Monday 05/11/2012, 11:25

In the coming weeks (hopefully starting this week) I will be running a little test. In the previous week Kalindra was in play and I noticed a ridiculous amounts of games where I knew I had the game won or lost after the 1st round.

I will be be tracking the following stats from my games:
Games played
number of games decided prior to playing a single round
number of games decided after the 1st round
number of games decided after 2 rounds
number of games that was determined on the outcome of round 3

Monday 05/11/2012, 12:33

Good luck. I hope your results will prove conclusive smiley

Monday 05/11/2012, 15:17

Nice experiment. Are you willing to post the deck(s) you are using for this? I think that has more impact on the data you're trying to collect than anything else.

Monday 05/11/2012, 18:05

Wait one sec, if the game is decided after the 1st round, your not the best player are you? Nah joking I like the experiment.

By the way the easy result that everyone here knows is that determined on the outcome of round 3 is going to be a huge one, 50/50's anyone? Yeah post your deck aswell.

Saturday 10/11/2012, 05:44

Decided to go with a deck that isnt able to make huge gaps except for Beeboy
Mafia Street Thugs

some decks have more capability to create insurmountable life gaps.... mono Jungo, anything with Kalindra
some decks will have less ability to create life gaps.. mono Montana that does not draw Avola, Roots without Gretchen

its a deck that doesnt really force too many 50-50s so i think its a good one to test. pussycat/Uranus decks force a lot more 50-50s than this


round - number of matches decided
0 - 2
1 - 17
2 - 23
3 - 12

most games decided after 2nd round. not that many 3rd round deciders... and by the way, 3rd round does not always mean 50-50, but a guess is often involved but the odds of a right guess differs.

lol games decided with no pillz played is awesome.... not

far too many games decided immediately after opening round.

the results above are by no means conclusive, but they are in line with the observations of many players. some decks will force more wins in the 1st round, some will not.

way more ELO than i wanted to play this week smiley

Saturday 10/11/2012, 06:24


I couldn't help but notice that the number of games decided by 1st round and 3rd round are looking pretty similar... Maybe that's not too bad. Excellent observation about 3rd rounds not always being 50/50s. I (ideally) try to force a situation that will allow me a favorable split in round 3 creating a win. If there are even 3 possible ways I can pill that make sense, and the opponent absolutely has to guess which I picked, I feel pretty good about the match, win or lose (as I managed to skew the odds in my favor.)

Proud of you for putting up with so much ELO! smiley

Saturday 10/11/2012, 06:51

No i dont think round 1 and round 3 look very similar...
17 vs 12.... factor in the 2 games decided with no pillz played

round 1 is stupidly important and round 2 has become the guessing/coin flip round

the amount of high gapping cards is pretty nuts... the addition of some very good 2 star cards has allowed people to load up on more nuking cards.

Jose Star, Jean, Treeman are all silly good 2 star cards

Saturday 10/11/2012, 07:22

Other than that I think that 1 and 3 are the most similar rounds in terms of how decisive for the game they are, I completely agree with you. You're right that the trend is towards great low-stars and huge nukes, and that this makes deck building, well... A bit dumber than it used to be.

But, let's put it this way: only about a third of the games in your experiment were decided in round 1. I don't think that's really more than one might expect of a game with 4 rounds, only 3 of which can be deciding rounds.

Saturday 10/11/2012, 08:49

0 rounds played - 3.7% of the games decided - this is ridiculous. this will never be good. This is as much fun as Ambre+Kolos with no DR/SOA

1 round played - 31.5% of the games decided. to say it is 1/3 of the deciding rounds and therefore it is ok... well, no this is not ok. this is a strategy game.. not a coin flip game where who ever wins the first one is the winner, and then we flip the coin 3 more times just for kicks and giggles. This is not high quality game play, this is why cards like Sigma, Sylth are banned, and Kolos and Smokey (when 12 life was the rule) were banned. Reducing the game to a 1st round gamble is not that much fun

2 round played - 42.6% of the games decided. This is probably about the right amount...

3rd round played - 22.% of the games decided. i think more of this should happen... fun as long as it is not 50-50 guess work i think smiley

Sample size was only 54 games. I will do it again next week maybe. I imagine the trend of opening rounds being decisive will hold true. If Kalindra is free next week, ill run her

Saturday 10/11/2012, 14:07

Quick question, though: Are you counting 1st turn resigns where your opponent still had a chance to win, but opted not to play, because that chance was low?

That's definitely something to take into consideration. A lot of good ELO players will resign in round 1 if the opening play has reduced their chances of winning significantly below 50%... That's because they can sit there and waste their time or move on to play 2-3 more favorable games that will offset that loss. This is the same "lose fast" strategy that you see in DTs, except that in ELO, there is virtually no penalty for a resign. What's true over the span of an hour still holds true over the span of a week.

That may have inflated your Round 1 numbers a bit. Not the fault of the game in those cases that things were decided so early.


Answer to this subject