL Docs | Google! | Asking Smart Questions | Internet Acronyms ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
quote:
Original post by Silvermyst
I like it.

I''ve been wondering why there still is no real MMO sportsgame out there where players can compete against eachother while building a career.

Seems to me that there are enough players out there who take pleasure in proving themselves better than others. Why not create a real competition for them?


I guess in a way u could say tony hawks is simalar to this but not online or any games where u build a carreer in sports before tony hawks, if there is one? spawned this idea.

Share this post


Link to post
Share on other sites
Sounds like a cool idea...I personally find the concept of having to train and market your character to be the most compelling part of your idea. In regard to preventing ball hogging, I would leave that job to the coach of the team (whether AI controlled or human); if it''s human controlled then it''s easy...the coach controls which players come in and out and take out ball hogs at any time. If it''s AI controlled then it gets a little more complicated because you''d really want to simulate tactics (however I guess you could just get play books and choose the best play based on the current state of the game); however in general any player that deviates too substantially from the coaches play can be forced to sit on the bench. If "in the huddle" talk is allowed, the players may be able to influence the coach''s decision too.

Share this post


Link to post
Share on other sites
quote:
Original post by bob_the_third
Sounds like a cool idea...I personally find the concept of having to train and market your character to be the most compelling part of your idea.

Thanks. I thought that was one of the coolest parts of it too.

quote:

In regard to preventing ball hogging, I would leave that job to the coach of the team (whether AI controlled or human); if it''s human controlled then it''s easy...the coach controls which players come in and out and take out ball hogs at any time.

Precisely, and in the huddle talk - as well as talk while on the court, during sideouts, etc - should be critical to a social game. The key is simply to give people a lot of flexibility and support technologies like VoIP. Instantly, players are able to talk to each other - a critical part of sports in real life.

quote:

however in general any player that deviates too substantially from the coaches play can be forced to sit on the bench.

Well, not exactly. If a player deviates from the called play, but still puts up numbers, passes and does all the things that win games, the coach may decide to let them play. Many coaches allow their better players to be like on-court coaches, particularly point guards (and to a lesser extent, two-guards). If, OTOH, the player is following the general playbook but not passing the ball or taking good shots - generally making bad decisions - then [s]he should be benched. So it''s a balance of playbook versus production, and at different times and for different players different factors will be emphasized. Early in the season or in pre-season games, the playbook will be emphasized, particularly for rookies and "junior" players. At playoff time, OTOH, production is just about all that counts, provided you''re not disrupting the flow of the team.

Great comments. Thanks!

[ GDNet Start Here | GDNet Search Tool | GDNet FAQ | MS RTFM [MSDN] | SGI STL Docs | Google! | Asking Smart Questions | Internet Acronyms ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
I just wanted to re-iterate a point that got glossed over before. There will always be many more minor league players than major league, many more players than coaches, and many more coaches than owners.

Adding teams doesn''t change the proportions. For each owner you are going to need an entire stable of coaches and players to play for you. If owning is popular, how do you ensure that? If you get 5000 people signed up to play as owners, and just 10000 as players, what do you do?

Share this post


Link to post
Share on other sites
quote:
Original post by AnonPoster
I just wanted to re-iterate a point that got glossed over before. There will always be many more minor league players than major league, many more players than coaches, and many more coaches than owners.

Adding teams doesn''t change the proportions. For each owner you are going to need an entire stable of coaches and players to play for you. If owning is popular, how do you ensure that? If you get 5000 people signed up to play as owners, and just 10000 as players, what do you do?

There are two options. The first is that the ownership simulation - like the sports simulation - require advancement in that you can''t just walk in and "be" an owner. You have to make investments in teams or "rise through the ranks" to hit an ownership position, if you even want to. Very few teams are actually owned by just one person; it''s usually a family or a group of associates. Furthermore, there are other, probably more strategically important, positions than owner - such as General Manager. I think the Scout role could also play a part here, so that could be a crossover role, working with the GM to make smart acquisitions.

The second solution is to simply create a team of al AI players for every new person that decides to be an owner right out the box. If the guy makes smart decisions, his team value will rise and real human players will sign. If the guy makes bad decisions, the team will fall and either be bought out by a larger team/organizations, its assets being absorbed and/or redistributed as the acquiring organization sees fit.

Since the management sim is still far off, I have time to adjust, tweak and otherwise correct this model. My focus now is on the sports sim.

[ GDNet Start Here | GDNet Search Tool | GDNet FAQ | MS RTFM [MSDN] | SGI STL Docs | Google! | Asking Smart Questions | Internet Acronyms ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
The trouble with mixing AI and humans in the same game is keeping the abilities even. If the AI players are better than human players, then owners will not want humans on their team, and that is terrible. If the AI is worse than the humans, then teams that for one reason or another are composed predominantly of AI will be very handicapped. And if they are equal, then there is the problem that there may not be enough hard decisions for the players to find the game fun.

On the other hand all-human teams are prone to schedule trouble. You could take the step of incorporating player scheduling into the game, so that players join teams that are naturally on at the same time they are. Or you could dissociate the human controlling the avatar from the avatar itself. Coaches/GMs could trade avatars, statistical representations of player capability, and then could get the actual game played by whoever happens to be on at the time. Players'' stats would be kept according to how well they used the avatars they controlled. The trouble with this is that players might not feel very attached to the outcome of the game.

While the idea of a stratified MMO game where some players are leaders and others are followers is fantastic, the design is troublesome :} I think it could be made to work, but you have to choose among the lesser of several evils. But I think in the long run it might contribute to player equality and game accessibility, because newer or less skilled players would not be so overshadowed by the more dedicated players.

Share this post


Link to post
Share on other sites