<< Back to General Forum .GrayedOutTextBox { color: Gray; }

 Posts 1 - 30 of 48   1  2  Next >>
Analysis: Which picks win?: 6/24/2015 13:58:26

smileyleg
Level 61
Report
```
ALL			Top50			DIFF
Picks	     #	Pick%	Win%	     #	Pick%	Win%	Pick%	Win%
1,2,3	 94650	50.9	49.7	  3201	39.7	48.7	-11.2	- 1.0
1,2,4	 25253	13.6	50.8	  1207	15.0	50.3	+ 1.4	- 0.5
1,2,5	 12200	 6.6	54.9	   805	10.0	57.1	+ 3.4	+ 2.2
1,2,6	  2030	 1.1	57.0	   127	 1.6	52.8	+ 0.5	- 4.2
1,2,7	  1373	 0.7	28.3	     6	 0.1	50.0	- 0.7	+21.7
1,3,4	 21478	11.5	50.9	  1120	13.9	50.4	+ 2.3	- 0.5
1,3,5	  5914	 3.2	49.5	   349	 4.3	48.4	+ 1.1	- 1.0
1,3,6	   785	 0.4	54.5	    67	 0.8	50.7	+ 0.4	- 3.8
1,3,7	   837	 0.4	29.4	     1	 0.0	 0.0	- 0.4	-29.4
1,4,5	  6029	 3.2	50.8	   358	 4.4	49.2	+ 1.2	- 1.6
1,4,7	   159	 0.1	35.2	     0	 0.0	-	- 0.1	-
1,7,7	   124	 0.1	26.6	     0	 0.0	-	- 0.1	-
2,3,4	  9855	 5.3	48.2	   461	 5.7	46.0	+ 0.4	- 2.2
2,3,5	  3757	 2.0	47.8	   263	 3.3	48.3	+ 1.2	+ 0.5
2,3,6	  1167	 0.6	50.5	   104	 1.3	54.8	+ 0.7	+ 4.3
2,3,7	   423	 0.2	29.3	     1	 0.0	 0.0	- 0.2	-29.3

```

Table shows for each possible combination of picks how often it occurred and how often it won. The first set is for all games, the second set is for the 4,000+ games between top 50 players. The last two columns are differences in Pick% and Win% between the two groups. If there is a "7" in the pick that means the player was randomly assigned a territory because they picked less than six.

Some of the numbers make for too small of sample sizes to say much.

Things of note:

* Top50 players get 1,2,3 10% less than the whole (makes sense)
* Getting 1,2,3 is a slight disadvantage (I usually hate it)
* Getting 1,2,6 is the best for the whole, and I expect it would be in the top50 with more games (you get your top 2 and know all 3 of theirs)
Analysis: Which picks win?: 6/24/2015 14:05:19

Ƨillynamenace
Level 57
Report
the win% shows how often the players getting these picks win the game, right?

~~~~~yeah it should be so. that's why you say getting the "1,2,6" is the best.

Edited 6/24/2015 14:06:32
Analysis: Which picks win?: 6/24/2015 14:16:52

Hennns
Level 58
Report
This show how important it is to make 6 picks, just look at this.

2/3/6 wins 50.5% of the time
2/3/7 wins 29.3% of the time

Similar low win rates goes for other combinations with random picks too:
1/2/7 28.3% win rate
1/3/7 29.4% win rate
1/4/7 35.2& win rate
1/7/7 26.6% win rate

Always make 6 picks

Also Smileyleg, can you do this for territories blockaded too, I love your statistics :D
Analysis: Which picks win?: 6/24/2015 14:25:32

knyte
Level 58
Report
Thanks for doing this! I love the huge bump in 1/2/7- looks like that level of intel has very high value, especially when combined with the ability to go for the FTB/two-turn bonus you wanted to start with.

The difference also seems to imply that one distinguishing mark of a good player is their ability to use intel gleaned from picks (and counterpick their opponents so the 1/2/7 combo is actually valuable instead of just a collection of random badly-picked territories which won't let them take full advantage of the situation).

What's also interesting is that this occurs in 1/2/7 and not in 1/2/6, which seems to imply a few things as well.

Edited 6/24/2015 14:28:21
Analysis: Which picks win?: 6/24/2015 14:36:32

Beren • apex
Level 63
Report
@Hennns, I think there's a bit of a confounding effect here though. Anyone who is any good will always make 6 picks, so you know off the bat that if someone got their "7th" pick, they're likely to be a below-average player. You'd have to compare the win rates of players who get their "7th" pick to the win rates of players who make less than 6 picks but don't get a random pick. Your overall message is correct though. Always make 6 picks!

@knyte, the win rate for 1/2/7 among the top 50 players is rather meaningless, since it only happened 6 total times.

Edited 6/24/2015 14:37:24
Analysis: Which picks win?: 6/24/2015 14:40:59

knyte
Level 58
Report
@Beren: Ah, didn't notice the sample size.
Analysis: Which picks win?: 6/24/2015 14:51:06

Nex
Level 60
Report
How exactly does one get 7th pick on the 1v1 ladder? I had assumed this was impossible.
Analysis: Which picks win?: 6/24/2015 14:55:13

knyte
Level 58
Report
7th pick = you got something you didn't pick, so you made 5 or fewer picks.
Analysis: Which picks win?: 6/24/2015 15:16:14

Kenny • apex
Level 59
Report
125 and 126 are usually lethal if you picked appropriately because your 5th/6th in this scenario is generally a counter. You have intel on 2/3 picks and a counter. Your 6th could even be a counter to a triple pick. So yeah if I get 123, 125, or 126, I know I'm probably going to win.

I disagree about 123 being a disadvantage. The reason it happens a lot less between the top 50 players is because both players can read the map appropriately. Getting 123 usually means your opponent read the map wrong and you have really nice expansion.
Analysis: Which picks win?: 6/24/2015 15:16:23

smileyleg
Level 61
Report
Thanks all!

Hennns: Sure, I can do a blockade one.

Forbidden: As knyte said, and I explained in the text :)

Interesting thing I just noticed: if you don't get your first pick you have a definite disadvantage unless you get 2,3,6 -- although 2,3,6 does quite well among top50 players (I'm usually happy when I get it). That's more of an advantage for getting first pick than I would've guessed.
Analysis: Which picks win?: 6/24/2015 15:19:39

smileyleg
Level 61
Report
Well getting 123 can also mean you read the map wrong.

If you get 126 and lose you've definitely made a mistake (it's happened to me and I knew it).
Analysis: Which picks win?: 6/24/2015 23:25:22

Widzisz • apex
Level 60
Report
```Picks 	#	Win%
1,7,7	124	26.6```

lol :D Losing after your opponent got 2 random picks must look lovely.
Analysis: Which picks win?: 6/24/2015 23:53:07

TBest
Level 60
Report
Interesting thing I just noticed: if you don't get your first pick you have a definite disadvantage

Fizzer made a thread about that, if memory serves me right. Well see if I can find it.

EDIT: I can't find it :( I think he said something about the player who gets 1st pick wins 10% more games. This is all memory based through.

I can be confusing it with the rating formula. There 1st pick is valued as 10 points.

Edited 6/25/2015 00:36:27
Analysis: Which picks win?: 6/25/2015 02:58:57

Cornucopia
Level 31
Report
Oh wow, this shit again? Try no one cares.
Analysis: Which picks win?: 6/25/2015 03:16:40

Donald Trump
Level 50
Report
^ Ignore this troll.
Analysis: Which picks win?: 6/25/2015 03:18:03

smileyleg
Level 61
Report
Why are you so sad Cornucopia?

You sure seem to care!
Analysis: Which picks win?: 6/25/2015 03:18:04

Nex
Level 60
Report
He's not in Kansas anymore
Analysis: Which picks win?: 6/25/2015 06:10:43

Dublin Warrior
Level 49
Report
what map? :)
Analysis: Which picks win?: 6/25/2015 09:13:31

DanWL
Level 62
Report
I always do at least 6 picks, to avoid randomness.

Cornucopia: Oh wow, this shit again? Try no one cares.

What happened to rule #1: You shall be respectful to other players at all times.?

Edit: Once (in https://www.warlight.net/MultiPlayer?GameID=8683194), I found out all 3 of a player's picks, on turn 0.

Edited 6/30/2015 20:40:28
Analysis: Which picks win?: 6/25/2015 09:34:58

knyte
Level 58
Report
I always do 7 picks, just to avoid the random one. I always new there was a slight possibility of getting a random one, but I've never had to use it.

I don't think that gives you any benefit. A "pick 7" scenario would be something like this:

Player A makes only 5 picks because he thinks he's read the map perfectly and that he should get at least 3 of his first 5 picks. Player B makes the same picks as Player A and ends up getting all 3 of theirs- so Player A is stuck with one random pick (a "pick 7").

Outside that, it's completely impossible to get a random pick if you make at least 6 picks each game. The only way to get a random pick is if you have an overlap of 3 or more picks with your opponent- the system doesn't just arbitrarily decide to screw you over, at least not to my experience. Yours might be different.
Analysis: Which picks win?: 6/25/2015 10:59:45

smileyleg
Level 61
Report
That's right, it is impossible to actually get your 7th pick in a 1v1 game with 3 starts. If both players pick the same 6 in order one of them will get 145 and the other will get 236.

Analysis: Which picks win?: 6/25/2015 12:20:16

Sabali
Level 56
Report
I think he said something about the player who gets 1st pick wins 10% more games.

Was this in games where both players picked the same pick as 1 or in all games? I would suspect the former.
Analysis: Which picks win?: 6/25/2015 14:05:28

knyte
Level 58
Report
You know, this data could totally change Warlight ladders/CLOTs. Bayeselo (which Fizzer uses or the ladders) does adjust for first-pick advantage using an experimental constant, but the one Coulon (maker of Bayeselo) derived came from chess- and I have no idea how Fizzer got his experimental constant.

So smileyleg's data can be used to get a good constant for first pick advantage. But it gets even better. Now we're even able to adjust for very specific situations (with more data, perhaps)- like getting 1,4,7 when your opponent gets 2,3,5. Or derive different constants for first-pick advantage depending on whether you had the same first pick.

If we had probabilities like that and were able to determine which changes were neglible and then go ahead and derive constants from the ones that weren't, well, that would totally change every CLOT out there and make the results much more accurate in the long run.
Analysis: Which picks win?: 6/25/2015 19:12:13

Sephiroth
Level 60
Report
DanWL, how in hell do you think there could be a possibility to get the 7th pick when the total territory limit is 6?

Edit: When I had to use 6 of my picks, my first 3 picks were all gone and I was left with picks 4-6.

bullshit, you ALWAYS get at least 1 in your first 2 picks and if you get your 6th pick you DIDN'T get your 4th and 5th

Edited 6/25/2015 19:13:04
Analysis: Which picks win?: 6/26/2015 02:13:36

Fallen Angel
Level 55
Report
@Dan, with 3 picks per player it is impossible to get 7th. Don't attempt to argue this one; just accept it as fact, please.
Analysis: Which picks win?: 6/29/2015 07:15:26

knyte
Level 58
Report
bullshit, you ALWAYS get at least 1 in your first 2 picks and if you get your 6th pick you DIDN'T get your 4th and 5th

Is there a non-cyclic way to handle picks? I.e., a purely random one where the system decides between P1 and P2 for each picking turn until one of them has 3 starts.

Obviously not on the Strategic 1v1 template or the 1v1 ladder, but does one exist? I haven't seen it used, ever.

EDIT: to clarify, not that I would ever want to use it

Edited 6/29/2015 07:16:04
Analysis: Which picks win?: 6/29/2015 08:07:43

TBest
Level 60
Report
"a purely random one where the system decides between P1 and P2 for each picking turn until one of them has 3 starts."

Well, there is automatic distribution. and that is the only option to manual :)
Analysis: Which picks win?: 6/29/2015 16:31:54

Sephiroth
Level 60
Report
Is there a non-cyclic way to handle picks?

There isn't. Even with random move order, once the game has decided priority the turn will happen in this fashion:
P1, P2, ... , Pn-1, Pn, Pn, Pn-1, ... P2, P1, P1, P2, ...

in a 1v1, it is always
P1, P2, P2, P1, P1, P2, ...
Analysis: Which picks win?: 6/29/2015 16:40:11

knyte
Level 58
Report
Ah. I almost want to try that, just to see how things would change now that I can no longer predict my pick combos (or at least rule out impossible ones like 1-2-6).
Analysis: Which picks win?: 6/29/2015 16:58:53

Widzisz • apex
Level 60
Report
(or at least rule out impossible ones like 1-2-6)

But 1-2-6 is possible.
 Posts 1 - 30 of 48   1  2  Next >>