On the Road to Expert 7: Analysis

Article written by DaRq_DarkJihad
Originally published on 11-11-1999 ; updated on 08-17-2014
Tags: Basics Archives

Article Downloads:
» Jihad vs. Armagedn

Ok since yesterday I was a little devoid of ideas and gave you a mediocre post today I come back with one of the things that I really think separates the experts + those who will be from those who aren't. Frankly even Sammy could learn from this post... There are not one but TWO times that you need to go into game analysis mode in AOK. Immediately Post-Game and during recorded games. Every loss you have you need to do BOTH. I recommend refusing to play 95% of the time if record game is not on.

ONE) Post-Game Analysis.... Do NOT simply quit game immediately and assume you will come back and watch the recording, immediately after the game you will have a clear perspective of what happened and a better ability to analyze the late stages of the game.

First, you want to resign, not quit so the map is completely revealed. Try to resign when you're beat but not decimated, it tends to keep the map in a more "usable" state.. ie you can see more what your opponent is doing.

Assuming you got beat.. which is the best time... look all around the map. Note your opponents building placement & number. Ask yourself these questions: How many villagers did my opponent have and what were they on ? How many TCs ? How many villagers at Each ? Did your opponent spot wall ? If so where? If so think back to the game and note if it was effective or not and why. What buildings did he build and more importantly how far are they away from his original town (ie the more forward the later to a certain extent) ? What troop mix did he beat you with, and more importantly, what was he/she preparing at there buildings to back it up ? Note the placement of all town centers and ask yourself, could i have grabbed that spot earlier ? Was it important? Did their TC placement help defeat my attack ? Why ? If i had gone a different way would/could my attack have succeeded (IE I should have scouted more) ? Did they use terrain to their advantage ? If so how.. I generally spend 3-4 minutes checking over the map after a game to get a better feel on how and what my opponent was doing.

TWO) Recorded Game Analysis New to AOK, recorded games offer a second and better way to analyze your games and mistakes & triumphs real-time, however I still recommend post-game analysis as well.

Ooh I almost forgot(this is edited in) I often click as if I'm playing, and especially with methos, matty games I find that not only was I looking at the same spot but I did almost exactly the same things (ie the units move exactly where i sent them at nearly the same time i clicked, as if I was moving them even though I can't) I find it really helps me "get in the head" of the player I'm watching.

During a recorded game you aren't doing exactly the same things as post-game. I generally believe you should start as yourself with reveal on. Quick take a look and then flip between you and your opponent every 15-20 seconds or so, flipping on and off reveal to get a feel for both what you did see and what you could have seen with better scouting. I really don't recommend fast-forward but whatever floats your boat i suppose. Watch your own and your opponents resource distributions and villager count. If he gets ahead of you, how did he do it. Does he have more resources with a similar number of villagers ? How did he/she do it ? Also watch your opponents scouting patterns. Every once in a while click between military and economy military maps. If you/they had any idle villagers look at where they were and what they were doing, so next game you can be sure villagers there don't go idle. If your opponents has very few/no idle villagers, look at what they did to keep them busy and eliminate pointless micromanagement.

As you move to feudal & castle watch your own and opponents buildings going up and forward builders if any. Where did they place those builders/buildings ? Should you have moved your own buildings ? As you attack + troop movements begin note if your troops went along the best path, try and figure out which the best path would be (the one thats safest and will leave you unobserved and coming from an unexpected direction) ? Did you pick the right units to attack with ? Were you easily countered ? If so what should you have added to the mix to beat that ? While you're being attacked note: Was there anything i should have made earlier as defense based on my scouting ? Could I have easily blocked/impeded the route that the enemies troops took ? Did they take me unexpected ? Did i need to upgrade my troops more ? What upgrades did my opponent get ? Did superior range/attack or armor contribute to my defeat or my ability to repel the attack. If my TCs or archers had one more range would that have made a huge difference ?

As you watch the game try and also get a clear sense of the following: Should I have went more economy or military and when ? Should I have allocated resources differently ? If so when would I have had to start getting what I needed ? Did my opponents attack use more or fewer resources to achieve similar results as mine ? If fewer why ? Did my opponents overall strategy defeat mine or was it a failure in execution ? Finally, what was the most significant thing that I/my opponent did that ensured my victory/defeat ?

You've got to think of these questions and try to answer them all while looking from both players perspectives (like the computers pseudo-multitasking, only from one at a time, but changing them quickly 15-20 secs)

YOU see now why I DON'T Recommend fast forward :-)

Me and Armagedn play a random/random/random cases + get islands.. watch and try and pick out the mistakes and good parts on each team.. trust me theres alot there especially for me as it was getting late and I had just lost to a teut tcer and was a little peeved (note definitely flip between reveal/no reveal in this game.. .scouting was a major problem of mine)

Jihad and Armagedn
Article Downloads:
» Jihad vs. Armagedn