Probability of outcomes of each "roll" of the dice

[Jimbo1958]

Here is a question for math wizards, computer geeks and experienced gamers...
When it comes to probable outcomes of each play in a game, how does the online version of Strat-o-matic work versus the board game?

Here is the reason I ask the question. Take two players with identical stats for hitting (.300 batting average, 30 home runs). While their stats are the same, their cards may be very different. One might hit a homer with a roll of 1-7 while the other might do so with a roll of 1-2, 1-3, 1-4. I know that the odds are the same, but I always opt for the the player with the 1-7, all things being equal.

When choosing players for the online game, I can reference the player cards, but I don't know if my player selection strategy (however flawed it might be) works because the computer version may rely on some random computer program that does really does not use a simulated "roll" of the dice that "benefits" my player with the 1-7 homer.


I haven't gotten much sleep lately, so don't be harsh on me if this sounds and actually is illogical.

[cummings2]

Well, I am neither a wiz, a geek nor experienced...if anything I am more of a Dumb, Nerdy Virgin (I think I'll try this as my new come on line )

Still here are my thoughts, as far as numbers generated randomly, the important element would be that there are 2 sets of numbers from 1 to 6 and the result added - hence the "simulated roll" you refered to.

I guess there is reason to believe the roll is simulated in some way otherwise players with an INJ=6 would get as injured as players with an INJ=1, and defense up the middle would have the same impact as in the corners...I think it's a safe bet to say that most SOM-OL players have experienced INJ and DEF results that would validate assuming the rolls are simulated.

Just my two cents, keep in mind they're the thoughts from a dumb, nerdy virgin

[bjs73]

I believe that I've got the probabilities correct but please someone else come and correct me if I'm wrong. I will focus on any position player's "1" column and you can roll it out to the "2-6" columns because they are the same.

For the sake of simplicity the first number represents 1d6 followed by a dashed number which represents 2d6.

1-2 = 0.46%
1-3 = 0.93%
1-4 = 1.39%
1-5 = 1.85%
1-6 = 2.31%
1-7 = 2.78%
1-8 = 2.31%
1-9 = 1.85%
1-10 = 1.39%
1-11 = 0.93%
1-12 = 0.46%

Take all of these percentages and multiply by 6 columns and you get:
99.96%
:?:
Seriously, the missing .04% is there, I just rounded the above numbers to the nearest hundredth. There are some incredibly long decimals after those figures so if you really want to find that missing .04%, use your own calculator! LOL

How did I arrive at the figure? First die is a 1/6 chance multiplied by the outcome of two dice. There are 36 possibilities on a 2d6 throw. 1/36 to roll a 1 or 12, 2/36 to roll a 3 or 11, 3/36 to roll a 4 or 10, etc., etc.

Hope this helps?

P.S. As far as I can tell, rolls are simulated in the computer game just like the board game. So, for your comment about all things being equal, on a 1-2 or 1-3 or 1-4 being equal to a 1-7 is correct.

[Valen]

The only advantage of 1-7 over 1-2, 1-3, 1-4 in the dice game would be if you believed the dice were giving unequal rolls. 7 should be rolled 6 out of 36 times unless the dice are not weighted correctly resulting in an unequal distribution. Or if you were good enough with rolling the dice to influence what was rolled. I was once accused of being that good whle playing another dice game.

There would be no value in preferring one over the other unless you believed the computer random dice number generator was not producing numbers in equal distribution. This would be difficult if not impossible to determine without seeing the dice rolls that were supposedly used to determine results. Doubt we will ever see that. But if we did I would be that some like penngray would be extracting the data programatically and analyzing it to determine if there were any discrepancies in random number distribution. Not saying that I would blame him for doing so. Call it the strato version of stealing the other team's signs. :lol:

[MARCPELLETIER]

Actually, dices that have holes into it (99% of all dices) are indeed asymmetrically weighted and probably biased. Moreover, all dices (100%) always have opposite faces worth 7, so 1 is opposed to 6. Therefore, if a bias is affecting a dice, it necessarily favors either 1 or 6, but not both.

I could never figure out, though, if it was 1 or 6 that had the edge.

[bleacher_creature]

Here is a link that I believe accurately presents dice probability percentages as related to SOM, board game or computer.

[url]http://strato.berce.us/dice.htm[/url]

On "split results" (HR 1-16 for example), the computer simply uses a 1-20 dice roll simulator. The percentages Strat uses are extremely accurate, despite what critics might say. In addition, strat does some averaging against the league/history I believe, so that you can play teams from 1968 (pitching dominant era) vs teams from other eras, and have things "balance out".

I believe this even applies to NL players vs AL players within a given year.