Questionable AI decisions

[ravinhood]

If the devs wanted to they could program the AI to make it so elite a human player would barely stand a chance.

This ^ is what I'm looking for in a good baseball stragety sim. But, in a way that makes it somewhat fair to the player (human) me and I still have a "chance" if I do everything right. I mean as it is now even on hard or hardest trading settings I can still finagle the trade I want if I take enough time to do it. A lot of it falls to being able to pick up cheap good guys still in the FA pool. They should have the AI try to DRAIN the AI pool every year as much as possible.

I do have to disagree with you about no game is perfect. Front Page Sports Baseball came about as close to perfect as they came. I do remember the trading of players was a lot harder than it ever has been with Out of the Park. I honestly believe their trading system was setup on a player for player point system and the numbers had to match pretty darn close to be able to get a trade. This game if you throw enough money and other crappy players at the computer AI it will accept not taking into account everything it should have. I think a lot of it has to do with basically unlimited players to get in the FA pool and in Front Page Sports Baseball you couldn't do that. There were limits.

[ra7c7er]

Originally Posted by ravinhood

This ^ is what I'm looking for in a good baseball stragety sim. But, in a way that makes it somewhat fair to the player (human) me and I still have a "chance" if I do everything right. I mean as it is now even on hard or hardest trading settings I can still finagle the trade I want if I take enough time to do it. A lot of it falls to being able to pick up cheap good guys still in the FA pool. They should have the AI try to DRAIN the AI pool every year as much as possible.

I do have to disagree with you about no game is perfect. Front Page Sports Baseball came about as close to perfect as they came. I do remember the trading of players was a lot harder than it ever has been with Out of the Park. I honestly believe their trading system was setup on a player for player point system and the numbers had to match pretty darn close to be able to get a trade. This game if you throw enough money and other crappy players at the computer AI it will accept not taking into account everything it should have. I think a lot of it has to do with basically unlimited players to get in the FA pool and in Front Page Sports Baseball you couldn't do that. There were limits.

I don't think you can count a game from the 90s that allowed 10 man rosters as the best ever. The reason that game was hard was because of all the limitations it had. Plus the financials were nothing compared to what modern games are dealing with. Not to mention all the bugs the game had and the fact that players only had 5 ratings. The games rating engine couldn't differentiate power hitters from contact hitters. Making RL power hitters worthless in the game. Base stealing went by player speed only. And the pitcher injury bug which meant 3 pitchers in consecutive pitches would be injured (which was never fixed btw because the devs claimed they couldn't figure out why it was happening). Hardly the best game ever.

As for the FA pool I completely argree with you. OOTP tends to way overdo it on FA pool. Personally every few seasons I remove the bottom 10-15% of players (guys who are never going to play anyway). OOTP could really use a retirement algorithm for the FA pool. Something like if you're postional rank in the "world" is below 2 times the worst active player then the player would either switch positions to one where he meets the criteria or he has a high chance of retiring. One of the issues I see with OOTP is that you can't have one entry pool for an entire world. If I have 5 leagues and I want all of them to have a draft I have to create 5 drafts worth of players and 5 drafts worth of extra players. If a player could just set his world to create X number of entry players a year and have every league draft from that pool. In an order the player chose you could VASTLY eliminate some of the garbage FA players that will never make a team and still keep the overall quality of the players reasonable.

[ravinhood]

To me it was the limitations that made the game great. Being realistic doesn't matter to me as much as "challenge" does and you can't deny the challenge that game gave. Front Page will always remain the BEST in my book and a horde of other players book from the demand to see it come back as I have seen. Not to mention the excellent graphics for the time and the crowd noises. It might have not match MLB tit for tat but it never meant to. It brought to game players what they wanted good animation with a fair stat game. OOTPBB is just the opposite it bring a good stat game with a piss poor animation game and career mode game and trading game. I weigh all the elements I enjoy and Front Page is the BESTEST ever with MicroLeague coming in 2nd

[Kobeck]

microleague was awsome but just a different animal than OOTP

FP was good, but IIRC crashed more than Lindsey Lohan

[ra7c7er]

Originally Posted by ravinhood

To me it was the limitations that made the game great. Being realistic doesn't matter to me as much as "challenge" does and you can't deny the challenge that game gave. Front Page will always remain the BEST in my book and a horde of other players book from the demand to see it come back as I have seen. Not to mention the excellent graphics for the time and the crowd noises. It might have not match MLB tit for tat but it never meant to. It brought to game players what they wanted good animation with a fair stat game. OOTPBB is just the opposite it bring a good stat game with a piss poor animation game and career mode game and trading game. I weigh all the elements I enjoy and Front Page is the BESTEST ever with MicroLeague coming in 2nd

Obviously their is no reasoning with you. You claim you want OOTP to give you everything and be just like RL (as you've said in other threads) and now here you're saying you want something as limited as a nearly 30 year old game that allowed 10 man rosters and couldn't even properly handle batter statistics. One or the other dude. One or the other. Also can you show me the horde of players wanting Front Page back?


I've agreed with some of what you've said and others have too. but all you seem to want to do is argue in every thread you've posted in. If you like the other games so much go play them. You can get Front page as freeware now. Go play the perfect Front Page Baseball and every time you lose three pitchers on three pitches because of the bug just remember it's perfect.

I'm ignoring you now (as I recommend others do to) so I don't have to deal with it anymore so reply if you want but I won't see it.

[ra7c7er]

Originally Posted by Kobeck

microleague was awsome but just a different animal than OOTP

FP was good, but IIRC crashed more than Lindsey Lohan

Microleague was good for what it was but it to was so incredibly limited.

And yes FP crashed constantly. Sierra actually recommended saving after making any major changes to your team and after every day in the game.

[Kobeck]

Originally Posted by ra7c7er

I've seen squeeze plays. They are very rare. On the same point not a single squeeze I've ever seen has worked.

A lot of managers in game won't take minor league deals right away when signing comes around. What happens is that the AI takes the best available at the time because it has to fill the position. So as a human player if you wait you'll be able to pick up better left over managers because later in the off season they will take lesser jobs. Just like players will take less later in the off season or after the season starts. As much as I'd like to see it change trades, FA, and hiring and firing are some of the hardest things to program in a sports game. Heck look at EA and madden it still sucks there and they've had 25 years and umpteen hundreds of programmers working on Madden. I hardly place blame on a small group of 6-7 who are making a stats only baseball game.

Basically you're stuck programming one of three ways for the AI to sign both players and coaches.

1. have AI go after and sign players immediately with the best available they can get at the end of the season. (this is the setting most sports games use, not just OOTP)
Pros- AI has a good chance of getting some good players before the human player can sign them all (under normal circumstances), Quality players spread around the league.
Cons- AI will routinely outspend itself because player demands are highest at the end of seasons. Easy for human player to exploit because human player can go after players the AI won't even look at since the AI is stuck on best available player and doesn't look at the FA pool as a whole nor looks at what type of secondary needs the team has.

2. Have all AI teams wait till the end of the off season to sign players and personnel
Pros- AI is less likely to overspend to get players which makes the overall financial structure of the game more stable.
Cons- Much much easier for the human player to take advantage of. One or two high budget AI teams will sign a significantly higher amount of quality players leading to a insurmountable deficit in quality from the top of a league to the bottom. Basically bottom tier team will never be able to make it back to the top.

3. Sign fully on a as need basis based on current rooster.
Pros - AI will never go after players it doesn't need. Harder for human player to exploit because AI will typically have more money. AI team will usually always have a good spread of players among all positions.
Cons- AI still goes after best available. AI will still overpay. AI rarely signs upgrades or futures for older players. High amount of quality players left in FA pool after signings are over (could potentially be a good thing if you have more than one lead fed from the same pool of players as more quality players will trickle down to lower leagues)

Those are the three main ways sports games cover signings. As you can see each has pros and cons. The same is true for all AI decisions each way of programing the AI has pros and cons. And we as players WILL ALWAYS latch onto the cons and run with them. Sometimes the cons can be minimized or even eliminated but sometimes they can't. Again look at all the problems games like Madden still have. Many of the same things that plagued Madden 10 years ago still have issues today and again they have 10xs the programmers and 1000000x the budget.

If the devs wanted to they could program the AI to make it so elite a human player would barely stand a chance.

Back on the signing players and coaches thing. IMO the best way for OOTP or any sports game to sign FAs would be to try an mimic real life. Obviously the best players go first and usually the top contracts are considered to be overblown but contracts quickly fall back to reasonable levels after the top 1% of players are signed. (OOTP tries to do this, I think, but it runs into trouble because every AI team is going after every player at the same time not spread out like IRL). Player demands wouldn't be the starting point for contracts. Player demands would be pretty close to the signing point or close to the point where a contract becomes toxic. Contracts wouldn't be able to be backloaded to extremes. In past madden games they curbed the backloaded contract exploit by coding into the game no more then a ?22%? (I think) raise from year to year. (I vaguely remember it being a sliding scale based on the first years salary but 22% was I think the largest year to year margin you could have). The FA signing period would be more drawn out. As it is in the game the first week-two weeks is a mad dash and then signing is down to nothing. If signing was more drawn out it wouldn't really help the AI any but it would be a bit more realistic. AI would take more consideration into what it actually needed. If a team (minors included) had a glut of quality players at a position they shouldn't be going after high priced FAs. The same goes for if a team has an aging high end player they should be going after reasonable younger players of that position for the future. If the team has a 5 star player at a position locked into a long contract they shouldn't be going after another high end player as a backup. They should be looking at 2.5-3 star guys as backups. Or properly evaluating their minors and bringing someone up. Then replacing that guy lower down the FA list. Overall the AI in normal conditions should want to play moneyball (i.e. best value not best available) with modifiers based on budget, current payroll, short term needs, long term needs, etc moving the scale up or down as to how free wheeling the AI is with money.

I could go on but this post is already way to long. Anyway OOTP isn't perfect but it isn't horrible either. It's actually pretty good. their are flaws here and there but that's with all games. And I can say from personal knowledge the devs here are WAY WAY more receptive to comments and criticism then most.

Again (I'm almost done really), the financial aspects of a game are hardest to program. I'm sure the devs could completely rework the financials to make them work better BUT all that time would come at a overall cost. Would you buy OOTP17 if it was the exact same as OOTP16 but with better financials? Remember everything else is the same and oh yeah they didn't have time to update rosters either.

In the end no game is perfect and the way some people are acting currently on the forums it's either pile of garbage or rose tinted glasses. This game is neither it's right where it should be for a game of it's type. It's the best baseball sim on the market by a long shot and we even as players should want nothing more then to keep it that way. Petty back and forth does nothing to further the game.

nice read.

I've always thought a variable that take into account players hometown/college/previous MLB stints vs offers.

Allow a small home team advantage if say the guy grew up in Georgia, played BB at Alabama and the Braves and Yankees made offers. Yankees would have to bring way more real $'s into the contract.

Having said that I would think prestige of the MLB clubs should be included. I think it is in that a level 2 league dont usually sign Alex Rodriguez types. Could just be a $ thing.

[ra7c7er]

Originally Posted by Kobeck

nice read.

I've always thought a variable that take into account players hometown/college/previous MLB stints vs offers.

Allow a small home team advantage if say the guy grew up in Georgia, played BB at Alabama and the Braves and Yankees made offers. Yankees would have to bring way more real $'s into the contract.

Having said that I would think prestige of the MLB clubs should be included. I think it is in that a level 2 league dont usually sign Alex Rodriguez types. Could just be a $ thing.

Loyalty does affect a teams ability to resign players and for cheaper. I think their is a small percentage of players that want to play near their hometown already in the game not sure how big it is. If it's not in the game already it might be coming now that owner goals include things like hometown players and attendance. Hometown and local players are more important in OOTP16 then in previous versions of the game.

I don't think college and previous experience make any difference but I do know that in the past people have made suggestions about where players went to college should have some small effect.

Not sure if you play fictional or not but the game does have a league reputation system mostly used for independent and international leagues. It's a 1-10 scale with 0 included. I haven't made a 0 reputation league yet but I want to. High loyalty players will be more willing to stay with their teams even in lower reputation leagues. You'll get some 4.5-5 star guys willing to stay and make less. Which I think is somewhat realistic.

[Dyzalot]

AI decides to make a pitching change in the top of the ninth but since the pitcher will be leading off the bottom of the ninth, it makes a double switch. Sounds good so far. But then when the bottom of the 9th starts the AI pinch hits for the guy he used for the double switch. Note that the the guy who pinch hit was also a middle infielder and the pitcher that pitched the ninth for the opponent threw with the same hand as the previous pitcher, so that isn't what confused the AI either.

[ra7c7er]

Originally Posted by Dyzalot

AI decides to make a pitching change in the top of the ninth but since the pitcher will be leading off the bottom of the ninth, it makes a double switch. Sounds good so far. But then when the bottom of the 9th starts the AI pinch hits for the guy he used for the double switch. Note that the the guy who pinch hit was also a middle infielder and the pitcher that pitched the ninth for the opponent threw with the same hand as the previous pitcher, so that isn't what confused the AI either.

Yeah a dev is already looking into it. He's admitted it isn't supposed to be happening as much as it does.

[Dyzalot]

Going into the top of the ninth with a one run lead the AI brings in the closer but decides to make a double switch because the pitcher's spot is due up next replacing the excellent defensive catcher with a very mediocre one. I don't think double switches should be happening in this spot.

[bazeall]

This is for trade AI. I would not mind the extra time it took.

Any time I trade with a computer team, let it sim 5 seasons forward both with and without the trade. This way, it could evaluate the trade.

[Dyzalot]

Originally Posted by Dyzalot

Going into the top of the ninth with a one run lead the AI brings in the closer but decides to make a double switch because the pitcher's spot is due up next replacing the excellent defensive catcher with a very mediocre one. I don't think double switches should be happening in this spot.

So should I be posting this one as a bug as well?

[Lukas Berger]

Originally Posted by Dyzalot

So should I be posting this one as a bug as well?

I would, it seems fairly questionable.

[Dyzalot]

Unnecessary double switch in the bottom of the ninth when the closer is brought in which removes the good defensive center fielder for a poor defensive center fielder. There is no reason to double switch in the ninth if bringing in a new pitcher with the lead unless it leads to an improvement in the defense as well.

[Dyzalot]

Runners on 1st & 2nd, bottom of the ninth with the score 3-1. AI pinch runs for guy on 2nd.

[Dyzalot]

Down 4-0 in the bottom of the ninth the runner tries to go "first to third" against Puig's arm.

['94 EXPOS]

Originally Posted by Dyzalot

Down 4-0 in the bottom of the ninth the runner tries to go "first to third" against Puig's arm.

Looks like an exciting baseball play to me. Bad idea to run? Probably. I have watched a couple of baseball games in real life where players have been thrown out at a base before.

Who was running? Was he fast but a poor baserunner?

[Dyzalot]

Originally Posted by '94 EXPOS

Looks like an exciting baseball play to me. Bad idea to run? Probably. I have watched a couple of baseball games in real life where players have been thrown out at a base before.

Who was running? Was he fast but a poor baserunner?

McCutchen was running. Speed = 10, running = 15 (1-20)

[ra7c7er]

I feel like it was a great play too. Just because it's Puig doesn't mean people aren't going to try. Not sure what you're trying to show with that one.

As for the others you keep posting up. The devs already know and this isn't the bug forum. One of the devs even came on this thread and said he's looking into it you don't need to keep posting it up. Everyone, including the devs, know it's happening.