Surprised by the AI

[orca]

In my first campaign game I've been very impressed with the play of the AI. Inevitably there is the occasional stupidity, but it usually does a good job. I do think that the rebels accept battle too often, but that's my only concern.

Anyway I wanted to tell the story of how the AI really got me. I'm playing the British in the 1775 scenario. In 1775 I just hold my own - Brown raises hell in the South, Dinwiddie sadly doesn't manage to last the year, and Gage holes up in Boston. When Howe arrives I send three regiments to Canada, and with these I'm able to secure Ticonderoga by the end of the year. Before Washington shows up Howe manages to win a hard fought battle outside Boston with heavy losses, and I think I see that the rest of the rebels retreat to Newbury. Spend the winter beseiging Newbury and keeping the army supplied, and it turn out there are only three regiments there.

Early in 1776 I've collected enough troops in Canada to begin an offensive towards Alabny. I have 3 regiments head south from Ticonderoga and three from Oswego. Turns out I meet very little opposition, just a couple of militia regiments that are easily killed. The main army with reinforcements splits up to secure all of New England, and Parker's fleet lands Cornwallis directly into New York City. There are now lots of rebel armies running around, some quite strong, and I spend the rest of the 1776 campaigning season running them all to ground and destroying them. By winter I have 8 or so small armies scattered from Norwich to Stanwix, but all of New England and New York are completely controlled (well Maine is still in revolt, but that is easily cleared early in '77). Prevost has used the regulars from Florida and Jamaica to secure all of Georgia and 96.

My plan for 1777 is to really shut down all rebel recruitment by taking Pennsylvania to shut down the middle states and the Carolinas to shut down the deep south. As an afterthought I gather a couple of regiments on Lake Ontario to go take Pittsburgh.

The spring and summer campaigns go well. I trap 4 regiments in Charleston and take it. I then have enough strength to squeeze the rebels out of South Carolina, catching Marion and destroying his partisans. By the end of the summer I hold Charlotte and have no known rebels behind my lines. In the north I land an army from Newport in Head of Elk, while armies from West Point and New York push on Princeton and whatever the town by Allentown is. The force at Head of Elk fails to catch anyone rushing to support Philadephia and I begin to secure Pennsylvania. I still haven't seen many rebels, and the garrison in Philadelphia is minimal. I can't believe the AI has abandoned the area, so I advance methodically and wait for more of my forces to arrive from New England. By the end of the summer I have the line of the Susquehanna, and can see rebel armies moving around on the right bank. I send one army to Wyoming, send one across the Susquhanna a bit below that, with a third army to hold the left bank and two other armies heading for Baltimore and the area west of it, so that they can move against rebels trying to escape my (presumably) crushing victory in the upper Susquehanna.

Well, Phipps on the right bank wins two battles at some cost and is then mauled by another rebel army. No battle in Wyoming, but I end up with a seige of a garrison of just a couple of armies (my intelligence is still pretty poor). No sight of rebels on the left bank, and no sight of any enemy forces in Maryland. My offensive has hit thin air, Phipps' army will be licking it's wounds for the winter, but at least I have Wyoming.

In October I send armies to secure Maryland, confident that the rebels must bhave escaped further south. I send the army from the left bank to reinforce the seige of Wyoming, which I assault. The advance guard in the south is in Hillsboro, and although the surrounding countryside is unsecured Charlotte is well under control and can send help quickly. So the Deep South is mine, and Pittsburgh is under seige (the last American objective in the West). I've pretty much met my goals for 1777.

And here is where the AI gets me. In October Washington shows up with 6 regiments on the Monogahela. My army runs like heck for Niagara, but get sinto a fight and I escape with 2 much weakened companies of regulars left and have to send that regiment all the way back to Albany, abandoning Niagara on the way (the rebels take it during the winter). The American army in Wyoming turns out to be almost a big as the reinforced army beseiging it, and although my assault shows Lee that Americans are no match for my regulars, the beseigers are badly damaged and I'm not willing to risk losing a half dozen regiments in further assaults that may not suceed, especially with winter coming. I find I don't have enough supply trains nearby to support a winter seige of Wyoming, so I withdraw back across the Susquehanna.

So I assumed that the Americans would withdraw south to protect Virginia, their best recruiting ground. Instead they snuck most of their troop into Wyoming under my nose, ensuring that they would be able to recruit in Pennsylvania in 1778, while sending the rest out West causing me to lose control of everything between Oswego and Detroit. A very impressive trick that I couldn't imagine any AI to be capable of.

Turned out OK in the end. The drawback with this AI response is that it left Virginia exposed, and in Spring 1778 I was able to enter Virginia in force and secure North Carolina. I also squashed Wyoming in the spring and spent the rest of the year chasing down the armies from Virginia and Wyoming. I didn't get Pittsburgh until spring 1779 when I finally won.

It wasn't a game winning move from the AI, but it threw me for a loop. I really do love the AI in this game, AGEOD has done a great job with it.

[Pocus]

Thanks much. Nice post too. That would be very cool if you posted that on the Matrix forum