Lord-Willi
Joined: Aug 02, 2003
|
 Posted:
Jul 30, 2003 - 10:30 |
 
|
That might be correct, but seeing the actual dice would be rather interesting. I play elves and therefore may TD without much luck, but i there is no turn i can get past without using rather a Reroll or suffering a turnover.
Luck may be 50% statistically, but that may mean 3,5 as average or binary just 1 or 6 cut in half, the first one would bring me victory the second is just crap.
Mathematics and Statistics cannot be compared as Luck ! |
|
|
Sinner
Joined: Aug 02, 2003
|
| Posted:
Jul 30, 2003 - 10:46 |
|
| Skar2 wrote: | | An intelligent player would make the safest moves before the critical moves and therefore have more successful actions (and higher luck). |
Did you know: it´s more likely to succeed a two dice roll than a dodge (if you are AG4 and have block or AG3 without block). Just wanted to mention, cause then I most recently graduated to an intelligent player. Before I dodged/ failed before the block, because I said safety first (you really want to protect those fragile witch elves), but now I will play bash first, LOL... as if my darkies were bashy...
Dark
So long |
_________________
Sinner
Darkie's Dreams - successfully cherrypicking any race, any coach, any rating, any number of DP since 20/09/2003 ... and still winning! |
|
locutus
Joined: Aug 02, 2003
|
| Posted:
Jul 30, 2003 - 12:44 |
|
Well, if i tell you the truth, that is a question i have decided to consider not important.
There are a few posts related to this and also the two other big questions (TR/ST and Coach rating). The thing is that its very difficult to create a formula that satisfies everybody.
First thing is that the risk depends on the teams, its easier to dodge with an elf with dodge than picking up the ball with a dwarf runner with sure hands. So is it risky to dodge with the elf?, is it risky to pick up the ball with the dwarf?. Too many things to consider, i wouldnt change the formula for luck calculation, mainly cause its not my objetive to be the unluckiest player in fumbbl and because everybody have lucky matches and unlucky matches.
The only thing i think could be useful would be to have the number of POW, SKULLS,ETC... maybe grouped by three,two or one dice rolled.
Anyway, i respect skar2 aproximation to luck calculation and would be very interested in reading a document that explains a theory for correct luck calculation. 
--locutus-- |
|
|
Zhluhur
Joined: Aug 02, 2003
|
| Posted:
Jul 30, 2003 - 13:02 |
|
It would be cool to see the last , say last 20 dice rolls you and you opponent rolled. Show them on the client as little dices graphics above or under the field.
ie.
Your Dices: 1 2 5 6 2 4 5 2 4 6 5 1 2 3 4 5 2 1 4 5 2
Game field:XXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXxx
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXx
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
Opp. Dices: 1 2 3 4 5 1 3 6 6 6 6 6 2 1 1 1 2 3 2 3 4 5 3
So you can point on that 8 1s in a row that made you loose your match |
_________________
*-* Let the games begin! *-* |
|
dertre
Joined: Aug 02, 2003
|
| Posted:
Aug 28, 2003 - 14:58 |
|
arf.. Mathematician : you forgot a modifier due to racial characteristic....
and also one due to the kind of coffe a player may drink for the misclick risk or one for..........
and one.... etc...
the main problem with modelisations : you can take in account whatever you want and their will always be someone to be unhappy cause you negliged some value or approximate.
Thoughts of a bbwl player about modelisation and mathematician in BBowl |
|
|
Mordachai
Joined: Aug 02, 2003
|
| Posted:
Aug 28, 2003 - 15:21 |
|
I like Zhluhur's idea. Have the client showing the latest dice roll for each player is good. Of course it isn't a priority, but still good. |
_________________
"I love this show!!!" - Gir |
|
freak_in_a_frock
Joined: Aug 02, 2003
|
| Posted:
Aug 28, 2003 - 16:07 |
|
it's not the value of the dice that causes luck it is when you get them, for example 1,3,5,5,4,6,6,1 seems like a fairly lucky set of rolls when you look at them, but these are the exact rolls that killed my ogre
GFI 1 failed, pro 3 failed, armour 5+5=10 failed, injury 4+6 injured, 6 dead, apoth 1 failed
so a table of rolls does not take a lot of things into consideration, i think the luck factor is a nice edition to have, it means we can blame someything else other than our appalling play  |
|
|
Colin
Joined: Aug 02, 2003
|
| Posted:
Aug 28, 2003 - 17:12 |
|
In my own opinion, the luckometer is fairly accurate - it isn't so good at recording crucial rolls at key times, because it doesn't recognise them, but I think it does do a decent job of recording more long-term luck trends. I cobbled together a spreadsheet of my own luck vs the luck of my opponents in the games I played them, and discovered an average of 5% luck in their favour, which I felt was right, as I'd had the worst of it in most of the games (I still won most of them  ). It would be nice to see these included in coachs' stats: Avg Luck, Avg Opponent Luck, Luck Difference. |
_________________
Join The Cult of Tzeentch, mutate randomly! | Hug a newb! Join the Faculty of Academy Instructors! |
|
grep-v
Joined: Aug 02, 2003
|
| Posted:
Aug 29, 2003 - 12:11 |
|
| freak_in_a_frock wrote: | it's not the value of the dice that causes luck it is when you get them, for example 1,3,5,5,4,6,6,1 seems like a fairly lucky set of rolls when you look at them, but these are the exact rolls that killed my ogre
GFI 1 failed, pro 3 failed, armour 5+5=10 failed, injury 4+6 injured, 6 dead, apoth 1 failed |
Uhm, perhaps you got something wrong...
Your rolls: 1 3 1
Your opponent's rolls: 5 5 4 6 6
Seems like "a fairly lucky set"? Not really ....
And fairly unlikely too. Each chance for bad things to happen was 50% or less, and six bad things happened in a row (GFI, Pro, AV, Inj, dead, Apo) adding up. Seems to me as if you were in the receiving end. |
|
|
Christer
Joined: Aug 02, 2003
|
Personally, I would love changing the luck value into a "risk" value which gets higher if you do lots of difficult rolls. I'm guessing that most of the players with a high win percentage will have a low risk style of play while the weaker coaches do more and harder rolls.
It would most likely be helpful in a "learn how to win" perspective as well. |
|
|
boobis
Joined: Aug 02, 2003
|
| Posted:
Aug 29, 2003 - 16:31 |
|
| Christer wrote: | Personally, I would love changing the luck value into a "risk" value which gets higher if you do lots of difficult rolls. I'm guessing that most of the players with a high win percentage will have a low risk style of play while the weaker coaches do more and harder rolls.
It would most likely be helpful in a "learn how to win" perspective as well. |
This sounds like a great idea. Have someone proposed it to SkiJunkie?
/Joel |
|
|
CircularLogic
Joined: Aug 22, 2003
|
| Posted:
Aug 29, 2003 - 17:33 |
|
The stats would be easiest like this:
Failed 2+-rolls: x% (#rolls) (#failure)
Failed 3+... etc
So you can see, if you were particulary unlucky (failed 80% of your 3+rolls) or risky (high number of rolls compared to your opponent) and other stuff. My last game had about 40% failed 2+rolls.
Same with other stuff...
Failed 2d-blocks (with block)
Failed 2d-blocks (without block)
Failed to break armor at AV x
Injuryroll -> stunned: x% etc. |
|
|
slackman
Joined: Aug 02, 2003
|
| Posted:
Aug 29, 2003 - 21:28 |
|
u know guys, this has been brought up time and time again. ski has managed to add your dice rolls into the log file at the end of the game, but it only shows the total rolled for both of you. oddly enough, i found that there seems to be numbers rolled in pairs, 1s and 6s are rolled equally, 2s and 5s rolled equally, and 3s and 4s rolled equally. unfortunately, there seems to be a trend in my games where the 1-6 pair gets rolled more often than the 2-5 pair, which in turn occurs more often than the 3-4 pair. this coincides w/ what i have observed directly in the game. i roll double 1s, and my opponent almost certainly will roll a 5,6 block as his first action. or a 6, 5, 6 pass action (pick up ball, throw, catch). but the luck meter only slightly changes for both of us, as it takes the % chance to succeed and compares that to whether it did or not. using a reroll eliminates the first roll from the luck calculation entirely, so im only credited w/ rolling a single 1, instead of two. i think this may go a LONG way in skewing the actual luck % from what most people would regard as "lucky".
i think the problem w/ getting everything seperated is that its going to take processing power away from running the game and putting it into statistics creation. not to mention its going to take a considerable amount of coding to put each and every roll into the right place in the matrix. i think its worth seeing, but there are more important issues, like fixing bugs and implementing the last of the weapons.
slackman42 |
|
|
Delta
Joined: Aug 02, 2003
|
| Posted:
Aug 29, 2003 - 23:22 |
|
For every player that has bad luck...surely there is another that has good luck?
I don't think that calculating luck values from the dice rolls will ever give an accurate reading, as there are instances when a 1 is better than a 6. (well...injuries anyway!!)
I think the 'risk' idea is good but could perhaps stifle the gameplay a bit for some coaches. The coach with the flashy style and 'longshot plays' may end up having to play safe every drive just to keep in the game. (Although it would also broaden their coaching ability and teach them to adapt)
I'm pretty happy with the way things work as they are. (although I don't get to play enough!!).
You get good runs and bad runs.
Play on!!  |
|
|
Jared
Joined: Aug 16, 2003
|
| Posted:
Sep 04, 2003 - 13:26 |
|
as a recipient of frequent terrible luck (making up for my real world lucky lucky lucky factor) id like a stat fro apothecary effectiveness just so i can monitor how many times a darn 1 comes up with that irritating awefnopieng |
_________________
http://fumbbl.com/FUMBBL.php?page=group&op=view&group=2440
XXXtreme ball with added XX |
|
|
|
| |
Gnomes are trash
Search
