I'm going to look at some ladder games between top players. My theory: In games between top 1v1 players, "luck" determines more games than skill.

"luck" = luck of picks, cumulative luck, first/last order luck, guessing luck

luck of picks = all things being equal, one player gets the better end of the picks when both make the same pick(s) in the same order

cumulative luck = attack/defense luck throughout the game

first order luck = all things being equal, one player gets the first/last move in a 50-50 situation

guessing luck = there is no intelligence, a player has no idea where the enemy is but makes a lucky guess

I agree only with first order luck.
Other cathegories are bullshit 1 and 4, or mainly irrilevant 2.

Anyway I partially agree with you since some top players' games are only decided by luck. Of course my statement is not appliable to szezi's games. In those games if he wins, it's only for his superior skills; if he loses, it's only for luck. I don't know why but if you will analyze those games you will agree with me:O

Game 1: Niko vs Sze http://warlight.net/MultiPlayer?GameID=3504533

Key moments in game:

- cumulative luck (turn 3): Sze missed 1 of 2 3v2s and didn't get W Africa's bonus, Niko is one turn ahead in growth
- cumulative luck (turns 2, 4-5): in 4 4v2 attacks, Niko had 3 standing armies remaining, enabling him to (a) take W US with 3 less armies and be able to deploy 3 more in Scand and (b) expand into and take Canada quicker and with less troops deployed
- guessing luck (turn 10): Sze guesses Niko has Ant (and Aust?), he is wrong (and taking down that wasteland makes a big difference in Niko's army advantage: from 34-30 turn 9 to 56-29 turn 10)...this could be considered a bad move (guess for Ant and give up Russia/Scand?), though the "luck" up to this point was decidedly in Niko's favor, possibly forcing Sze to make something happen.

Game is essentially over now. Who had better "luck" when it mattered (the earlier the "luck," the more its influence on the game)? Who won?

Oh wait I have misunderstood what you meant with cumulative luck, in that case cumulative luck passes from the cathegory "irrilevant" to the cathegory "bullshit".

Games to look at: start from #1, most recent.

Game 2: Sze vs Oliebol http://warlight.net/MultiPlayer?GameID=3549676

pick luck: 3-4 picks are the same. Sze got the 3, which is strategically better than the 4
attack luck (turn 3): Oli hits Tibet 3v2 and misses. Sze is taking E China and now knows Oli is coming, advantage Sze (however, Sze's pick of CA enabled him to flow better and take that bonus smoother and with less armies deployed than Oli in Ant -- if Oli had CA he could have hit Tibet 4v2, so his bad luck was caused by a weaker 5 pick)
until the end: Sze makes better moves (blockade, better delays, etc.)

Result: Sze has better "luck" and played better. Easy win.

Zibik vs Sze http://warlight.net/MultiPlayer?GameID=3485495

- pick advantages (skill): Sze gets intel of Zibik in Scand with 3 pick, while Zibik takes India with his 3 pick (slow to get bonus) <-- key to game
- skill: Sze delays and ensures he enters Scand no matter what (given Zibik's limited moves)
- attack luck (turn 9): zibik is far behind, so he tries to take W Africa on the cheap with 3 3v2 attacks, 1 misses

Game over. Skill won.

Tim vs Oli http://warlight.net/MultiPlayer?GameID=3550688

nothing notable until turns 4 and 5:
- turn 4: Oli could have delayed to at least match Timi's delays, he doesn't and Oli kills the neutral in China (a minor point that would have saved Oli 1 army and cost Timi 1-2 armies on the neutral, giving Oli more standing armies -- net gain of 3-4 -- to attack with next move) <-- hard to guess
- turn 4 (cumulative luck): Timi makes 2 4v2 attacks in Indo, both result in 3 standing armies (ideal result, quite lucky)
- turn 5 (cumulative luck): thanks to Timi's luck turn 4, he can try to take Indo on the cheap with 2 3v2s (which work)
- result of turns 4-5: with normal luck, Timi is not able to get Indo on turn 5 and Oli busts Indo or at least forces Timi to deploy close to 100% in Philippines...instead Timi keeps Indo and gets a 17-12 income advantage.

Game over. "Luck" wins.

Art vs Timi http://warlight.net/MultiPlayer?GameID=3553273

pick luck: Art and Timi have the same 1st pick, Artham gets it (skill: and Timi's 2-4 picks aren't as good as Art's 2-3)
cumulative luck (turn 4): 3 3v2 attacks drop in Greenland and Art's income goes from 8 to 17 in one turn (W Russia also falls with a 4v2 and 2 2v2s)
cumulative luck (turn 5): Timi tries 3 3v2s to take E Africa, one fails (net effect of these 3v2 attacks turns 4-5: Art +5, Timi -4)

Game over. By the time the skillful fighting happens turn 6, Art already has insurmountable advantages in standing armies and an income advantage of 17-12. "Luck" gave Art these advantages. Early skill in picking gave him minor positional advantages. And Timi's India pick was misused, to ensure turn 6 would be as dire as it was.

Timi vs unknown http://warlight.net/MultiPlayer?GameID=3471141

- pick luck: same 4 pick in Scand. unknown gets Scand, Timi gains intel advantage (2 known picks of unknown's vs 1 known pick of Timi's) and has it countered in Ukraine
- turns 2-3: Timi jumps the gun in Scand (bad guessing luck) and misses 1 of 2 3v2s to take E China, advantage unknown, who is able to take W Africa with full deployment two turns in a row, all with 4v2s...thankfully for Timi, unknown doesn't press the advantage and by attacking to see SA, unknown's initial advantage is lost due to (a) lack of aggression (though unknown didn't have the intel to be sure about making an aggressive move) and (b) Timi's intel on SA
- turn 6 (guessing luck): Timi successfully blockades SA on the cheap when unknown could have gone for the jugular (by neutralizing CA or even taking SA and then neutralizing CA)
- turns 4-8 (cumulative luck in CA/SA): bad luck for unknown turn 4 (4v5 loses 1 more than normal), unknown gets 1-2 back with good luck 17v17 the next turn, 2 4v2s result in 3 standing armies each, which enables Timi more flexibility in his deployment to defend CA and contest Scand (the difference between a possible 22v14 and 22v15 is 33% vs 2% success)
- turn 10 (guessing luck): unknown guesses Timi doesn't have a priority card and instead of delaying and attacking where Timi will go, unknown attacks where Timi was
turn 10 (skill): blockade of 42 is overkill, not letting Timi waste armies on killing off half of unknown's stack in SA leads to a shift in army strength from turn 9 (from 37-32 advantage to 38-42 disadvantage)
- turn 11 (defense luck): Timi gets nice defense luck that saves 1-2 armies from dying on unknown's 28v23 attack. This enables Timi to hit Scand 5v3 instead of 4v3 or 3v3. Thus, instead of unknown gaining parity in income and have more armies in Scand/Russia, Timi ends the game.

Game over. More luck when it mattered for Timi. Overall skill: slight advantage to Timi (unknown's blockade and low risk early/in SA when Timi blockaded it). Deciding factor: "luck."

Melan, feel free to continue looking at the recent games for players ranked #3 through HHH to analyze games between top ladder players (those once ranked #1 or #2). I believe the results will be the same as the games I just looked at. My guess...

Luck (L) vs Skill (S) as basis of victory for more or less equally skill players:

Majority of games (50%-65%): L
Some games (10%-15%): all S
Some games (20%-30%): L and S both contribute, unclear which is most important
Some games (5%-15%): winner has L and S advantage
Some games (5%-15%): winner has S advantage and L disadvantage

If you disagree, though, show the data. I'm saying the above based on about 1000+ games against top players in 1v1s. And I showed a handful of random games to support my ideas. Where is the basis of your "bullshit?" A misunderstanding of definitions (which are deliberately unclear/simple because the games below help define them)? A limited understanding of why games are won or lost?

To further define luck: "luck" that happens away from where the action is doesn't really matter. When and where the "luck" happens is essential to any discussion of how "luck" influences a game's outcome. Also, if you think my "accumulative luck" is the same as the statistics for members that are called accumulative luck, you are mistaken. I simply call it that because it needs a name. "Attack/defense luck at key moments" could also be its name. The statistics tool's accumulative luck can be misleading: all attacks/defenses are equally measured, regardless of strategic importance, location and number (eg, is good luck on a 4v2 or 2v1 attack equivalent to having good luck on a 20v13 or 6v4 attack? according to the stats it is; in terms of effect on the outcome of the game, they can't be compared).

As for my game with Timi - in my humble opinion the key was that Timi as you said misused the India pick. Furthermore, he failed to play an order delay on turn 6. Had he done that, he would regain at least income balance. I still would have had more standing armies though, but that not necesairly would have to be decisive.

As for picks - I was thinking about that actually quite a lot here, mainly becouse we both had the exact same picks, but in diffrent order. I also assumed the scenario, where I didn't get the first pick and in my opinion that would have worked equaly well, so in general I think my picks were simply superior to Timi's and that combined with Timi's mistake (in my opinion) in SEA was the decisive factor in the game.

best regards.

PS. I did of course have quite a bit luck with capturing Greenland, but I don't consider it to be decisive.

Sorry for double-posting, but I've just thought of a thing concerning this topic.

How about instead of just pointing out the games, where luck was an important (if not decisive) factor, we try to simulate what moves (great skill) could offset that luck. In other words, how could an omnipotent WL player could have won them?

in conclusion, gui/myhand/szew and all those are consistently lucky most of the time, while really SKILLED players like me get unlucky too much

Between 2 top poker players luck decides who is the winner.
I need proofs to say that this statement is bullshit?
All those 3 cathegories of "luck" could enter in a cathegory of skill: 1=being able to pick (when I pick I always consider as a possibilty the situation in wich my opponent does my same picks, f.e.). 2=luck management. 4= being able to make a good attack.

I wouldn't exactly agree that a game between two top poker players is decided by luck, but it is irrelevant as poker is nothing like stratigic 1 v 1 settings. I elaborated on this matter in another topic, so no point reapetin myself.

if the opponent takes the same picks and the picks give one player an advantage then luck is clearly an important factor, since both players were equally skilled at the start, there's no argument about that

And how often does that happen? Same picks meaning also same order of picks obviously.

Id say no more then once in 100 games. That's hardly a statistically signifcant amount. You can't make a generalisation based on something that's less propoable then error margain.

Warlight, what Artham and Trollussa are saying is that you need to craft a strategy that works in worst-case scenario as well.

- turn 11 (defense luck): Timi gets nice defense luck that saves 1-2 armies from dying on unknown's 28v23 attack. This enables Timi to hit Scand 5v3 instead of 4v3 or 3v3. Thus, instead of unknown gaining parity in income and have more armies in Scand/Russia, Timi ends the game.

28 kill in average 28*0,6=16,8 and unknown killed 17, nothing irregular. Good defense-Luck only matters about the troops the attacker loses, not about the troops that survive (that would be negative attackers luck!)



About my game against Artham, i agree about my mistake in picks, i realised it when i saw what i received, but that was too late