<< Back to General Forum   Search

Posts 1 - 27 of 27   
How to Improve: 3/5/2014 15:30:48

Qi 
Level 55
Report
1. Beginner: Play anyone anywhere; and think, ask questions, self-criticize. Learn the mechanics and some basic approaches to the game.

2. Intermediate: Watch what better players do. Look for patterns. If you don't know why they did something, ask (yourself or them) why and look for answers. Pay attention to what weaker players do too. Sometimes, in their madness, they have innovative or useful strategies. Usually, you can learn what not to do. You can learn from the weak and the strong.

3. Advanced: (a) When you have stopped improving, play stronger players or players with different styles. Play to win. And watch the history afterwards to see what you and the opponent did well and not so well.
(b) Watch the games of top players. Look for patterns, strengths, weaknesses.

4. Cream of the Cream: (a) Keep sharp by playing good players.
(b) Invite the best players to a real-time game. Play to get in their head. Focus entirely on what you think they are thinking. Don't try to win. Just try to get in their head. Consider it a form of meditation. Then, after you invariably lose, go back and immediately look at the history. Since you spent the last half hour trying to get in their head, reviewing the history should offer greater insights. I do this sometimes and it is more rewarding than simply playing to win or watching a game I didn't play. By trying to get in the opponent's head, it forces me to rethink the game's logic as if I were playing for the first time. It also might help for future games: understanding the enemy's logic now gives you insights in future games, and showing the enemy a weak game could throw off the enemy's guesswork or predictions in future head-to-head games. If you play your best every time, the enemy can understand your logic better.
(c) Experiment. Try out new ideas, approaches, strategies. Even if you lose, at least it was fun and different.

Edited 3/5/2014 15:36:29
How to Improve: 3/5/2014 15:57:26


ps 
Level 60
Report
or you can just scratch all that Qi wrote, make a custom template on an unknown map and only play there.
How to Improve: 3/5/2014 15:58:12


Unreality 
Level 50
Report
very true

and i have a few more suggestions for intermediate+ players

Many times it's about choosing between
a) Making your best move
b) Countering enemies best move
c) Countering opponent's move that counter your best move
d) and deeper like inception... counter the counter that counter...

one or two critical moment of doing the right thing could win you the game, so in order to increase your chance of success:
Analyze your opponent's battle history, ppl made mistakes, ppl have bad habit, ppl usually follow a pattern vs intermediate players (top vs top they could do risky moves)

It is always the best thing to try to list out the possible moves of your opponent first, then have an idea of the chance of happening of each of his possible options (based on your study to his battle record), then deduce a best move that could counter most of his highest chance options

Like, most often ppl wont hit the most obvious place that you will guard, and might go thr a longer safe route to find your weak point, which means sometimes you could just give up defending that 'most obvious place' and then ambush his longer 'safer' route.

Counter is the key, and could give you the highest reward, but that depends on how well you know your opponent (from his history). If you failed to predict... you will be forced to make your own best move... which could be easily countered.

Edited 3/5/2014 17:21:55
How to Improve: 3/5/2014 16:09:03


ChrisCMU 
Level 60
Report
I was just talking about this with a clan mate yesterday. The biggest thing IMO is after you play for a bit, you MUST make the transition from %75 luck games. You need to play more games/tournaments where the settings are %16 luck or less, have advanced attacks (attack only, transfer only, by %) and are not FFA or Multiattack. I am not saying never play these, but if they make up the bulk of what you play, your growth will be stunted A LOT. High luck games do not reward strategy like low luck games do.
How to Improve: 3/5/2014 16:21:08

Qi 
Level 55
Report
I played half my games on 75% luck. I think there is definitely a skill involved to how you manage the luck. But most people make attacks on 75% luck as if it were 16% and then complain when they fail. Well, there are ways to deal with this so it doesn't turn you off from 75% luck.

(a) Don't waste armies if failure means your positioning or situation changes too much. (The opposite is also true: be more willing to waste armies going for bonuses if you can afford it.) (b) If taking a certain territory is essential to the game, use more armies. Doing a 4v2 or 5v2 when that territory is crucial is simply poor risk management. (c) If you get seriously screwed by luck and lose, get angry and really go with this anger. It was exciting, right? Now go back and play again. Soon, the enemy will get just as angry. It feels great to watch him get so worked up over the game's luck, right? (d) If you lose to someone you have a bit of a rivalry with, blame the luck!

Edited 3/5/2014 16:23:59
How to Improve: 3/5/2014 17:09:00


Mr. Gentleman*SEAHAWKSWONSUPERBOWLXLVIII*
Level 58
Report
Great post guys! I will head your words of wisdom and seek to grok your thoughts!
How to Improve: 3/5/2014 17:13:05


Mr. Gentleman*SEAHAWKSWONSUPERBOWLXLVIII*
Level 58
Report
-------Math failure------

forums 1 Gentleman 0

-------Math failure------

please disregard my input.

Edited 3/5/2014 18:37:17
How to Improve: 3/5/2014 17:30:32


Richard Sharpe 
Level 59
Report
Gentleman, your math is WAY off.

The odds of succeeding with five 90% attacks is 59%. The odds of succeeding with any four (thus having one failure) is 92%.
How to Improve: 3/5/2014 17:46:35


Mr. Gentleman*SEAHAWKSWONSUPERBOWLXLVIII*
Level 58
Report
yeah i was just guesstimating it, point was not expecting a failure is fool hardy. but thanks for your accurate maths!

Edited 3/5/2014 17:47:38
How to Improve: 3/5/2014 17:53:43


[WM] ᵀᴴᴱ𝓕𝓻𝓲𝓭𝓰𝓮 
Level 59
Report
Wow. I don't feel comfortable telling that - but in your case - sorry, but revision of BASIC probability maths is the only way you can improve. This game involves so much luck and probs management, that with your current understanding you have absolutely no chance for playing at a decent level..
How to Improve: 3/5/2014 18:30:03


Mr. Gentleman*SEAHAWKSWONSUPERBOWLXLVIII*
Level 58
Report
lol jesus I had other things going on. just put the first thing that crossed my mind. I recognize I might not be as math oriented as others, that much is true. However, there are tools to compensate for that such as calculators etc.

Now I may not be good at number crunching but I am great at abstract thinking. which is also an important factor.

Sorry for the bad maths I do apologize. May the math gods forgive me!

I am an artist that is what I do for a living. So I will compensate for my poor number crunching using tools provided and I will maximize other attributes that can be beneficial.

Now I would be happy to duel you and learn from you since clearly you are the superior player

Also my wife has a genius level I.Q. went to college at 16 got her masters in quantum physics is a God damn human calculator and she sucks at this game. I mean awful, it drives her nuts. Sort of why we work we both think differently.

I think it is important to recognize your strengths and weaknesses, so you can identify areas in which you need to compensate for.

Again I apologize for the laziness in my post. Sometimes I lack the focus to follow through with the details. However, there are people on this forum that are not and my point was validated with a more accurate answer. I appreciate that.

Edited 3/5/2014 18:35:37
How to Improve: 3/5/2014 18:31:39


MusicManiac
Level 54
Report
Guys, is there any way to delete messages? so there would be more free space?)
On forum, i mean)

Edited 3/5/2014 18:31:48
How to Improve: 3/5/2014 18:50:51


Taishō 
Level 57
Report
Richard Sharpe, show your work if you're gonna make claims of that nature.

The chance of a coin landing on tails is always 50% regardless. If you flip it 100 times it will never change, but always remain 50%.

So please explain to me why 5 90% moves have a 59% success rate? Each move has a 90% chance for success, the second-fifth move's chance for success doesn't change with the first moves failure or success.

Edit: Gentlemen!! Change yo' name! Also, if that's your wife in your picture that would make her hot and smart. Nice combination, bro.

Edited 3/5/2014 18:53:14
How to Improve: 3/5/2014 18:58:12


Richard Sharpe 
Level 59
Report
Hauptman, it is called binomial distribution. Pretty simple stuff really.

The odds of a 90% action succeeding 5 straight times is 0.9^5 = 0.59049

By no means did I claim that the result of one trial impacts the probability of any other trial. That is just foolish. Instead I showed what the true probability of each case succeeding was and then moved on to the odds of having at least 4 of the 5 succeed.

Since I am feeling generous, here is the math for getting 4 of 5 successes:
5*[(0.9^4)*0.1] = 0.32805
Thus getting 4 or 5 successes has a 0.91854 chance of occurring.

Even more generous! Here is a good link to perform such calculations:
http://stattrek.com/online-calculator/binomial.aspx

Edited 3/5/2014 19:14:08
How to Improve: 3/5/2014 19:09:54


Mr. Gentleman*SEAHAWKSWONSUPERBOWLXLVIII*
Level 58
Report
Thanks for the posting the math Sharpe.

This was sort of the point I was trying to make earlier. That expecting all 5 out of 5 attacks to succeed while possible is unlikely. However is would be reasonable to assume that 4 of 5 would succeed.
How to Improve: 3/5/2014 19:31:33


[WM] ᵀᴴᴱ𝓕𝓻𝓲𝓭𝓰𝓮 
Level 59
Report
Hauptmann - pretty simple is an understatement really..

Of course every single attack will remain 90% plausible. The past attacks don't change the upcoming probability.

But the overall probability of ALL or ALL BUT ONE, or any given combination of outcomes is not equal to the probability of a single event happening.

If you have two 90% attacks, each of them will have prob. of 90%, but:

probability of BOTH succesful would be 90% of 90%..
in maths:
0.9 * 0.9 = 0.81 or in other words 81%
on the other hand
probability of first one succesful and the other one failing would be 90% of 10% ....
0.9 * 0.1 = 0.09 ...... 9%
futhermore - probabiliby of ONE succesful and the other failing...
0.9 * 0.1 + 0.1 * 0.9 = 0.09 + 0.09 = 0.18 .......18%
probability of both failing:
0.1 * 0.1 = 0.01 ..... 1%

so we have:
81% for both attacks successful
18% for one succesful, one failed
9% for first succesful, second failed
9% for first failed, then succesful
1% for both failed

going further:
99% for AT LEAST ONE succesful
19% for at least one failed
etc..

the above was a simple calc for two equally probable attacks. It's not much different for 5..

all succ: 0.9 * 0.9 * 0.9 * 0.9 * 0.9 = 0.59049 ......= .59.049%
none succ: 0.1 * 0.1 * 0.1 * 0.1 * 0.1 = 0.00001 ..... = 0.001%
exactly 1 failed 5 * (0.9 * 0.9 * 0.9 * 0.9 * 0.1) = 5 * 0.06561 = 0.32805 .... = 32,085%
AT LEAST ONE succ 1 - 0.00001 = 0.9999 .... = 99.99%


the fun starts when you have combinations of different probability attacks, where you attack single teritorry, with a couple of teritorries you NEED to take and a couple you would benefit from taking, but are not that necessary.. then guesstimating or counting skills start to matter..
How to Improve: 3/5/2014 19:38:50


Taishō 
Level 57
Report
Fridge and Sharpe, I'll be badgering you two more come September when I start graduate school.

This is where theory and practice clash I guess, but of course it helps with micro management in important games, much appreciated.
How to Improve: 3/5/2014 19:45:10


Richard Sharpe 
Level 59
Report
It should be noted that the exact same mechanics are utilized in determining the success of any given attack or defense. Just change the 0.9 to 0.6 or 0.7 as needed. Of course, luck then factors into the equation but I'll leave that to you to figure out.
How to Improve: 3/5/2014 23:30:33

Qi 
Level 55
Report
I have never calculated the probability of multiple attacks. I estimate. It gets me close enough. Risk management doesn't require exact calculations sometimes.

Edited 3/5/2014 23:50:42
How to Improve: 3/5/2014 23:50:48

JSA 
Level 59
Report
One other important thing to improving. I once remember Trollussa saying in a forum post about how if you want to improve, you must first remind yourself you are a noob and can still learn. It's nearly impossible to improve at a very high level unless you have a humble heart and realize you make mistakes. Be open to criticism and accept it. Once you get good, don't believe everything an instructor tells you. Work it over in your own mind and then decide if it is true or not.

The very top players have reasons they are the best: Sze is a human calculator and perfects every move. Gui and Gnuffone aim to get in their opponent's head (Gui more on play, Gnuffone more on picks). Dead piggy has played so much against great players and has experimented; he's had lows but when he is on a high, he is nearly invincibile.

Those are quite probably the 4 best strategic 1v1 players in the game. They each have a different way of doing things, but they all have very effective styles that work for them. The trick is to find what kind of player you do best as, and changing it to fit yourself better. I once memorized Szeweningen's picking style and could pick his picks as well as he could. But unlike him, I did not have the high level of play he did at that time. I am now much better because I've learned how to play better, although I lack the focus needed on picking lately.

If you want to do well, focus as often as possible. Focusing only sometimes is a much slower route to improving, as I have figured out :p

Set a goal. For instance, my goal now will be to memorize one of their picking styles and perfect it, and focus on every game I play.

The ideal player in my opinion: Mix Dead piggy's total mastery of Medium Earth with Gnuffone's counters with Szeweningen's safety. Get in your opponent's head like Gui does. Vary up your style so you are not too predictable. Study these player's picks, and perfect all of theirs.
How to Improve: 3/6/2014 11:49:54

Widzisz • apex 
Level 60
Report
I believe 4 c) in Qi's post should be implemented earlier. At stage 2 imo.

Why all Your points besides this one seem to say "look at others"? Is this intentional or no?

I don't believe that's all. I believe it's better to figure out some things on your own, or at least try to.
I mean, instead of constantly looking at what other players do, try to use your imagination and think what you can do (at least from time to time).

Yea, it may take a longer to get better, you fail more often, and it may seem a bit pointless to re-invent the wheel, but I think it pays off in deeper understanding of the game, and later you rely on your own ideas, not someone else's.

So, I would add somewhere around point 2: "Think and test your own ideas sometimes.".
How to Improve: 3/6/2014 13:13:42

Qi 
Level 55
Report
I simply mean that you develop what works for you in stages 1 amd 2. In stage 3, try to pay more attention to learning the styles of others. In stage 4, you should understand enough styles to begin more radical experimentation to try to figure out styles and strategies that nobody knows. Stage 4 experimentation is meant to push the realm of possibility and challenge a good/great player to reinvent himself and the game.

This is why I don't mention experimentation earlier. Earlier experimentation is simply learning the ropes of what most people recognize as good play. Stage 4 experimentation is Isaac Newton alchemy.

Edited 3/6/2014 14:11:32
How to Improve: 3/6/2014 14:03:06

Widzisz • apex 
Level 60
Report
Ah, ok. So I just misunderstood a bit, and apparently there is "think" in 1st stage. Should be in 2nd too though :P
Still, experimentation as a way to grasp the basic doesn't seem as significant as it should be, at least to me.
How to Improve: 3/6/2014 14:18:38

Qi 
Level 55
Report
True. I think when I wrote it, I had certain assumptions about the readers: inteliigent enough to make it to stage 4, and clever enough to figure out the basics.
How to Improve: 3/6/2014 18:54:56


Despicable
Level 48
Report
Hi I've read most of what you guys said, its been improving my game. I would consider myself an intermediate player, for which is why I need you guys to help me be better. So I would like to play some games with the expert players, multi-day no boot practice games on any map you guys want. It would be great if you played with me, I really want to learn the game to its highest potential. please mail me or reply here if you would like to help me and the map in which to play in. Also tell me if it will be a team game or FFA.

Edit:
I have a 35% win rate, which is low for you, but the last few games I've played in I've won most of them thanks to your tips.

Edited 3/6/2014 19:02:29
How to Improve: 3/7/2014 12:49:26


professor dead piggy 
Level 59
Report
"Stage 4 experimentation is Isaac Newton alchemy." - Qi

This is my absolute favourite. What kinds of things have you seen that you consider stage 4? For me it was tiny blockades. Teddy was the first big name that i noticed who made tiny irritating blockades in semi unimportant territories. He then kept pressure up somewhere else and never gave his opponent 11 free incoem to break a blockade of 7 until the game was alreadd decided. Very cool and fun.
How to Improve: 3/7/2014 13:26:29

Qi 
Level 55
Report
Impaller did this sometimes, if he even bothered to blockade.
Posts 1 - 27 of 27