Well, the guts of this blog have sat on my profile page for far too long and so I’ve finally decided to hammer out the kinks and post it. I had meant to post it around the time when we had lots of complaints about the RNG, so here is my perspective and hopefully some new insights.

There are several aspects involved (and this blog has turned out longer than anticipated) so I’ll address each in one in turn:

1) Luck Meter
2) Broken RNG/Sooo unlucky
3) Fun Fun Fun

Luck Meter

Quoted from site rules and reference link:
http://fumbbl.com/help:Luck

"As many of you have seen, JavaBBowl shows a number in the titlebar called Luck. As with any statistical figures, it is very important to understand exactly what they represent. The luck value is a semi-statistical figure that represents the outcome of your rolls compared to the statistically expected average. Please take a moment to re-read what you just read, and consider what it really means…”

The explanation that follows is useful in understanding the nature of the luck meter, but it can be a bit much to digest when first read. The first thing to accept is that there is no way to accurately measure each coach’s luck in the game. This is because the “luck” of one coach is not independent of the other. What is good for one coach is generally bad for the other and possibly to varying degrees. Take for example a Turn 15 zombie to zombie long bomb to keep alive the chance of a winning touchdown. This is of course very lucky for the coach controlling the zombies, but if the opposing coach is using elves then whilst it’s not good news it might not be all that bad. He may have the players in place to recover the ball and respond with his own touchdown.

The next thing to realize is that the luck meter is pretty good at what it does. It seems to me that a good deal of thought went into its design and coming up with something significantly better would not be easy However, it is useful to understand it’s limitations so that you can interpret the readings in a more meaningful manner.

The meter judges the results of your actions in a generalized manner by assigning qualitative values (success, neutral and failure) to the outcome of your rolls. In the main these values will mirror the assessment of the coach taking the actions but sometimes it may not. Sometimes you’ll need a push rather than a POW e.g. Surfing.

The luck meter is also unable to interpret the importance of certain rolls and does not consider timing. Getting a POW when blitzing the ball carrier is much “luckier” than getting a POW on a nothing block you make at the end of your turn. Causing injuries early in the game is “luckier” than causing them later.

Although unsubstantiated, I don’t believe that the luck meter takes into account rolls from the kick-off table or the use of rerolls to get that “successful” result, and we should all be familiar with how lucky “getting the ref” can be ;-).

Lastly, as the luck meter starts with zero data for each game, the value is highly inaccurate for the first few rolls that are being made. It also means that bad luck is often emphasized as turnovers mean a smaller data set (yes you have been unlucky, it’s just not all that unlikely…).

Broken RNG/Sooo unlucky

Just how unlucky have you been? I’m sure we’ve all had matches here we have felt under the cosh and nothing seems to be going right. But if you make 36 two dice blocks (with block skill) a match you should expect double skulls and if you dodge all your elves out of TZs during your turn then expect to fail eventually. There is also no rule that you should get double skulls on the 36th block or fail the 36th dodge. Luck is chaotic and random means that it could equally happen on the first action you taken.

One of the great laymen myths is the Law of Averages. Blind belief that “It will all even out in the end.” The truth is that if we assume that the dice rolls are independent of each other then our expectation is that you will have a normal distribution of rolls from here on in. So as the number of rolls you make tends towards infinity the statistical deviance you experienced early on will become insignificant compared to the "expected" normal distributed rolls from the "infinite" number of rolls games you have made.

Sometime ago I was going add a section in my profile for the luckiest sequence of moves I’d seen an opponent do. I was prompted into doing this by seeing a Human catcher dodge into 3 longbeard tackle zones, pick up the ball, dodge into 2 longbeard TZs, dodge twice into 1 TZ, dodge out of a TZ run on and make 2 GFI’s to end up out of blitzing range…

Now both I and my opponent at the time both thought this was fantastically lucky and as mentioned before I was going to add a section in my profile to keep tabs on whether this feat would be beaten. However, when I worked out the probability of this occurring I found out to my disappointment that it was only about 900 to 1.

“900 to 1 you say that’s damn unlikely!” But is it? I’m sure I’m not the only coach who eschews two dice blocks (with block) without reroll to score a TD. Sometimes I’ll not take a two dice block with reroll in favour of scoring the winning TD. What risk am I avoiding? Quad skulls, which is a 1 in 1,296 chance.

Quad skulls is not such an unnatural occurrence in games I’m sure most coaches have experienced it and while I can’t remember getting “hex” skulls, I do remember getting triple skulls and double skulls with a skull/pow on the reroll. The point is that unlikely occurrences aren’t as uncommon as we might believe. I mean we are interacting in a system that produced Lenny, http://fumbbl.com/FUMBBL.php?page=player&op=view&player_id=2927502. Surely anything can happen.

Why not take a step back and look at it objectively? If you still find that you’ve been unlucky outside the wide probability bounds of the fumbbl arena then just consider what Forest Gump said whilst out on a run, having stepped in dog do, “[@#%!] It happens.”

(Note to fact police: I have taken some license to recreate the actual sequence of moves for readability but suffice to say it was something similar, particularly in terms of probability)

Fun Fun Fun

We are all here to have some fun. Each of us derives this pleasure in our own manner. Some like to win, some like to kill, maim and slaughter, and some coaches just like to compete. But most of us fall into each of those categories to some degree.

You will have bad luck in matches and in some matches nuffle will $@#%! slap you. As a coach you will be impotent as you just cannot do anything with the dice rolls you are getting. But before you bemoan your bad luck and start taking it out in chat. Take a moment to consider that you are both there to have fun and in order to play bloodbowl you need two coaches. If you ruin someone else’s fun they may not want to play you again. I know I’ve turned down rematches because of bad previous experience, even very favourable TR/TS offers.

I’m not saying this because “I don’t moan in matches and you shouldn’t”. Because I have moaned in matches but hopefully I’m still getting better at controlling it. I try not to and instead attempt to make fun of my predicament. I find some good chat lightens my mood and the feeling of being put upon. One of the things that made me see this way was a blog by a new coach talking about how he lucked out and beat a good coach. In the blog I remember he mentioned that the good coach bemoaned his luck for most of the match.

We were all “noob’s” at some point, so just consider a match that you’ve played “well” in, by your current standards, you’ve gotten a bit of luck and pulled off a fantastic win against one of the established coaches with a monster CR. How does it feel to have the other coach moaning all match about his luck, either directly or indirectly belittling your achievements?

I feel that fumbbl is a community that (broadly) only benefits from numbers and diversity in its ranks. So before you go and put off a new fumbler by bemoaning your luck consider that we are all here for fun and not just for yours. So give a back a little “fun” and hold your tongue lashing because if you are as good as you think you are then you’ll be getting it back with interest ;-).