instead of booting? (i need the help on me stats! the dice here are overly UN-friendly to me).

http://www.wargear.net/games/view/3958/Board

Could try and Message him.
Boots should count towards Ranking...I assume they count towards your stats too.

i did already. if i boot i get a win in my stat column?

I dunno, try it and let's see

i guess it does. it went up with the 'win'

Congratz on the Win weathertop!!!!

Now that was a hard fought victory. I suggest you put the link on your profile page!

Cool, that's good to know! Thanks for the test!

hey at least i would have won it if they stuck around. :^|

and for the way my rolling has been, EVERY win is precious. my typical unmodified roll is say 6v1 and lose all.

I wish there was an over-all luck-stat so we could brag about how poor our luck is. I love to talk about every roll and how unlikely the outcome was, much to everyone else's dismay.

Though I guess it's still better than my habit slapping people in the face every time they do something differently than how I would have done it.

Andernut wrote: I wish there was an over-all luck-stat so we could brag about how poor our luck is. I love to talk about every roll and how unlikely the outcome was, much to everyone else's dismay.

Though I guess it's still better than my habit slapping people in the face every time they do something differently than how I would have done it.

Would this be a per-game stat or an overall thing?

The problem on WF was the way the stats went screwy whenever you didn't have 6v6 dice rolls. I'm not great on the math so if someone fancies putting together something on how I could do a stats calculator I'd be happy to add it in.

An overall thing, I can figure out the odds, but know nothing of programming and what kind of load calculating 10v4 and 3v7 and 2v12 etc. would add to servers. Just a fun statistic.

The LuckStat plug-in was quite popular, but yeah, it was based on 6-sided die. So, the actual number it would give was something like (number of expected losses) - (number of armies lost) right? If so, it shouldn't be too bad to program, even for arbitrary dice. Next time I have free time, I'll see what I can produce (unless someone else wants to). Then again, Andernut not having a way to back his dice complaints might be fun too :)

-Hugh

I've already done this for the WF script, it was added in the recent update before I went on vacation. It properly accounts for odds for any dice modifications and abandoned territories.

I would be happy to rewrite the client-side javascript used in the script to server side php (i think tom is using php?).

What I would need is the format for the history logs.

Tom you have my email so drop me a note if you are interested. Since it's one of the main things I am interested in I would be happy to write this for the site.

By the way I had thought that the turn-based dice with modifiers would be difficult, and it was mildly so, but figuring out a luck statistic for BAO play is turning out to be much harder.

For instance, assume you have a BAO turn where you are attacking someone with 100 armies, and the dice are set so that you should expect to kill with 50% of them.

So you attack and it turns out that there are 5 defenders. You roll 100 10s. What is your luck for that roll? Is it:

+50, because you rolled 50 more killing dice than you should have?
-45, because you should have killed 100 armies, but you only killed 5?
0, because you killed everyone you should have?

BTW the server load is minimal, it can calculate each modification in about 5ms, and does not increase with the number of turns.  It does increase with the number of total dice sides involved, so as long as you avoid 100 sides vs. 67 sided dice, it'd be OK.

IRoll11s wrote:

For instance, assume you have a BAO turn where you are attacking someone with 100 armies, and the dice are set so that you should expect to kill with 50% of them.

So you attack and it turns out that there are 5 defenders. You roll 100 10s. What is your luck for that roll? Is it:

+50, because you rolled 50 more killing dice than you should have?
-45, because you should have killed 100 armies, but you only killed 5?
0, because you killed everyone you should have?

When calculating the odds in turn-based play, the event we examining is each individual roll, but in Simulplay the event is a whole battle, as you can't stop after each roll. You are rolling a whole cup of dice at once.

I would say for determining your offensive luck, you want to know what is the probability of rolling more kill dice than there are defenders.

Nope. Start over. This line of thinking is heading us into territory where any time you attack with fewer armies than there are defenders you will end up with negative luck.

Maybe we're better off leaving offensive luck as some kind of (#rolls better than kill value)-(#rolls expected better than kill value)+(#armies counterkilled)-(#expected counter-kill losses).

This is, of course, assuming Simultaneous play uses the same dice scheme here as it did in the old world. Surely there's a less convoluted way that still allows for a similar style of play.

The best I could come up with was to have two statistics.. 'luck' which is:

Number of dice that rolled a kill - number of dice that 'could' have rolled a kill (and the equivalent for defense), which is I think what you wrote in your 4th paragraph. This way you avoid under and over kills.

You could also have a secondary stat called 'Allocation', which is how close you were to committing the right number of armies to attacks and defense.

Otherwise you're right, I originally wrote it the first way and checked a long BAO game, and everyone's luck was like -600, -900. The longer the game the higher the negative.