How to whip the AI

[akuenzi]

To the great minds that post here:  What are the ways to most effectively crush one's AI competition?  I know there are many statements here on just how stupid the AI is, but in the games I play, sometimes they are pretty good, and stubbornly stick around when I'd like to drive them off the map.  The question is how to do that.

 

1.  Buy them on the stock market.  Okay, I've done that.  Once a person has control, you can do all sorts of fun things with them.  However, I've been in situations where I'm not able to purchase a majority of the stock to do this.

 

2.  Merge their railroad in with mine.  I've never actually done this in gameplay, but I instinctively don't like it because it ends up giving one of my AI competitors a boatload of money in the process to buy them out.  Sure, grease a palm with enough cash, and it can be done... but I'd rather leave my AI competitor penniless and begging on the street (because he's a bad guy, of course :) ).

 

3.  Compete head-to-head where the competitor is operating.  I've tried this before, and all it has accomplished is throwing my own money away.  I build a separate track and network of stations as close as possible to my competitor, then try to beat him with whatever he's hauling.  The theory being, take away from what he can haul, his revenue and profits should go down, right?  Even with stocking this railroad with locomotives from other slave companies, I still haven't been able to defeat the AI with this tactic.  Somehow they still survive and all I do is grind money.

 

4.  Creative terra-forming.  This random thought just came to mind -- I suppose with a little effort one could mess with their track slope and mess with the speed with which they deliver their cargo.  Might be fun... but then again, it also costs money, potentially a lot of money and I'm not sure how much good it would do.

 

5.  Lay rails across the competitor's rails and create blockage.  Yeah -- park the locomotive right in the way of oncoming traffic and stop the whole line!  Wish it was that easy.  I actually tried this a time or two and couldn't ever get it to work just the way I wanted and is only of temporary impact.  It also costs money, when I hoped it would break down right across his rails and make a permanent blockage.  Running my trains on his rails doesn't work as his always takes priority.

 

6.  Another thought comes to mind.  Is it possible to bulldoze industry that you own?  I know we can't bulldoze stuff within the AI's station area, but what if it's an industry and we buy it?  Does anyone know if we can bulldoze it then?

 

So much of the above is mere gimmickry.  In actuality, I'd like to defeat them head-to-head (#3 above).  In the real world, isn't that how it's done by most of us?  If we want to beat the competition, we need to do it faster-better-cheaper.  Sure would be nice if it could be done in here as well.  Even when I've micromanaged the stuffing out of it, I still can't defeat the AI when going head-to-head against them.  Does anyone else know how to do this effectively?

[j7n]

Case 3 could work better but also become unrealistic if you create 2 stations near each competitors station as close as the space permits. Then have a train go between the nearby stations, dumping any cargo whether demanded or not. Your trains would be ready most of the time to pick up loads rather than en-route between distant stations.

Case 5 cannot be implemented when you need it. If you are lucky and have an AI build over your existing track, then it can work. I've used this method several times. I think the throttle setting has an impact on priority. Blocked trains will eventually start up and gain priority after waiting a really long time.

Something related to case 5, but not quite the same is to divert AI trains running on your track onto spur lines, bulldoze the connection to the main network, and hold those trains hostage forever. Such trains will take up smaller amount of space than a blocked line, and I think you can even move them later if you need the land. The AI might even upgrade those trains (but don't take my word for it).

I did attempt Case 3 in the Antarctica scenario. It removed profit from AI companies, and didn't cost as much to run. However, enemy companies were allowed to repeatedly declare bankruptcy halving their slowly built up debt, and would not be liquidated. A company getting liquidated happens once in a blue moon. In the end I had to propose a merger to purchase their debt with real cash.

[outofmage]

1. If you have trouble controlling AI, you can try my last hand of stock method.

 

2. I don't like to merge AI's crap either. At least if I merge an AI, I am planning to destroy my company too.

 

3. I have never tried this.

 

4. That's quite creative, maybe you need to save/load to find out which route computer prefer to connect, then mess up with the terro they tend to build tracks on.

 

5. If you want to block AI's trains, you can buy a cheap train and run on AI's busiest track, then press Ctrl+B, your train will break on AI's track and block the track. You don't need to build track across AI's railroad, and it's not going to work if AI build their track first. The priority will always be local trains>foreign trains. But broken trains will always have higher priority despite the track ownership (with normal throttle). 

 

j7n's method will also work, but you need to seduce AI onto your track to do so, but this method is permanent, mine needs to press Ctrl+B every 5 or 6 months.

 

6. You can bulldoze your own industry, but not within station's radius owned by other company. Even if you can, I believe it's more of your lost than AI's. AI can always get 80% of base value even this station does not demand this cargo at all, so they still get quite some revenue, but you lost half of the purchase price paid for this industry and a stable stream of income every year. 

[MaglevForever]

Looking at all these ideas, I see quite a few that I have never thought about trying. Good work to think them all up. I agree that head-to-head seems most realistic to me too. This is what I imagine I do. This is not a set of rules just a guide on how I typically play. If I called it something it would be: control everything and be one step ahead. It is possible in most games a notable exception being some of those with large pre-built AI.

 

I will have an expansion race with the AI (they always win but it keeps them busy). I use them to build up my net worth by investing smartly in them. If they pick good cities I sometimes invest all my money in them while issuing stock and steadily building my railway up. Then I will buy my company out once real bond-fueled growth is coming (preferably keeping a controlling stake in the AI as well). After 5-8 years most of the AI's growth is over especially in the second century (I prefer playing in the second century). Hopefully they buy back stock to increase my PNW value and pay me some dividends. If I am playing a medium sized map by about year 15 I should have expanded to all the cities of a good size that the AI didn't connect to. Then I can start saving and waiting for opportunities to buy them out. If their chairmen end up rich I don't mind, if they start a new company it wont make much money anyway. Sometimes the AI chairmen decide that a share is good value especially if you are selling it. If you can trick them into buying enough on margin it is possible to completely bankrupt these chairmen them when a economic downturn comes around. Like someone said on here in real life it is called insider trading.

 

If the scenario involves the DB 18 201 the AI can be a money pit. If they make a large network they will try to change their routes to long haul after they have reached a certain size and 7-8 years have passed. I recently tried to win the campaign entitled "America after the flood" without touching track or trains. I started a company then increased my PNW via stock buy-back then retired. Then I tried to invest smartly in the AI to get 25 million PNW. I managed 23. As soon as they changed their routes at the 7-8 year mark they went from quite profitable to barely breaking even. Personally, I wish they were a lot smarter. I almost feel sorry for them sometimes.

[akuenzi]

Then I can start saving and waiting for opportunities to buy them out. If their chairmen end up rich I don't mind, if they start a new company it wont make much money anyway.

Maybe this is ultimately the only way it can be done, assuming one 'misses' the opportunity to own more than 50% of them to take control and then bury their company... then again, maybe there will be opportunity to try again once they start up a new company.

How about this -- is it at least possible to 'quarantine' a competitor by laying small bits of track around his entire railroad? Is it possible to lay it in such a way that he can't go across it, and thereby never expand? I'll have to give that a try. Maybe he could still cross it, but at the very least it would force him to pay me for it...

[jeffryfisher]

More tips:

 

* Use tons of bonds to win the early expansion race.

 

* Once any company builds a station within a town's radius, no AI will ever build a station there. If disconnected track is allowed (and if you have territory rights), you can invest in some stations to claim important cities even before you can afford to wire them up.

 

* Make sure you set your game preferences to "liquidate bankrupt companies", or else their corpses will remain as permanent roadblocks after you crush them.

 

* Only play scenarios (maps) that allow chairmen to be fired, otherwise they'll continue forever like zombie black-holes of infinite debt.

 

* When an AI player is deep in debt (margin), and if there are several blocks of stock available in the market, try a "bear raid". That's where you sell short in order to push your opponent into a margin call. If he's super vulnerable, his selling will depress the price even more, forcing more selling. He could be wiped out completely, and several stocks could then become available for pennies on the dollar. You can then snap up the cheap shares to make a killing and merge or take control (if control is allowed).