And Eden shows exactly how little he knows about the web and hacker attacks...

DDOS stands for Distributed Denial of Service. It is a bunch of computers all hitting the same site at the same time to either crash a router or, at a minimum, deny others the ability to get in. It doesn't delete data. Yeesh.

For the love of God, don't put such wonderful data in a pie chart.
http://www.gilliganondata.com/index.php/2009/12/02/how-succinctly-can-i-explain-why-pie-charts-are-evil/
http://blog.revolutionanalytics.com/2009/08/how-pie-charts-fail.html
http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=00018S

Eden, I took another look at your Turkey game. At first glance, it looks completely insane, but after that, you seem to have caught everyone off guard. Nicely done! I wish I could read the press.

Also, I'm more familiar with the other version of DDoS, Distributed Denial of Sex. I suspect a vast network hitting me and plugging all my pheromone ports. Are you responsible, Draugnar?

"For the love of God, don't put such wonderful data in a pie chart."

Yeah, use a cake graph. teehehehehehehehehehe

Bring the game ID up on here and you can read the global press at least. Of course that may not tell you much.

Hah, I forgot you could get the global press. That's some entertainment right there!

Best performing turkish openings are very rare... Why? Well they perform well because only the rare genius uses them, and because the sample for them is very small it is easier for a small number of wins to push the percentage up.

Great stats, i'll analyise a bit more when i get a chance...

Statistics are like a bikini. What they reveal is suggestive, but what they conceal is vital.

Statistics are like a lamppost to a drunk - for support, not illumination.

Statistics are like drunk girl in a bikini, easy to unveil .... WHUU?!?

Statistics are like analogies...none of them are perfect, some dont make sense and they crap standing up.

"And Eden shows exactly how little he knows about the web and hacker attacks...

DDOS stands for Distributed Denial of Service. It is a bunch of computers all hitting the same site at the same time to either crash a router or, at a minimum, deny others the ability to get in. It doesn't delete data. Yeesh."

...do you have any concept of sarcasm whatsoever?

"Eden, I took another look at your Turkey game. At first glance, it looks completely insane, but after that, you seem to have caught everyone off guard. Nicely done! I wish I could read the press."

The press was basically "guys it's game 4 and me, joey and drano ended up 3wding the first 3 games, so this game really doesn't mean anything unless someone gets like a 30+ center win. so I'm gonna do something totally stupid and see what happens. don't fuck it up, k? k."

and they didn't

^ Don't you just love the leagues? Speaking of which, I need to get with fortknox about them and the F2F in Cinci.

IMPORTANT - If any of you have noticed that the percentages don't always add up to 100%, I don't keep resigns in the table (the fifth possibility for an opening. I don't have any games that are still playing in the db dump). But this shouldn't happen in the CD free tables. It seems the script for isolating CDs still lets some of the resigns slip through. I'll try to figure out why this happens. I had assumed that it was OK since the diagnostic at the top says that it has ignored as many CDs as there are in the database, but there have been some problems there.

Statistics are like bikinis. They are sexy and we want to see the bare things underneath. And asking for them gets you a slap and a restraining order.

Wait, what were we talking about?

What I love is whenever the collective WebDip brain gets ahold of new statistics about moves and outcomes, the way people tend to play on the site changes accordingly. For bonus points, figure out how people will implement what they've learned in this thread and then counteract it in unexpected ways.

I think everyone will do the "hold Spring 01" move. So I'm going to hold Spring *and* Fall 01.

The Leagues are the best. I miss them. Can they restart this summer?

Let's see, how will things change here? I'll go pull some unexpected numbers, but the first thing that jumped out at me: Why is the Edinburgh variation so much better, statistically, than the Yorkshire variation for England?

Don't worry, draug has been working on the leagues for ages now. I'm sure there'll be an announcement soon.

Wait, so according to these statistics, the Yorkshire pudding opening doesn't work usually. Huh. Who'd have thought?

Hey, don't knock it! It has 0% defeats :D

@ President Eden - The Edinburgh variation on what? I assume you mean the London to North Sea, Edinburgh to Norwegian Sea, Liverpool to Edinburgh/Yorkshire opening? I suppose the move to Edinburgh allows for all the same convoying options as the Yorkshire move, plus the option of a convoy to Norway via the Norwegian sea, and is thus a more versatile opening. Though I agree it hardly seems to account for a huge difference.

Yessir, that one. It is more versatile, no doubt, but the inability to cover London and secure Norway if France and Russia give you fits always made me think the Yorkshire variation was better. Seems the stats don't validate that.

north sea can cover london and either norwegian can cover norway or north sea (or convoy to norway).

It is also a good opening to give england a shot at belgium.

It is also a very anti-russian opening.

Not saying it is good or bad, but it is quite versatile. But any opening will have major risks if it gives you a lot of potential for growth. "Why move tri-alb? Italy can royally screw you!" Same preface. It is assumed that if it does well, it isn't because of the opening as much as the diplomacy and the way the game moves.

I'd like to also see the average number of years these games last, if, for nothing else, to prove how much more the game is than the intro move...

Gotta disagree with you, fortknox. The strongest anti-Russian opening around is undoubtedly the one that puts your army in Norway and fleet in Barents Sea in A01. You have that opening with both the Yorkshire and Edinburgh variants. Edinburgh variant isn't anymore anti-Russian than Yorkshire variant. And vis-a-vis Yorkshire it's no better at getting Belgium.

It *is* more versatile, in that it gives you the option of convoying via NWS and using NTh for something else, but it doesn't lead to any higher potential for growth than the Yorkshire variant. You're still only looking at Belgium and Norway (substitute Denmark for Belgium in some circumstances; never Holland). The difference is that if France and Russia do bother you, you lose that potential, because you can't take Norway. (And even if Russia doesn't - can you imagine ENC -> NTH in A01, knowing that you were going to cover London? Yikes!)

I'd also like to see average number of game years. And if it's not too complicated, average percentage of seasons (spring, autumn) that each country survived with the openings - would really help elucidate the survivability of the openings in question. (So say the game went to 1915, but Austria died in 1902. Austria survived for 7% of the seasons in question with that opening.)

Another reason for the separation between the move to Edinburgh and Yorkshire could be that people are more likely to move to Edinburgh if they feel confident that France is on their side, which means they already have a diplomatic leg up versus the average opener to Yorkshire.

Maybe so. I see it as a safety measure, but I can see others not doing the same.

The problem, im guessing, isnt the data, its what youre doing with the data. Percentages arent the best method of analyzing data, and when it comes to stastical inferrences, id say theyre among the worst by themselves. is there a significance between 5 and 10%? what does it mean to have a 20% survival percentage? alone, theyre simply numbers. if youre going to make assumptions then you need to know additional information. knowing the expected values is handy, as is coupling the percentages with a t-test or f-test or ANOVA testing to get a bigger picture.

I would argue that neither of the two Northern Opening variations is inherently better than the other and that the most important thing is that England convoys the army regardless of where it is at the start of F01.