yeah obv i usually go for what i think is the best play, but you win like 2 DCs a week, if you went 6 months without winning (like me) you'd eventually start doing random stuff to try and break the mold, every frigging day i do the frigging normal dynasty strat with whatever comes up in the first couple of racks every time i put my whole frigging mind on how to play each rack sometimes i've sat there for 2 minutes each rack thinking about the play, where to put what and every frigging day someone just does better no matter what i do, someone does better. I start with 5s major chain because i get 3 5s in the first rack, i don't get a 5 for the rest of the game, I look at other's games and they just discard all the fives from the start as if they know that there isn't fives it's so ****ing confusing I have no idea how to be so frigging psychic...

thanks for the help btw, though i am really bad at reviewing my games as i do not know in my head what the racks were/what the options i had at the time was etc, I know i placed that 4 there for a reason, nothing i do is ever random or unthoughtful, it's just i can't remember what the reason was at the time.

it's worth noting that winning 10% of the DCs is an amazing record, I think i'm winning fewer than 1 in 20, so if you're comparing yourself with the best every day, of course you feel rotten. If you're winning 1 a month, you're doing ok, i think, so a goal of just winning one is pretty reasonable, but would also be decent achievement.

I think my basic opening strat tends to be:
* try to fill three of the outer chain squares with a single number asap
* only fill the fourth one with the same number if a) i'm playing 6s round the outside, or b) there are far fewer of the third chain (i.e. the potential minor outer chain).
* inner chain one of 5s and 6s
* don't commit to playing 3s until there are boat loads of them, but retain the ability to connect them all if i do then commit
* perfect the bonuses

I never play 2s on the sides any more unless forced, I'd much rather play a 2 in the centre square than a 1.

I've done a small analysis of your DC game from yesterday, how I would play it from round 5 onwards. Can't upload it atm as imageshack is being crap, will post tomorrow. Obv. I don't want to give away too much, especially to Gaby, Dynasty and TH who all read this thread, but I don't think the analysis will give away any of the things I think I have an edge over them on. So I don't really want to start dissecting my games, but w/e as a one-off, and the game plays out quite nicely but is interesting enough to be useful.

It might be useful though if, when you play the Daily Challenge tomorrow, you took screenshots of each round, and posted with reasoning, and we could maybe point out a couple of errors. Hopefully the Challenge will be relatively normal and not abs. mental like a few have been recently.

Donk, don´t try and compare yourself with Paul, his stats are ludicrous and he´ll give Dynasty a run for his money, mark my words.

But taking screenshots after every rack is a good idea to try and find your leaks. First try and find them yourself, after that show them here.

The mostly likely conclusion will be that you are a scrub

Today my score will not be enough for top10 I guess :

Ending my streak of 1-2-3-1 places

My setup after 4 racks was awesome though and could easily have won.

i'm going for back to back wins today. If my score doesn't stand up, i'm calling rigged.

individual player pages don't seem to be working

Percentages that a player wins the DC he plays. In (parenthesis) the amount of games played. I have looked up the highest ranked players and the players with highest wins.

Paul101 13,22% (174)
Derwi 11,50% (226)
Dynasty 9,92% (484)
Radisto 9,60% (177)
krohn 9,48% (211)
Gaby 9,19% (370)
TH10 9,03% (144)
Alex 8,98% (490)
mrgibson 7,88% (406)
Zac 6,68% (464)
bingobango 6,21% (161)
Sugartastic 6,10% (344)
anton-09 5,75% (435)
Shadowlike 5,65% (283)
Popcorn 5,48% (383)
dogfloss 4,50% (467)
Luke 2,56% (156)

So this is the DC from the 8th, Zac gave me the exact racks. There's no really tricky spots imo, but it plays out in a way that I can point out a few things. Gonna start with your setup after Round 4:



So this is pretty damn good, a couple of small things I'm not a fan of, but nothing major, and it ends up being the nut layout. From hereon, this is how I would play it. I'm not gonna give every reason for everything, but just point out some main things.

Round 5:

12455



So the key here is to notice that we put the 5s around top right instead of bottom left. Normally we should go for bottom left (as I did in my game), but because your 3s setup is so good, we can go around top right without blocking the 3s, leaving the bottom left open to possibly snake 3s or 5s later. Like I said earlier, top and bottom zone contain both a 1 and a 2, so we put the 1s and 2s in there first.

Round 6:

12446



This round is pretty self-explanatory, ideally we'd like to put the 6 in the bottom zone so we don't block the 3s, but we put it in the left zone instead because of bonus considerations (you have to put the 2 in the top zone, meaning you have to put the 6 in the left zone). Blocking the 3s around top left is not a huge problem anyway.

Round 7:

11256



So Rounds 3 and 7 are generally the most important rounds, because you normally have to take some sort of gamble, and work out which is your best one. However, here, playing it correctly, we can leave virtually everything open. The 2 in the left zone is trivial, as we want it there and definitely don't want anything else there. The key considerations now are that we want to 4-zone the 5s, place the 1s not next to another 1 and avoid blocking our 6s. The 3s are less important, but if we can avoid blocking them too then great. Normally you'd have to sacrifice at least one of your priorities, and do some calculations as to EV and chance of success, and perform a balancing act, but here we can place everything so everything is left open. A quick thought tells you that this is the (only) optimal solution. The 1s must go in the centre and right zones to avoid blocking the 6s. Then we can place the 6 in the bottom zone and connect the 5s via the top zone, and luckily we can achieve all our aims.

Round 8:

46666



Again, an easy round, the only thing to note here is that if you're awake then you can luck into slightly improving your third chain while connecting all the 6s.

Round 9:

12345



We get a great score and come joint second in the DC that day, but had the best setup going into Round 9.

So while it might seem that the top players are psychic, really what's going on is that we're just leaving our games as flexible as possible so that we can adapt to lots of different racks well.

Hope I haven't just absolutely owned myself with this analysis.

*blushes*

Your setup that day is not MUCH better than mine. I will full snake my 6s earlier than you. If less than 4x6s show up, I beat your 6-chain.

6s came flooding, but fortunately for me 3s did not, and I could 3-zone my 4s quite easily.

That was Donk's setup after Round 4, not mine btw (mine wasn't good). But your one might well be better than his. Either way, no humongous mistakes I wanted to comment on. I made an error though, you (Gaby) had the best setup going into round 9, Donk should've had the second-best though.

i've still got 4k on the field.

Not any more.. Apparently Paul is a good teacher!

I wrote a play by play commentary to today's DC, hopefully there will be some leaks that can be worked on, though I doubt it as I played it really well imo and got a very good score even though kokiri still sniped first (I'm looking forward to seeing kokiri's game). I'll post it tomorrow.

I think writing down all my thoughts really helped me, I found that is the way I improve at things is to write things down.

Also FUUUUU kokiri, I saw my score and my set up and I was like "ive got to win this one" then I saw your score and I was like FUUUUUUUUU. This was by far my best showing at a DC ever.

holy aboutfacements batman, NH.

I think I did one thing that was contrary to the schema i wrote above that was arguably a bit dubious, but which came good in the later rolls.

uuhmm you won one ....

Yeah but like 19 other people joint won, it's a little tainted lol.

I have been doing some calculations for outer chain strategy.

Suppose you have set up an outer chain of 3x6s in the first 4 racks, which is a pretty standard play, if you follow Dynasty strat.

In Rack 5 you get one 6, which is easy to place.



Now in Rack 6 you get another 6, which you will probably place like this (POSITION 1):


Today I want to suggest another strategy, which is to place like this (POSITION 2):


The first placement is the natural placement and allows you to create a full chain if 2x6s come in the racks 7-8.
The second placement is a bit risky and could possibly have you end up with a broken 6-chain as a sacrifice to maximize possibility to get a full chain of 6s.
If only one more 6 arrives in rack 7-8, you place like this:

And you hope to get a 6 in rack 9 and complete a 5-zone chain, while "normal" strategy would have created a 6-chain covering only 4 zones.

Let´s compare POSITION 1 and 2.
If rack 7 and 8 produce 2x6s or more (51.55%), both strategies will win and 5-zone the outer chain.

If rack 7 and 8 produce 1x6 (32.30%), POSITION 1 will yield a 4-chain of 6s. POSITION 2 will depend on rack 9 to produce at least one 6, make a full chain and beat POSITION 1.
Probability for at least 1x6 in rack 9 = 59.81%
So:
POSITION 1 will win 0.3230*0.4019= 12.98%
POSITION 2 will win 0.3230*0.5981=19.32%

If rack 7 and 8 produces 0x6 (16.15%), POSITION 1 will yield a 4-chain of 6s, while POSITION 2 will depend on rack 9 to achieve the same result.
POSITION 1 will win 0.1615*0.4019= 6.49%
POSITION 2 will tie 0.1615*0.5981=9.66%

So:
POSITION 1 will win 12.98+6.49=19.47%
POSITION 2 will win 19.32%
And there is a tie 51.55+9.66=61.21%

The strategies are practically equivalent. If the second 6 appears in rack 6, you can follow either strategy.
Actually, in the daily challenge, you should choose POSITION 2, because you will have a better chance to outplay the field who will probably choose for POSITION 1.

************************************************** ****************************************

Now suppose a slightly different scenario: the second 6 comes in rack 7, not in rack 6. Still we compare both strategies, but now with only two racks to come.
If rack 8 produces 2 or more 6s, there will be a tie = 19.62%

If rack 8 produces 1x6 (40.19%):
POSITION 1 will win if 0x6 appear in rack 9 = 0.4019*0.4019=16.15%
POSITION 2 will win if any 6 appears in rack 9 = 0.4019*0.5981=24.04%

If rack produces 0x6 (40.19%)
POSITION 1 will win if 0x6 appear in rack 9 = 0.4019*0.4019=16.15%
POSITION 2 will tie if any 6 appears in rack 9 = 0.4019*0.5981=24.04%

So in this scenario:
POSITION 1 will win = 32.30%
POSITION 2 will win = 24.04%
There will be a tie = 19.62+24.04=43.66%

So in this scenario, the risky strategy is inferior (but still very playable if you aim for a DC win). If the second 6 appears in rack 7, you are better of following standard strategy.

Of course all this is looking at the position in a vacuum: other things might influence the probabilities (bonus, other chains etc).

For example if your board after 5 racks looks like this:


And you get a 5 and a 6 in rack 7, you might even consider placing like this:


And 5-zone the 5s while having decent possibilities to 5-zone the 6s as well.

so in between writing and posting my analysis i apparently logged in (which is a lie) and now I've lost my many paragraphs and pictures analysis. so tilted right now i spent ages on that ****.

Couldn't play the challenge yesterday because TC was down for over an hour when I tried to play and from looking at the winner board it looks like that would have been the perfect challenge for me

Don't worry, I won it for you

I'm getting there, only two more wins until I've got 10!

I've been making some improvements and doing a few experiments in an attempt to speed up the site. One of the changes I made a couple of days ago would lock up the site if someone performed a certain series of actions, but it should be fixed now.

I also have a bunch of monitoring set up now, so I should be able to track down the source of problems much more quickly.

so i'm *HOLD*ing for my 3rd win in 5 days. I'm a bit surprised today's score is holding up (in a 3 way tie) but i'll take it.

edit: although i note no dynasty in the scores yet, so there's definitely room for a late dagger in the heart